colin/nixpkgs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

nixpkgs docs: format =)

Browse Source
This commit is contained in:
Graham Christensen 2018-05-01 19:54:21 -04:00
parent fd2dce9708
commit 77161de454
No known key found for this signature in database
GPG Key ID: ACA1C1D120C83D5C
27 changed files with 8153 additions and 6710 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1218
doc/coding-conventions.xml
View File

File diff suppressed because it is too large Load Diff

559
doc/configuration.xml
View File

@ -1,42 +1,51 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-packageconfig"> xml:id="chap-packageconfig">
<title>Global configuration</title>
<title>Global configuration</title> <para>
Nix comes with certain defaults about what packages can and cannot be
<para>Nix comes with certain defaults about what packages can and installed, based on a package's metadata. By default, Nix will prevent
cannot be installed, based on a package's metadata. By default, Nix installation if any of the following criteria are true:
will prevent installation if any of the following criteria are </para>
true:</para> <itemizedlist>
<listitem>
<itemizedlist> <para>
<listitem><para>The package is thought to be broken, and has had The package is thought to be broken, and has had its
its <literal>meta.broken</literal> set to <literal>meta.broken</literal> set to <literal>true</literal>.
<literal>true</literal>.</para></listitem> </para>
</listitem>
<listitem><para>The package isn't intended to run on the given system, as none of its <literal>meta.platforms</literal> match the given system.</para></listitem> <listitem>
<para>
<listitem><para>The package's <literal>meta.license</literal> is set The package isn't intended to run on the given system, as none of its
to a license which is considered to be unfree.</para></listitem> <literal>meta.platforms</literal> match the given system.
</para>
<listitem><para>The package has known security vulnerabilities but </listitem>
has not or can not be updated for some reason, and a list of issues <listitem>
has been entered in to the package's <para>
<literal>meta.knownVulnerabilities</literal>.</para></listitem> The package's <literal>meta.license</literal> is set to a license which is
</itemizedlist> considered to be unfree.
</para>
<para>Note that all this is checked during evaluation already, </listitem>
and the check includes any package that is evaluated. <listitem>
In particular, all build-time dependencies are checked. <para>
<literal>nix-env -qa</literal> will (attempt to) hide any packages The package has known security vulnerabilities but has not or can not be
that would be refused. updated for some reason, and a list of issues has been entered in to the
</para> package's <literal>meta.knownVulnerabilities</literal>.
</para>
<para>Each of these criteria can be altered in the nixpkgs </listitem>
configuration.</para> </itemizedlist>
<para>
<para>The nixpkgs configuration for a NixOS system is set in the Note that all this is checked during evaluation already, and the check
<literal>configuration.nix</literal>, as in the following example: includes any package that is evaluated. In particular, all build-time
dependencies are checked. <literal>nix-env -qa</literal> will (attempt to)
hide any packages that would be refused.
</para>
<para>
Each of these criteria can be altered in the nixpkgs configuration.
</para>
<para>
The nixpkgs configuration for a NixOS system is set in the
<literal>configuration.nix</literal>, as in the following example:
<programlisting> <programlisting>
{ {
nixpkgs.config = { nixpkgs.config = {
@ -44,187 +53,197 @@ configuration.</para>
}; };
} }
</programlisting> </programlisting>
However, this does not allow unfree software for individual users. However, this does not allow unfree software for individual users. Their
Their configurations are managed separately.</para> configurations are managed separately.
</para>
<para>A user's of nixpkgs configuration is stored in a user-specific <para>
configuration file located at A user's of nixpkgs configuration is stored in a user-specific configuration
<filename>~/.config/nixpkgs/config.nix</filename>. For example: file located at <filename>~/.config/nixpkgs/config.nix</filename>. For
example:
<programlisting> <programlisting>
{ {
allowUnfree = true; allowUnfree = true;
} }
</programlisting> </programlisting>
</para> </para>
<para>
<para>Note that we are not able to test or build unfree software on Hydra Note that we are not able to test or build unfree software on Hydra due to
due to policy. Most unfree licenses prohibit us from either executing or policy. Most unfree licenses prohibit us from either executing or
distributing the software.</para> distributing the software.
</para>
<section xml:id="sec-allow-broken"> <section xml:id="sec-allow-broken">
<title>Installing broken packages</title> <title>Installing broken packages</title>
<para>
<para>There are two ways to try compiling a package which has been There are two ways to try compiling a package which has been marked as
marked as broken.</para> broken.
</para>
<itemizedlist> <itemizedlist>
<listitem><para> <listitem>
For allowing the build of a broken package once, you can use an <para>
environment variable for a single invocation of the nix tools: For allowing the build of a broken package once, you can use an
environment variable for a single invocation of the nix tools:
<programlisting>$ export NIXPKGS_ALLOW_BROKEN=1</programlisting> <programlisting>$ export NIXPKGS_ALLOW_BROKEN=1</programlisting>
</para></listitem> </para>
</listitem>
<listitem><para> <listitem>
For permanently allowing broken packages to be built, you may <para>
add <literal>allowBroken = true;</literal> to your user's For permanently allowing broken packages to be built, you may add
configuration file, like this: <literal>allowBroken = true;</literal> to your user's configuration file,
like this:
<programlisting> <programlisting>
{ {
allowBroken = true; allowBroken = true;
} }
</programlisting> </programlisting>
</para></listitem> </para>
</listitem>
</itemizedlist> </itemizedlist>
</section> </section>
<section xml:id="sec-allow-unsupported-system">
<section xml:id="sec-allow-unsupported-system">
<title>Installing packages on unsupported systems</title> <title>Installing packages on unsupported systems</title>
<para> <para>
There are also two ways to try compiling a package which has been marked as unsuported for the given system. There are also two ways to try compiling a package which has been marked as
unsuported for the given system.
</para> </para>
<itemizedlist> <itemizedlist>
<listitem><para> <listitem>
For allowing the build of a broken package once, you can use an environment variable for a single invocation of the nix tools: <para>
For allowing the build of a broken package once, you can use an
<programlisting>$ export NIXPKGS_ALLOW_UNSUPPORTED_SYSTEM=1</programlisting> environment variable for a single invocation of the nix tools:
</para></listitem> <programlisting>$ export NIXPKGS_ALLOW_UNSUPPORTED_SYSTEM=1</programlisting>
</para>
<listitem> </listitem>
<para> <listitem>
For permanently allowing broken packages to be built, you may add <literal>allowUnsupportedSystem = true;</literal> to your user's configuration file, like this: <para>
For permanently allowing broken packages to be built, you may add
<literal>allowUnsupportedSystem = true;</literal> to your user's
configuration file, like this:
<programlisting> <programlisting>
{ {
allowUnsupportedSystem = true; allowUnsupportedSystem = true;
} }
</programlisting> </programlisting>
</para> </para>
</listitem> </listitem>
</itemizedlist> </itemizedlist>
<para> <para>
The difference between an a package being unsupported on some system and being broken is admittedly a bit fuzzy. The difference between an a package being unsupported on some system and
If a program <emphasis>ought</emphasis> to work on a certain platform, but doesn't, the platform should be included in <literal>meta.platforms</literal>, but marked as broken with e.g. <literal>meta.broken = !hostPlatform.isWindows</literal>. being broken is admittedly a bit fuzzy. If a program
Of course, this begs the question of what "ought" means exactly. <emphasis>ought</emphasis> to work on a certain platform, but doesn't, the
That is left to the package maintainer. platform should be included in <literal>meta.platforms</literal>, but marked
as broken with e.g. <literal>meta.broken =
!hostPlatform.isWindows</literal>. Of course, this begs the question of what
"ought" means exactly. That is left to the package maintainer.
</para> </para>
</section> </section>
<section xml:id="sec-allow-unfree">
<section xml:id="sec-allow-unfree">
<title>Installing unfree packages</title> <title>Installing unfree packages</title>
<para>There are several ways to tweak how Nix handles a package <para>
which has been marked as unfree.</para> There are several ways to tweak how Nix handles a package which has been
marked as unfree.
</para>
<itemizedlist> <itemizedlist>
<listitem><para> <listitem>
To temporarily allow all unfree packages, you can use an <para>
environment variable for a single invocation of the nix tools: To temporarily allow all unfree packages, you can use an environment
variable for a single invocation of the nix tools:
<programlisting>$ export NIXPKGS_ALLOW_UNFREE=1</programlisting> <programlisting>$ export NIXPKGS_ALLOW_UNFREE=1</programlisting>
</para></listitem> </para>
</listitem>
<listitem><para> <listitem>
It is possible to permanently allow individual unfree packages, <para>
while still blocking unfree packages by default using the It is possible to permanently allow individual unfree packages, while
<literal>allowUnfreePredicate</literal> configuration still blocking unfree packages by default using the
option in the user configuration file.</para> <literal>allowUnfreePredicate</literal> configuration option in the user
configuration file.
<para>This option is a function which accepts a package as a </para>
parameter, and returns a boolean. The following example <para>
configuration accepts a package and always returns false: This option is a function which accepts a package as a parameter, and
returns a boolean. The following example configuration accepts a package
and always returns false:
<programlisting> <programlisting>
{ {
allowUnfreePredicate = (pkg: false); allowUnfreePredicate = (pkg: false);
} }
</programlisting> </programlisting>
</para> </para>
<para>
<para>A more useful example, the following configuration allows A more useful example, the following configuration allows only allows
only allows flash player and visual studio code: flash player and visual studio code:
<programlisting> <programlisting>
{ {
allowUnfreePredicate = (pkg: elem (builtins.parseDrvName pkg.name).name [ "flashplayer" "vscode" ]); allowUnfreePredicate = (pkg: elem (builtins.parseDrvName pkg.name).name [ "flashplayer" "vscode" ]);
} }
</programlisting> </programlisting>
</para></listitem> </para>
</listitem>
<listitem> <listitem>
<para>It is also possible to whitelist and blacklist licenses <para>
that are specifically acceptable or not acceptable, using It is also possible to whitelist and blacklist licenses that are
<literal>whitelistedLicenses</literal> and specifically acceptable or not acceptable, using
<literal>blacklistedLicenses</literal>, respectively. <literal>whitelistedLicenses</literal> and
</para> <literal>blacklistedLicenses</literal>, respectively.
</para>
<para>The following example configuration whitelists the <para>
licenses <literal>amd</literal> and <literal>wtfpl</literal>: The following example configuration whitelists the licenses
<literal>amd</literal> and <literal>wtfpl</literal>:
<programlisting> <programlisting>
{ {
whitelistedLicenses = with stdenv.lib.licenses; [ amd wtfpl ]; whitelistedLicenses = with stdenv.lib.licenses; [ amd wtfpl ];
} }
</programlisting> </programlisting>
</para> </para>
<para>
<para>The following example configuration blacklists the The following example configuration blacklists the <literal>gpl3</literal>
<literal>gpl3</literal> and <literal>agpl3</literal> licenses: and <literal>agpl3</literal> licenses:
<programlisting> <programlisting>
{ {
blacklistedLicenses = with stdenv.lib.licenses; [ agpl3 gpl3 ]; blacklistedLicenses = with stdenv.lib.licenses; [ agpl3 gpl3 ];
} }
</programlisting> </programlisting>
</para> </para>
</listitem> </listitem>
</itemizedlist> </itemizedlist>
<para>A complete list of licenses can be found in the file <para>
<filename>lib/licenses.nix</filename> of the nixpkgs tree.</para> A complete list of licenses can be found in the file
</section> <filename>lib/licenses.nix</filename> of the nixpkgs tree.
</para>
</section>
<section xml:id="sec-allow-insecure">
<title>Installing insecure packages</title>
<para>
<section xml:id="sec-allow-insecure"> There are several ways to tweak how Nix handles a package which has been
<title> marked as insecure.
Installing insecure packages </para>
</title>
<para>There are several ways to tweak how Nix handles a package
which has been marked as insecure.</para>
<itemizedlist> <itemizedlist>
<listitem><para> <listitem>
To temporarily allow all insecure packages, you can use an <para>
environment variable for a single invocation of the nix tools: To temporarily allow all insecure packages, you can use an environment
variable for a single invocation of the nix tools:
<programlisting>$ export NIXPKGS_ALLOW_INSECURE=1</programlisting> <programlisting>$ export NIXPKGS_ALLOW_INSECURE=1</programlisting>
</para></listitem> </para>
</listitem>
<listitem><para> <listitem>
It is possible to permanently allow individual insecure <para>
packages, while still blocking other insecure packages by It is possible to permanently allow individual insecure packages, while
default using the <literal>permittedInsecurePackages</literal> still blocking other insecure packages by default using the
configuration option in the user configuration file.</para> <literal>permittedInsecurePackages</literal> configuration option in the
user configuration file.
<para>The following example configuration permits the </para>
installation of the hypothetically insecure package <para>
<literal>hello</literal>, version <literal>1.2.3</literal>: The following example configuration permits the installation of the
hypothetically insecure package <literal>hello</literal>, version
<literal>1.2.3</literal>:
<programlisting> <programlisting>
{ {
permittedInsecurePackages = [ permittedInsecurePackages = [
@ -232,47 +251,44 @@ distributing the software.</para>
]; ];
} }
</programlisting> </programlisting>
</para> </para>
</listitem> </listitem>
<listitem>
<listitem><para> <para>
It is also possible to create a custom policy around which It is also possible to create a custom policy around which insecure
insecure packages to allow and deny, by overriding the packages to allow and deny, by overriding the
<literal>allowInsecurePredicate</literal> configuration <literal>allowInsecurePredicate</literal> configuration option.
option.</para> </para>
<para>
<para>The <literal>allowInsecurePredicate</literal> option is a The <literal>allowInsecurePredicate</literal> option is a function which
function which accepts a package and returns a boolean, much accepts a package and returns a boolean, much like
like <literal>allowUnfreePredicate</literal>.</para> <literal>allowUnfreePredicate</literal>.
</para>
<para>The following configuration example only allows insecure <para>
packages with very short names: The following configuration example only allows insecure packages with
very short names:
<programlisting> <programlisting>
{ {
allowInsecurePredicate = (pkg: (builtins.stringLength (builtins.parseDrvName pkg.name).name) &lt;= 5); allowInsecurePredicate = (pkg: (builtins.stringLength (builtins.parseDrvName pkg.name).name) &lt;= 5);
} }
</programlisting> </programlisting>
</para> </para>
<para>
<para>Note that <literal>permittedInsecurePackages</literal> is Note that <literal>permittedInsecurePackages</literal> is only checked if
only checked if <literal>allowInsecurePredicate</literal> is not <literal>allowInsecurePredicate</literal> is not specified.
specified. </para>
</para></listitem> </listitem>
</itemizedlist> </itemizedlist>
</section> </section>
<!--============================================================--> <!--============================================================-->
<section xml:id="sec-modify-via-packageOverrides">
<title>Modify packages via <literal>packageOverrides</literal></title>
<section xml:id="sec-modify-via-packageOverrides"><title>Modify <para>
packages via <literal>packageOverrides</literal></title> You can define a function called <varname>packageOverrides</varname> in your
local <filename>~/.config/nixpkgs/config.nix</filename> to override nix
<para>You can define a function called packages. It must be a function that takes pkgs as an argument and return
<varname>packageOverrides</varname> in your local modified set of packages.
<filename>~/.config/nixpkgs/config.nix</filename> to override nix packages. It
must be a function that takes pkgs as an argument and return modified
set of packages.
<programlisting> <programlisting>
{ {
packageOverrides = pkgs: rec { packageOverrides = pkgs: rec {
@ -280,30 +296,27 @@ set of packages.
}; };
} }
</programlisting> </programlisting>
</para>
</para> </section>
<section xml:id="sec-declarative-package-management">
</section>
<section xml:id="sec-declarative-package-management">
<title>Declarative Package Management</title> <title>Declarative Package Management</title>
<section xml:id="sec-building-environment"> <section xml:id="sec-building-environment">
<title>Build an environment</title> <title>Build an environment</title>
<para> <para>
Using <literal>packageOverrides</literal>, it is possible to manage Using <literal>packageOverrides</literal>, it is possible to manage
packages declaratively. This means that we can list all of our desired packages declaratively. This means that we can list all of our desired
packages within a declarative Nix expression. For example, to have packages within a declarative Nix expression. For example, to have
<literal>aspell</literal>, <literal>bc</literal>, <literal>aspell</literal>, <literal>bc</literal>,
<literal>ffmpeg</literal>, <literal>coreutils</literal>, <literal>ffmpeg</literal>, <literal>coreutils</literal>,
<literal>gdb</literal>, <literal>nixUnstable</literal>, <literal>gdb</literal>, <literal>nixUnstable</literal>,
<literal>emscripten</literal>, <literal>jq</literal>, <literal>emscripten</literal>, <literal>jq</literal>,
<literal>nox</literal>, and <literal>silver-searcher</literal>, we could <literal>nox</literal>, and <literal>silver-searcher</literal>, we could
use the following in <filename>~/.config/nixpkgs/config.nix</filename>: use the following in <filename>~/.config/nixpkgs/config.nix</filename>:
</para> </para>
<screen> <screen>
{ {
packageOverrides = pkgs: with pkgs; { packageOverrides = pkgs: with pkgs; {
myPackages = pkgs.buildEnv { myPackages = pkgs.buildEnv {
@ -314,17 +327,17 @@ set of packages.
} }
</screen> </screen>
<para> <para>
To install it into our environment, you can just run <literal>nix-env -iA To install it into our environment, you can just run <literal>nix-env -iA
nixpkgs.myPackages</literal>. If you want to load the packages to be built nixpkgs.myPackages</literal>. If you want to load the packages to be built
from a working copy of <literal>nixpkgs</literal> you just run from a working copy of <literal>nixpkgs</literal> you just run
<literal>nix-env -f. -iA myPackages</literal>. To explore what's been <literal>nix-env -f. -iA myPackages</literal>. To explore what's been
installed, just look through <filename>~/.nix-profile/</filename>. You can installed, just look through <filename>~/.nix-profile/</filename>. You can
see that a lot of stuff has been installed. Some of this stuff is useful see that a lot of stuff has been installed. Some of this stuff is useful
some of it isn't. Let's tell Nixpkgs to only link the stuff that we want: some of it isn't. Let's tell Nixpkgs to only link the stuff that we want:
</para> </para>
<screen> <screen>
{ {
packageOverrides = pkgs: with pkgs; { packageOverrides = pkgs: with pkgs; {
myPackages = pkgs.buildEnv { myPackages = pkgs.buildEnv {
@ -336,31 +349,30 @@ set of packages.
} }
</screen> </screen>
<para> <para>
<literal>pathsToLink</literal> tells Nixpkgs to only link the paths listed <literal>pathsToLink</literal> tells Nixpkgs to only link the paths listed
which gets rid of the extra stuff in the profile. which gets rid of the extra stuff in the profile. <filename>/bin</filename>
<filename>/bin</filename> and <filename>/share</filename> are good and <filename>/share</filename> are good defaults for a user environment,
defaults for a user environment, getting rid of the clutter. If you are getting rid of the clutter. If you are running on Nix on MacOS, you may
running on Nix on MacOS, you may want to add another path as well, want to add another path as well, <filename>/Applications</filename>, that
<filename>/Applications</filename>, that makes GUI apps available. makes GUI apps available.
</para> </para>
</section> </section>
<section xml:id="sec-getting-documentation"> <section xml:id="sec-getting-documentation">
<title>Getting documentation</title> <title>Getting documentation</title>
<para> <para>
After building that new environment, look through After building that new environment, look through
<filename>~/.nix-profile</filename> to make sure everything is there that <filename>~/.nix-profile</filename> to make sure everything is there that
we wanted. Discerning readers will note that some files are missing. Look we wanted. Discerning readers will note that some files are missing. Look
inside <filename>~/.nix-profile/share/man/man1/</filename> to verify this. inside <filename>~/.nix-profile/share/man/man1/</filename> to verify this.
There are no man pages for any of the Nix tools! This is because some There are no man pages for any of the Nix tools! This is because some
packages like Nix have multiple outputs for things like documentation (see packages like Nix have multiple outputs for things like documentation (see
section 4). Let's make Nix install those as well. section 4). Let's make Nix install those as well.
</para> </para>
<screen> <screen>
{ {
packageOverrides = pkgs: with pkgs; { packageOverrides = pkgs: with pkgs; {
myPackages = pkgs.buildEnv { myPackages = pkgs.buildEnv {
@ -373,14 +385,13 @@ set of packages.
} }
</screen> </screen>
<para> <para>
This provides us with some useful documentation for using our packages. This provides us with some useful documentation for using our packages.
However, if we actually want those manpages to be detected by man, we need However, if we actually want those manpages to be detected by man, we need
to set up our environment. This can also be managed within Nix to set up our environment. This can also be managed within Nix expressions.
expressions. </para>
</para>
<screen> <screen>
{ {
packageOverrides = pkgs: with pkgs; rec { packageOverrides = pkgs: with pkgs; rec {
myProfile = writeText "my-profile" '' myProfile = writeText "my-profile" ''
@ -412,13 +423,13 @@ cp ${myProfile} $out/etc/profile.d/my-profile.sh
} }
</screen> </screen>
<para> <para>
For this to work fully, you must also have this script sourced when you For this to work fully, you must also have this script sourced when you are
are logged in. Try adding something like this to your logged in. Try adding something like this to your
<filename>~/.profile</filename> file: <filename>~/.profile</filename> file:
</para> </para>
<screen> <screen>
#!/bin/sh #!/bin/sh
if [ -d $HOME/.nix-profile/etc/profile.d ]; then if [ -d $HOME/.nix-profile/etc/profile.d ]; then
for i in $HOME/.nix-profile/etc/profile.d/*.sh; do for i in $HOME/.nix-profile/etc/profile.d/*.sh; do
@ -429,23 +440,22 @@ if [ -d $HOME/.nix-profile/etc/profile.d ]; then
fi fi
</screen> </screen>
<para> <para>
Now just run <literal>source $HOME/.profile</literal> and you can starting Now just run <literal>source $HOME/.profile</literal> and you can starting
loading man pages from your environent. loading man pages from your environent.
</para> </para>
</section> </section>
<section xml:id="sec-gnu-info-setup"> <section xml:id="sec-gnu-info-setup">
<title>GNU info setup</title> <title>GNU info setup</title>
<para> <para>
Configuring GNU info is a little bit trickier than man pages. To work Configuring GNU info is a little bit trickier than man pages. To work
correctly, info needs a database to be generated. This can be done with correctly, info needs a database to be generated. This can be done with
some small modifications to our environment scripts. some small modifications to our environment scripts.
</para> </para>
<screen> <screen>
{ {
packageOverrides = pkgs: with pkgs; rec { packageOverrides = pkgs: with pkgs; rec {
myProfile = writeText "my-profile" '' myProfile = writeText "my-profile" ''
@ -487,16 +497,13 @@ cp ${myProfile} $out/etc/profile.d/my-profile.sh
} }
</screen> </screen>
<para> <para>
<literal>postBuild</literal> tells Nixpkgs to run a command after building <literal>postBuild</literal> tells Nixpkgs to run a command after building
the environment. In this case, <literal>install-info</literal> adds the the environment. In this case, <literal>install-info</literal> adds the
installed info pages to <literal>dir</literal> which is GNU info's default installed info pages to <literal>dir</literal> which is GNU info's default
root node. Note that <literal>texinfoInteractive</literal> is added to the root node. Note that <literal>texinfoInteractive</literal> is added to the
environment to give the <literal>install-info</literal> command. environment to give the <literal>install-info</literal> command.
</para> </para>
</section> </section>
</section>
</section>
</chapter> </chapter>

38
doc/contributing.xml
View File

@ -1,35 +1,35 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-contributing"> xml:id="chap-contributing">
<title>Contributing to this documentation</title>
<title>Contributing to this documentation</title> <para>
The DocBook sources of the Nixpkgs manual are in the
<para>The DocBook sources of the Nixpkgs manual are in the <filename <filename
xlink:href="https://github.com/NixOS/nixpkgs/tree/master/doc">doc</filename> xlink:href="https://github.com/NixOS/nixpkgs/tree/master/doc">doc</filename>
subdirectory of the Nixpkgs repository.</para> subdirectory of the Nixpkgs repository.
</para>
<para>You can quickly check your edits with <command>make</command>:</para> <para>
You can quickly check your edits with <command>make</command>:
</para>
<screen> <screen>
$ cd /path/to/nixpkgs/doc $ cd /path/to/nixpkgs/doc
$ nix-shell $ nix-shell
[nix-shell]$ make [nix-shell]$ make
</screen> </screen>
<para>
<para>If you experience problems, run <command>make debug</command> If you experience problems, run <command>make debug</command> to help
to help understand the docbook errors.</para> understand the docbook errors.
</para>
<para>After making modifications to the manual, it's important to <para>
build it before committing. You can do that as follows: After making modifications to the manual, it's important to build it before
committing. You can do that as follows:
<screen> <screen>
$ cd /path/to/nixpkgs/doc $ cd /path/to/nixpkgs/doc
$ nix-shell $ nix-shell
[nix-shell]$ make clean [nix-shell]$ make clean
[nix-shell]$ nix-build . [nix-shell]$ nix-build .
</screen> </screen>
If the build succeeds, the manual will be in
If the build succeeds, the manual will be in <filename>./result/share/doc/nixpkgs/manual.html</filename>.
<filename>./result/share/doc/nixpkgs/manual.html</filename>.</para> </para>
</chapter> </chapter>

689
doc/cross-compilation.xml
View File

@ -1,308 +1,469 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-cross"> xml:id="chap-cross">
<title>Cross-compilation</title>
<title>Cross-compilation</title> <section xml:id="sec-cross-intro">
<section xml:id="sec-cross-intro">
<title>Introduction</title> <title>Introduction</title>
<para> <para>
"Cross-compilation" means compiling a program on one machine for another type of machine. "Cross-compilation" means compiling a program on one machine for another
For example, a typical use of cross compilation is to compile programs for embedded devices. type of machine. For example, a typical use of cross compilation is to
These devices often don't have the computing power and memory to compile their own programs. compile programs for embedded devices. These devices often don't have the
One might think that cross-compilation is a fairly niche concern, but there are advantages to being rigorous about distinguishing build-time vs run-time environments even when one is developing and deploying on the same machine. computing power and memory to compile their own programs. One might think
Nixpkgs is increasingly adopting the opinion that packages should be written with cross-compilation in mind, and nixpkgs should evaluate in a similar way (by minimizing cross-compilation-specific special cases) whether or not one is cross-compiling. that cross-compilation is a fairly niche concern, but there are advantages
to being rigorous about distinguishing build-time vs run-time environments
even when one is developing and deploying on the same machine. Nixpkgs is
increasingly adopting the opinion that packages should be written with
cross-compilation in mind, and nixpkgs should evaluate in a similar way (by
minimizing cross-compilation-specific special cases) whether or not one is
cross-compiling.
</para> </para>
<para> <para>
This chapter will be organized in three parts. This chapter will be organized in three parts. First, it will describe the
First, it will describe the basics of how to package software in a way that supports cross-compilation. basics of how to package software in a way that supports cross-compilation.
Second, it will describe how to use Nixpkgs when cross-compiling. Second, it will describe how to use Nixpkgs when cross-compiling. Third, it
Third, it will describe the internal infrastructure supporting cross-compilation. will describe the internal infrastructure supporting cross-compilation.
</para> </para>
</section> </section>
<!--============================================================--> <!--============================================================-->
<section xml:id="sec-cross-packaging">
<section xml:id="sec-cross-packaging">
<title>Packaging in a cross-friendly manner</title> <title>Packaging in a cross-friendly manner</title>
<section> <section>
<title>Platform parameters</title> <title>Platform parameters</title>
<para>
Nixpkgs follows the <link xlink:href="https://gcc.gnu.org/onlinedocs/gccint/Configure-Terms.html">common historical convention of GNU autoconf</link> of distinguishing between 3 types of platform: <wordasword>build</wordasword>, <wordasword>host</wordasword>, and <wordasword>target</wordasword>.
In summary, <wordasword>build</wordasword> is the platform on which a package is being built, <wordasword>host</wordasword> is the platform on which it is to run. The third attribute, <wordasword>target</wordasword>, is relevant only for certain specific compilers and build tools. <para>
</para> Nixpkgs follows the
<link xlink:href="https://gcc.gnu.org/onlinedocs/gccint/Configure-Terms.html">common
historical convention of GNU autoconf</link> of distinguishing between 3
types of platform: <wordasword>build</wordasword>,
<wordasword>host</wordasword>, and <wordasword>target</wordasword>. In
summary, <wordasword>build</wordasword> is the platform on which a package
is being built, <wordasword>host</wordasword> is the platform on which it
is to run. The third attribute, <wordasword>target</wordasword>, is
relevant only for certain specific compilers and build tools.
</para>
<para> <para>
In Nixpkgs, these three platforms are defined as attribute sets under the names <literal>buildPlatform</literal>, <literal>hostPlatform</literal>, and <literal>targetPlatform</literal>. In Nixpkgs, these three platforms are defined as attribute sets under the
All three are always defined as attributes in the standard environment, and at the top level. That means one can get at them just like a dependency in a function that is imported with <literal>callPackage</literal>: names <literal>buildPlatform</literal>, <literal>hostPlatform</literal>,
<programlisting>{ stdenv, buildPlatform, hostPlatform, fooDep, barDep, .. }: ...buildPlatform...</programlisting>, or just off <varname>stdenv</varname>: and <literal>targetPlatform</literal>. All three are always defined as
<programlisting>{ stdenv, fooDep, barDep, .. }: ...stdenv.buildPlatform...</programlisting>. attributes in the standard environment, and at the top level. That means
</para> one can get at them just like a dependency in a function that is imported
<variablelist> with <literal>callPackage</literal>:
<varlistentry> <programlisting>{ stdenv, buildPlatform, hostPlatform, fooDep, barDep, .. }: ...buildPlatform...</programlisting>
<term><varname>buildPlatform</varname></term> , or just off <varname>stdenv</varname>:
<listitem><para> <programlisting>{ stdenv, fooDep, barDep, .. }: ...stdenv.buildPlatform...</programlisting>
The "build platform" is the platform on which a package is built. .
Once someone has a built package, or pre-built binary package, the build platform should not matter and be safe to ignore. </para>
</para></listitem>
</varlistentry> <variablelist>
<varlistentry> <varlistentry>
<term><varname>hostPlatform</varname></term> <term><varname>buildPlatform</varname>
<listitem><para> </term>
The "host platform" is the platform on which a package will be run. <listitem>
This is the simplest platform to understand, but also the one with the worst name. <para>
</para></listitem> The "build platform" is the platform on which a package is built. Once
</varlistentry> someone has a built package, or pre-built binary package, the build
<varlistentry> platform should not matter and be safe to ignore.
<term><varname>targetPlatform</varname></term> </para>
<listitem> </listitem>
<para> </varlistentry>
The "target platform" attribute is, unlike the other two attributes, not actually fundamental to the process of building software. <varlistentry>
Instead, it is only relevant for compatibility with building certain specific compilers and build tools. <term><varname>hostPlatform</varname>
It can be safely ignored for all other packages. </term>
</para> <listitem>
<para> <para>
The build process of certain compilers is written in such a way that the compiler resulting from a single build can itself only produce binaries for a single platform. The "host platform" is the platform on which a package will be run. This
The task specifying this single "target platform" is thus pushed to build time of the compiler. is the simplest platform to understand, but also the one with the worst
The root cause of this mistake is often that the compiler (which will be run on the host) and the the standard library/runtime (which will be run on the target) are built by a single build process. name.
</para> </para>
<para> </listitem>
There is no fundamental need to think about a single target ahead of time like this. </varlistentry>
If the tool supports modular or pluggable backends, both the need to specify the target at build time and the constraint of having only a single target disappear. <varlistentry>
An example of such a tool is LLVM. <term><varname>targetPlatform</varname>
</para> </term>
<para> <listitem>
Although the existence of a "target platfom" is arguably a historical mistake, it is a common one: examples of tools that suffer from it are GCC, Binutils, GHC and Autoconf. <para>
Nixpkgs tries to avoid sharing in the mistake where possible. The "target platform" attribute is, unlike the other two attributes, not
Still, because the concept of a target platform is so ingrained, it is best to support it as is. actually fundamental to the process of building software. Instead, it is
</para> only relevant for compatibility with building certain specific compilers
</listitem> and build tools. It can be safely ignored for all other packages.
</varlistentry> </para>
</variablelist> <para>
<para> The build process of certain compilers is written in such a way that the
The exact schema these fields follow is a bit ill-defined due to a long and convoluted evolution, but this is slowly being cleaned up. compiler resulting from a single build can itself only produce binaries
You can see examples of ones used in practice in <literal>lib.systems.examples</literal>; note how they are not all very consistent. for a single platform. The task specifying this single "target platform"
For now, here are few fields can count on them containing: is thus pushed to build time of the compiler. The root cause of this
</para> mistake is often that the compiler (which will be run on the host) and
<variablelist> the the standard library/runtime (which will be run on the target) are
<varlistentry> built by a single build process.
<term><varname>system</varname></term> </para>
<listitem> <para>
<para> There is no fundamental need to think about a single target ahead of
This is a two-component shorthand for the platform. time like this. If the tool supports modular or pluggable backends, both
Examples of this would be "x86_64-darwin" and "i686-linux"; see <literal>lib.systems.doubles</literal> for more. the need to specify the target at build time and the constraint of
This format isn't very standard, but has built-in support in Nix, such as the <varname>builtins.currentSystem</varname> impure string. having only a single target disappear. An example of such a tool is
</para> LLVM.
</listitem> </para>
</varlistentry> <para>
<varlistentry> Although the existence of a "target platfom" is arguably a historical
<term><varname>config</varname></term> mistake, it is a common one: examples of tools that suffer from it are
<listitem> GCC, Binutils, GHC and Autoconf. Nixpkgs tries to avoid sharing in the
<para> mistake where possible. Still, because the concept of a target platform
This is a 3- or 4- component shorthand for the platform. is so ingrained, it is best to support it as is.
Examples of this would be "x86_64-unknown-linux-gnu" and "aarch64-apple-darwin14". </para>
This is a standard format called the "LLVM target triple", as they are pioneered by LLVM and traditionally just used for the <varname>targetPlatform</varname>. </listitem>
This format is strictly more informative than the "Nix host double", as the previous format could analogously be termed. </varlistentry>
This needs a better name than <varname>config</varname>! </variablelist>
</para>
</listitem> <para>
</varlistentry> The exact schema these fields follow is a bit ill-defined due to a long and
<varlistentry> convoluted evolution, but this is slowly being cleaned up. You can see
<term><varname>parsed</varname></term> examples of ones used in practice in
<listitem> <literal>lib.systems.examples</literal>; note how they are not all very
<para> consistent. For now, here are few fields can count on them containing:
This is a nix representation of a parsed LLVM target triple with white-listed components. </para>
This can be specified directly, or actually parsed from the <varname>config</varname>.
[Technically, only one need be specified and the others can be inferred, though the precision of inference may not be very good.] <variablelist>
See <literal>lib.systems.parse</literal> for the exact representation. <varlistentry>
</para> <term><varname>system</varname>
</listitem> </term>
</varlistentry> <listitem>
<varlistentry> <para>
<term><varname>libc</varname></term> This is a two-component shorthand for the platform. Examples of this
<listitem> would be "x86_64-darwin" and "i686-linux"; see
<para> <literal>lib.systems.doubles</literal> for more. This format isn't very
This is a string identifying the standard C library used. standard, but has built-in support in Nix, such as the
Valid identifiers include "glibc" for GNU libc, "libSystem" for Darwin's Libsystem, and "uclibc" for µClibc. <varname>builtins.currentSystem</varname> impure string.
It should probably be refactored to use the module system, like <varname>parse</varname>. </para>
</para> </listitem>
</listitem> </varlistentry>
</varlistentry> <varlistentry>
<varlistentry> <term><varname>config</varname>
<term><varname>is*</varname></term> </term>
<listitem> <listitem>
<para> <para>
These predicates are defined in <literal>lib.systems.inspect</literal>, and slapped on every platform. This is a 3- or 4- component shorthand for the platform. Examples of
They are superior to the ones in <varname>stdenv</varname> as they force the user to be explicit about which platform they are inspecting. this would be "x86_64-unknown-linux-gnu" and "aarch64-apple-darwin14".
Please use these instead of those. This is a standard format called the "LLVM target triple", as they are
</para> pioneered by LLVM and traditionally just used for the
</listitem> <varname>targetPlatform</varname>. This format is strictly more
</varlistentry> informative than the "Nix host double", as the previous format could
<varlistentry> analogously be termed. This needs a better name than
<term><varname>platform</varname></term> <varname>config</varname>!
<listitem> </para>
<para> </listitem>
This is, quite frankly, a dumping ground of ad-hoc settings (it's an attribute set). </varlistentry>
See <literal>lib.systems.platforms</literal> for examples—there's hopefully one in there that will work verbatim for each platform that is working. <varlistentry>
Please help us triage these flags and give them better homes! <term><varname>parsed</varname>
</para> </term>
</listitem> <listitem>
</varlistentry> <para>
</variablelist> This is a nix representation of a parsed LLVM target triple with
white-listed components. This can be specified directly, or actually
parsed from the <varname>config</varname>. [Technically, only one need
be specified and the others can be inferred, though the precision of
inference may not be very good.] See
<literal>lib.systems.parse</literal> for the exact representation.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>libc</varname>
</term>
<listitem>
<para>
This is a string identifying the standard C library used. Valid
identifiers include "glibc" for GNU libc, "libSystem" for Darwin's
Libsystem, and "uclibc" for µClibc. It should probably be refactored to
use the module system, like <varname>parse</varname>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>is*</varname>
</term>
<listitem>
<para>
These predicates are defined in <literal>lib.systems.inspect</literal>,
and slapped on every platform. They are superior to the ones in
<varname>stdenv</varname> as they force the user to be explicit about
which platform they are inspecting. Please use these instead of those.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>platform</varname>
</term>
<listitem>
<para>
This is, quite frankly, a dumping ground of ad-hoc settings (it's an
attribute set). See <literal>lib.systems.platforms</literal> for
examples—there's hopefully one in there that will work verbatim for
each platform that is working. Please help us triage these flags and
give them better homes!
</para>
</listitem>
</varlistentry>
</variablelist>
</section> </section>
<section> <section>
<title>Specifying Dependencies</title> <title>Specifying Dependencies</title>
<para>
In this section we explore the relationship between both runtime and
buildtime dependencies and the 3 Autoconf platforms.
</para>
<para>
A runtime dependency between 2 packages implies that between them both the
host and target platforms match. This is directly implied by the meaning of
"host platform" and "runtime dependency": The package dependency exists
while both packages are running on a single host platform.
</para>
<para>
A build time dependency, however, implies a shift in platforms between the
depending package and the depended-on package. The meaning of a build time
dependency is that to build the depending package we need to be able to run
the depended-on's package. The depending package's build platform is
therefore equal to the depended-on package's host platform. Analogously,
the depending package's host platform is equal to the depended-on package's
target platform.
</para>
<para>
In this manner, given the 3 platforms for one package, we can determine the
three platforms for all its transitive dependencies. This is the most
important guiding principle behind cross-compilation with Nixpkgs, and will
be called the <wordasword>sliding window principle</wordasword>.
</para>
<para>
Some examples will probably make this clearer. If a package is being built
with a <literal>(build, host, target)</literal> platform triple of
<literal>(foo, bar, bar)</literal>, then its build-time dependencies would
have a triple of <literal>(foo, foo, bar)</literal>, and <emphasis>those
packages'</emphasis> build-time dependencies would have triple of
<literal>(foo, foo, foo)</literal>. In other words, it should take two
"rounds" of following build-time dependency edges before one reaches a
fixed point where, by the sliding window principle, the platform triple no
longer changes. Indeed, this happens with cross compilation, where only
rounds of native dependencies starting with the second necessarily coincide
with native packages.
</para>
<note>
<para> <para>
In this section we explore the relationship between both runtime and buildtime dependencies and the 3 Autoconf platforms. The depending package's target platform is unconstrained by the sliding
window principle, which makes sense in that one can in principle build
cross compilers targeting arbitrary platforms.
</para> </para>
</note>
<para>
How does this work in practice? Nixpkgs is now structured so that
build-time dependencies are taken from <varname>buildPackages</varname>,
whereas run-time dependencies are taken from the top level attribute set.
For example, <varname>buildPackages.gcc</varname> should be used at build
time, while <varname>gcc</varname> should be used at run time. Now, for
most of Nixpkgs's history, there was no <varname>buildPackages</varname>,
and most packages have not been refactored to use it explicitly. Instead,
one can use the six (<emphasis>gasp</emphasis>) attributes used for
specifying dependencies as documented in
<xref linkend="ssec-stdenv-dependencies"/>. We "splice" together the
run-time and build-time package sets with <varname>callPackage</varname>,
and then <varname>mkDerivation</varname> for each of four attributes pulls
the right derivation out. This splicing can be skipped when not cross
compiling as the package sets are the same, but is a bit slow for cross
compiling. Because of this, a best-of-both-worlds solution is in the works
with no splicing or explicit access of <varname>buildPackages</varname>
needed. For now, feel free to use either method.
</para>
<note>
<para> <para>
A runtime dependency between 2 packages implies that between them both the host and target platforms match. There is also a "backlink" <varname>targetPackages</varname>, yielding a
This is directly implied by the meaning of "host platform" and "runtime dependency": package set whose <varname>buildPackages</varname> is the current package
The package dependency exists while both packages are running on a single host platform. set. This is a hack, though, to accommodate compilers with lousy build
systems. Please do not use this unless you are absolutely sure you are
packaging such a compiler and there is no other way.
</para> </para>
<para> </note>
A build time dependency, however, implies a shift in platforms between the depending package and the depended-on package.
The meaning of a build time dependency is that to build the depending package we need to be able to run the depended-on's package.
The depending package's build platform is therefore equal to the depended-on package's host platform.
Analogously, the depending package's host platform is equal to the depended-on package's target platform.
</para>
<para>
In this manner, given the 3 platforms for one package, we can determine the three platforms for all its transitive dependencies.
This is the most important guiding principle behind cross-compilation with Nixpkgs, and will be called the <wordasword>sliding window principle</wordasword>.
</para>
<para>
Some examples will probably make this clearer.
If a package is being built with a <literal>(build, host, target)</literal> platform triple of <literal>(foo, bar, bar)</literal>, then its build-time dependencies would have a triple of <literal>(foo, foo, bar)</literal>, and <emphasis>those packages'</emphasis> build-time dependencies would have triple of <literal>(foo, foo, foo)</literal>.
In other words, it should take two "rounds" of following build-time dependency edges before one reaches a fixed point where, by the sliding window principle, the platform triple no longer changes.
Indeed, this happens with cross compilation, where only rounds of native dependencies starting with the second necessarily coincide with native packages.
</para>
<note><para>
The depending package's target platform is unconstrained by the sliding window principle, which makes sense in that one can in principle build cross compilers targeting arbitrary platforms.
</para></note>
<para>
How does this work in practice? Nixpkgs is now structured so that build-time dependencies are taken from <varname>buildPackages</varname>, whereas run-time dependencies are taken from the top level attribute set.
For example, <varname>buildPackages.gcc</varname> should be used at build time, while <varname>gcc</varname> should be used at run time.
Now, for most of Nixpkgs's history, there was no <varname>buildPackages</varname>, and most packages have not been refactored to use it explicitly.
Instead, one can use the six (<emphasis>gasp</emphasis>) attributes used for specifying dependencies as documented in <xref linkend="ssec-stdenv-dependencies"/>.
We "splice" together the run-time and build-time package sets with <varname>callPackage</varname>, and then <varname>mkDerivation</varname> for each of four attributes pulls the right derivation out.
This splicing can be skipped when not cross compiling as the package sets are the same, but is a bit slow for cross compiling.
Because of this, a best-of-both-worlds solution is in the works with no splicing or explicit access of <varname>buildPackages</varname> needed.
For now, feel free to use either method.
</para>
<note><para>
There is also a "backlink" <varname>targetPackages</varname>, yielding a package set whose <varname>buildPackages</varname> is the current package set.
This is a hack, though, to accommodate compilers with lousy build systems.
Please do not use this unless you are absolutely sure you are packaging such a compiler and there is no other way.
</para></note>
</section> </section>
<section> <section>
<title>Cross packagaing cookbook</title> <title>Cross packagaing cookbook</title>
<para>
Some frequently problems when packaging for cross compilation are good to just spell and answer.
Ideally the information above is exhaustive, so this section cannot provide any new information,
but its ludicrous and cruel to expect everyone to spend effort working through the interaction of many features just to figure out the same answer to the same common problem.
Feel free to add to this list!
</para>
<qandaset>
<qandaentry>
<question><para>
What if my package's build system needs to build a C program to be run under the build environment?
</para></question>
<answer><para>
<programlisting>depsBuildBuild = [ buildPackages.stdenv.cc ];</programlisting>
Add it to your <function>mkDerivation</function> invocation.
</para></answer>
</qandaentry>
<qandaentry>
<question><para>
My package fails to find <command>ar</command>.
</para></question>
<answer><para>
Many packages assume that an unprefixed <command>ar</command> is available, but Nix doesn't provide one.
It only provides a prefixed one, just as it only does for all the other binutils programs.
It may be necessary to patch the package to fix the build system to use a prefixed `ar`.
</para></answer>
</qandaentry>
<qandaentry>
<question><para>
My package's testsuite needs to run host platform code.
</para></question>
<answer><para>
<programlisting>doCheck = stdenv.hostPlatform != stdenv.buildPlatfrom;</programlisting>
Add it to your <function>mkDerivation</function> invocation.
</para></answer>
</qandaentry>
</qandaset>
</section>
</section>
<para>
Some frequently problems when packaging for cross compilation are good to
just spell and answer. Ideally the information above is exhaustive, so this
section cannot provide any new information, but its ludicrous and cruel to
expect everyone to spend effort working through the interaction of many
features just to figure out the same answer to the same common problem.
Feel free to add to this list!
</para>
<qandaset>
<qandaentry>
<question>
<para>
What if my package's build system needs to build a C program to be run
under the build environment?
</para>
</question>
<answer>
<para>
<programlisting>depsBuildBuild = [ buildPackages.stdenv.cc ];</programlisting>
Add it to your <function>mkDerivation</function> invocation.
</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>
My package fails to find <command>ar</command>.
</para>
</question>
<answer>
<para>
Many packages assume that an unprefixed <command>ar</command> is
available, but Nix doesn't provide one. It only provides a prefixed one,
just as it only does for all the other binutils programs. It may be
necessary to patch the package to fix the build system to use a prefixed
`ar`.
</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>
My package's testsuite needs to run host platform code.
</para>
</question>
<answer>
<para>
<programlisting>doCheck = stdenv.hostPlatform != stdenv.buildPlatfrom;</programlisting>
Add it to your <function>mkDerivation</function> invocation.
</para>
</answer>
</qandaentry>
</qandaset>
</section>
</section>
<!--============================================================--> <!--============================================================-->
<section xml:id="sec-cross-usage">
<section xml:id="sec-cross-usage">
<title>Cross-building packages</title> <title>Cross-building packages</title>
<note><para>
More information needs to moved from the old wiki, especially <link xlink:href="https://nixos.org/wiki/CrossCompiling" />, for this section. <note>
</para></note> <para>
More information needs to moved from the old wiki, especially
<link xlink:href="https://nixos.org/wiki/CrossCompiling" />, for this
section.
</para>
</note>
<para> <para>
Nixpkgs can be instantiated with <varname>localSystem</varname> alone, in which case there is no cross compiling and everything is built by and for that system, Nixpkgs can be instantiated with <varname>localSystem</varname> alone, in
or also with <varname>crossSystem</varname>, in which case packages run on the latter, but all building happens on the former. which case there is no cross compiling and everything is built by and for
Both parameters take the same schema as the 3 (build, host, and target) platforms defined in the previous section. that system, or also with <varname>crossSystem</varname>, in which case
As mentioned above, <literal>lib.systems.examples</literal> has some platforms which are used as arguments for these parameters in practice. packages run on the latter, but all building happens on the former. Both
You can use them programmatically, or on the command line: <programlisting> parameters take the same schema as the 3 (build, host, and target) platforms
defined in the previous section. As mentioned above,
<literal>lib.systems.examples</literal> has some platforms which are used as
arguments for these parameters in practice. You can use them
programmatically, or on the command line:
<programlisting>
nix-build &lt;nixpkgs&gt; --arg crossSystem '(import &lt;nixpkgs/lib&gt;).systems.examples.fooBarBaz' -A whatever</programlisting> nix-build &lt;nixpkgs&gt; --arg crossSystem '(import &lt;nixpkgs/lib&gt;).systems.examples.fooBarBaz' -A whatever</programlisting>
</para> </para>
<note> <note>
<para> <para>
Eventually we would like to make these platform examples an unnecessary convenience so that <programlisting> Eventually we would like to make these platform examples an unnecessary
convenience so that
<programlisting>
nix-build &lt;nixpkgs&gt; --arg crossSystem.config '&lt;arch&gt;-&lt;os&gt;-&lt;vendor&gt;-&lt;abi&gt;' -A whatever</programlisting> nix-build &lt;nixpkgs&gt; --arg crossSystem.config '&lt;arch&gt;-&lt;os&gt;-&lt;vendor&gt;-&lt;abi&gt;' -A whatever</programlisting>
works in the vast majority of cases. works in the vast majority of cases. The problem today is dependencies on
The problem today is dependencies on other sorts of configuration which aren't given proper defaults. other sorts of configuration which aren't given proper defaults. We rely on
We rely on the examples to crudely to set those configuration parameters in some vaguely sane manner on the users behalf. the examples to crudely to set those configuration parameters in some
Issue <link xlink:href="https://github.com/NixOS/nixpkgs/issues/34274">#34274</link> tracks this inconvenience along with its root cause in crufty configuration options. vaguely sane manner on the users behalf. Issue
</para> <link xlink:href="https://github.com/NixOS/nixpkgs/issues/34274">#34274</link>
tracks this inconvenience along with its root cause in crufty configuration
options.
</para>
</note> </note>
<para> <para>
While one is free to pass both parameters in full, there's a lot of logic to fill in missing fields. While one is free to pass both parameters in full, there's a lot of logic to
As discussed in the previous section, only one of <varname>system</varname>, <varname>config</varname>, and <varname>parsed</varname> is needed to infer the other two. fill in missing fields. As discussed in the previous section, only one of
Additionally, <varname>libc</varname> will be inferred from <varname>parse</varname>. <varname>system</varname>, <varname>config</varname>, and
Finally, <literal>localSystem.system</literal> is also <emphasis>impurely</emphasis> inferred based on the platform evaluation occurs. <varname>parsed</varname> is needed to infer the other two. Additionally,
This means it is often not necessary to pass <varname>localSystem</varname> at all, as in the command-line example in the previous paragraph. <varname>libc</varname> will be inferred from <varname>parse</varname>.
Finally, <literal>localSystem.system</literal> is also
<emphasis>impurely</emphasis> inferred based on the platform evaluation
occurs. This means it is often not necessary to pass
<varname>localSystem</varname> at all, as in the command-line example in the
previous paragraph.
</para> </para>
<note> <note>
<para> <para>
Many sources (manual, wiki, etc) probably mention passing <varname>system</varname>, <varname>platform</varname>, along with the optional <varname>crossSystem</varname> to nixpkgs: Many sources (manual, wiki, etc) probably mention passing
<literal>import &lt;nixpkgs&gt; { system = ..; platform = ..; crossSystem = ..; }</literal>. <varname>system</varname>, <varname>platform</varname>, along with the
Passing those two instead of <varname>localSystem</varname> is still supported for compatibility, but is discouraged. optional <varname>crossSystem</varname> to nixpkgs: <literal>import
Indeed, much of the inference we do for these parameters is motivated by compatibility as much as convenience. &lt;nixpkgs&gt; { system = ..; platform = ..; crossSystem = ..;
</para> }</literal>. Passing those two instead of <varname>localSystem</varname> is
still supported for compatibility, but is discouraged. Indeed, much of the
inference we do for these parameters is motivated by compatibility as much
as convenience.
</para>
</note> </note>
<para> <para>
One would think that <varname>localSystem</varname> and <varname>crossSystem</varname> overlap horribly with the three <varname>*Platforms</varname> (<varname>buildPlatform</varname>, <varname>hostPlatform,</varname> and <varname>targetPlatform</varname>; see <varname>stage.nix</varname> or the manual). One would think that <varname>localSystem</varname> and
Actually, those identifiers are purposefully not used here to draw a subtle but important distinction: <varname>crossSystem</varname> overlap horribly with the three
While the granularity of having 3 platforms is necessary to properly *build* packages, it is overkill for specifying the user's *intent* when making a build plan or package set. <varname>*Platforms</varname> (<varname>buildPlatform</varname>,
A simple "build vs deploy" dichotomy is adequate: the sliding window principle described in the previous section shows how to interpolate between the these two "end points" to get the 3 platform triple for each bootstrapping stage. <varname>hostPlatform,</varname> and <varname>targetPlatform</varname>; see
That means for any package a given package set, even those not bound on the top level but only reachable via dependencies or <varname>buildPackages</varname>, the three platforms will be defined as one of <varname>localSystem</varname> or <varname>crossSystem</varname>, with the former replacing the latter as one traverses build-time dependencies. <varname>stage.nix</varname> or the manual). Actually, those identifiers are
A last simple difference then is <varname>crossSystem</varname> should be null when one doesn't want to cross-compile, while the <varname>*Platform</varname>s are always non-null. purposefully not used here to draw a subtle but important distinction: While
<varname>localSystem</varname> is always non-null. the granularity of having 3 platforms is necessary to properly *build*
packages, it is overkill for specifying the user's *intent* when making a
build plan or package set. A simple "build vs deploy" dichotomy is adequate:
the sliding window principle described in the previous section shows how to
interpolate between the these two "end points" to get the 3 platform triple
for each bootstrapping stage. That means for any package a given package
set, even those not bound on the top level but only reachable via
dependencies or <varname>buildPackages</varname>, the three platforms will
be defined as one of <varname>localSystem</varname> or
<varname>crossSystem</varname>, with the former replacing the latter as one
traverses build-time dependencies. A last simple difference then is
<varname>crossSystem</varname> should be null when one doesn't want to
cross-compile, while the <varname>*Platform</varname>s are always non-null.
<varname>localSystem</varname> is always non-null.
</para> </para>
</section> </section>
<!--============================================================--> <!--============================================================-->
<section xml:id="sec-cross-infra">
<section xml:id="sec-cross-infra">
<title>Cross-compilation infrastructure</title> <title>Cross-compilation infrastructure</title>
<para>To be written.</para>
<note><para>
If one explores nixpkgs, they will see derivations with names like <literal>gccCross</literal>.
Such <literal>*Cross</literal> derivations is a holdover from before we properly distinguished between the host and target platforms
—the derivation with "Cross" in the name covered the <literal>build = host != target</literal> case, while the other covered the <literal>host = target</literal>, with build platform the same or not based on whether one was using its <literal>.nativeDrv</literal> or <literal>.crossDrv</literal>.
This ugliness will disappear soon.
</para></note>
</section>
<para>
To be written.
</para>
<note>
<para>
If one explores nixpkgs, they will see derivations with names like
<literal>gccCross</literal>. Such <literal>*Cross</literal> derivations is
a holdover from before we properly distinguished between the host and
target platforms —the derivation with "Cross" in the name covered the
<literal>build = host != target</literal> case, while the other covered the
<literal>host = target</literal>, with build platform the same or not based
on whether one was using its <literal>.nativeDrv</literal> or
<literal>.crossDrv</literal>. This ugliness will disappear soon.
</para>
</note>
</section>
</chapter> </chapter>

1096
doc/functions.xml
View File

File diff suppressed because it is too large Load Diff

520
doc/languages-frameworks/beam.xml
View File

@ -1,124 +1,137 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-beam"> xml:id="sec-beam">
<title>BEAM Languages (Erlang, Elixir &amp; LFE)</title>
<title>BEAM Languages (Erlang, Elixir &amp; LFE)</title> <section xml:id="beam-introduction">
<section xml:id="beam-introduction"> <title>Introduction</title>
<title>Introduction</title>
<para>
In this document and related Nix expressions, we use the term,
<emphasis>BEAM</emphasis>, to describe the environment. BEAM is the name
of the Erlang Virtual Machine and, as far as we're concerned, from a
packaging perspective, all languages that run on the BEAM are
interchangeable. That which varies, like the build system, is transparent
to users of any given BEAM package, so we make no distinction.
</para>
</section>
<section xml:id="beam-structure">
<title>Structure</title>
<para>
All BEAM-related expressions are available via the top-level
<literal>beam</literal> attribute, which includes:
</para>
<itemizedlist>
<listitem>
<para>
<literal>interpreters</literal>: a set of compilers running on the
BEAM, including multiple Erlang/OTP versions
(<literal>beam.interpreters.erlangR19</literal>, etc), Elixir
(<literal>beam.interpreters.elixir</literal>) and LFE
(<literal>beam.interpreters.lfe</literal>).
</para>
</listitem>
<listitem>
<para>
<literal>packages</literal>: a set of package sets, each compiled with
a specific Erlang/OTP version, e.g.
<literal>beam.packages.erlangR19</literal>.
</para>
</listitem>
</itemizedlist>
<para>
The default Erlang compiler, defined by
<literal>beam.interpreters.erlang</literal>, is aliased as
<literal>erlang</literal>. The default BEAM package set is defined by
<literal>beam.packages.erlang</literal> and aliased at the top level as
<literal>beamPackages</literal>.
</para>
<para>
To create a package set built with a custom Erlang version, use the
lambda, <literal>beam.packagesWith</literal>, which accepts an Erlang/OTP
derivation and produces a package set similar to
<literal>beam.packages.erlang</literal>.
</para>
<para>
Many Erlang/OTP distributions available in
<literal>beam.interpreters</literal> have versions with ODBC and/or Java
enabled. For example, there's
<literal>beam.interpreters.erlangR19_odbc_javac</literal>, which
corresponds to <literal>beam.interpreters.erlangR19</literal>.
</para>
<para xml:id="erlang-call-package">
We also provide the lambda,
<literal>beam.packages.erlang.callPackage</literal>, which simplifies
writing BEAM package definitions by injecting all packages from
<literal>beam.packages.erlang</literal> into the top-level context.
</para>
</section>
<section xml:id="build-tools">
<title>Build Tools</title>
<section xml:id="build-tools-rebar3">
<title>Rebar3</title>
<para>
By default, Rebar3 wants to manage its own dependencies. This is perfectly
acceptable in the normal, non-Nix setup, but in the Nix world, it is not.
To rectify this, we provide two versions of Rebar3:
<itemizedlist>
<listitem>
<para>
<literal>rebar3</literal>: patched to remove the ability to download
anything. When not running it via <literal>nix-shell</literal> or
<literal>nix-build</literal>, it's probably not going to work as
desired.
</para>
</listitem>
<listitem>
<para>
<literal>rebar3-open</literal>: the normal, unmodified Rebar3. It
should work exactly as would any other version of Rebar3. Any Erlang
package should rely on <literal>rebar3</literal> instead. See <xref
linkend="rebar3-packages"/>.
</para>
</listitem>
</itemizedlist>
</para>
</section>
<section xml:id="build-tools-other">
<title>Mix &amp; Erlang.mk</title>
<para>
Both Mix and Erlang.mk work exactly as expected. There is a bootstrap
process that needs to be run for both, however, which is supported by the
<literal>buildMix</literal> and <literal>buildErlangMk</literal>
derivations, respectively.
</para>
</section>
</section>
<section xml:id="how-to-install-beam-packages">
<title>How to Install BEAM Packages</title>
<para> <para>
BEAM packages are not registered at the top level, simply because they are In this document and related Nix expressions, we use the term,
not relevant to the vast majority of Nix users. They are installable using <emphasis>BEAM</emphasis>, to describe the environment. BEAM is the name of
the <literal>beam.packages.erlang</literal> attribute set (aliased as the Erlang Virtual Machine and, as far as we're concerned, from a packaging
<literal>beamPackages</literal>), which points to packages built by the perspective, all languages that run on the BEAM are interchangeable. That
default Erlang/OTP version in Nixpkgs, as defined by which varies, like the build system, is transparent to users of any given
<literal>beam.interpreters.erlang</literal>. BEAM package, so we make no distinction.
</para>
</section>
To list the available packages in <section xml:id="beam-structure">
<literal>beamPackages</literal>, use the following command: <title>Structure</title>
<para>
All BEAM-related expressions are available via the top-level
<literal>beam</literal> attribute, which includes:
</para> </para>
<programlisting> <itemizedlist>
<listitem>
<para>
<literal>interpreters</literal>: a set of compilers running on the BEAM,
including multiple Erlang/OTP versions
(<literal>beam.interpreters.erlangR19</literal>, etc), Elixir
(<literal>beam.interpreters.elixir</literal>) and LFE
(<literal>beam.interpreters.lfe</literal>).
</para>
</listitem>
<listitem>
<para>
<literal>packages</literal>: a set of package sets, each compiled with a
specific Erlang/OTP version, e.g.
<literal>beam.packages.erlangR19</literal>.
</para>
</listitem>
</itemizedlist>
<para>
The default Erlang compiler, defined by
<literal>beam.interpreters.erlang</literal>, is aliased as
<literal>erlang</literal>. The default BEAM package set is defined by
<literal>beam.packages.erlang</literal> and aliased at the top level as
<literal>beamPackages</literal>.
</para>
<para>
To create a package set built with a custom Erlang version, use the lambda,
<literal>beam.packagesWith</literal>, which accepts an Erlang/OTP derivation
and produces a package set similar to
<literal>beam.packages.erlang</literal>.
</para>
<para>
Many Erlang/OTP distributions available in
<literal>beam.interpreters</literal> have versions with ODBC and/or Java
enabled. For example, there's
<literal>beam.interpreters.erlangR19_odbc_javac</literal>, which corresponds
to <literal>beam.interpreters.erlangR19</literal>.
</para>
<para xml:id="erlang-call-package">
We also provide the lambda,
<literal>beam.packages.erlang.callPackage</literal>, which simplifies
writing BEAM package definitions by injecting all packages from
<literal>beam.packages.erlang</literal> into the top-level context.
</para>
</section>
<section xml:id="build-tools">
<title>Build Tools</title>
<section xml:id="build-tools-rebar3">
<title>Rebar3</title>
<para>
By default, Rebar3 wants to manage its own dependencies. This is perfectly
acceptable in the normal, non-Nix setup, but in the Nix world, it is not.
To rectify this, we provide two versions of Rebar3:
<itemizedlist>
<listitem>
<para>
<literal>rebar3</literal>: patched to remove the ability to download
anything. When not running it via <literal>nix-shell</literal> or
<literal>nix-build</literal>, it's probably not going to work as
desired.
</para>
</listitem>
<listitem>
<para>
<literal>rebar3-open</literal>: the normal, unmodified Rebar3. It should
work exactly as would any other version of Rebar3. Any Erlang package
should rely on <literal>rebar3</literal> instead. See
<xref
linkend="rebar3-packages"/>.
</para>
</listitem>
</itemizedlist>
</para>
</section>
<section xml:id="build-tools-other">
<title>Mix &amp; Erlang.mk</title>
<para>
Both Mix and Erlang.mk work exactly as expected. There is a bootstrap
process that needs to be run for both, however, which is supported by the
<literal>buildMix</literal> and <literal>buildErlangMk</literal>
derivations, respectively.
</para>
</section>
</section>
<section xml:id="how-to-install-beam-packages">
<title>How to Install BEAM Packages</title>
<para>
BEAM packages are not registered at the top level, simply because they are
not relevant to the vast majority of Nix users. They are installable using
the <literal>beam.packages.erlang</literal> attribute set (aliased as
<literal>beamPackages</literal>), which points to packages built by the
default Erlang/OTP version in Nixpkgs, as defined by
<literal>beam.interpreters.erlang</literal>. To list the available packages
in <literal>beamPackages</literal>, use the following command:
</para>
<programlisting>
$ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -qaP -A beamPackages $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -qaP -A beamPackages
beamPackages.esqlite esqlite-0.2.1 beamPackages.esqlite esqlite-0.2.1
beamPackages.goldrush goldrush-0.1.7 beamPackages.goldrush goldrush-0.1.7
@ -128,34 +141,43 @@ beamPackages.lager lager-3.0.2
beamPackages.meck meck-0.8.3 beamPackages.meck meck-0.8.3
beamPackages.rebar3-pc pc-1.1.0 beamPackages.rebar3-pc pc-1.1.0
</programlisting> </programlisting>
<para> <para>
To install any of those packages into your profile, refer to them by their To install any of those packages into your profile, refer to them by their
attribute path (first column): attribute path (first column):
</para> </para>
<programlisting>
<programlisting>
$ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse
</programlisting> </programlisting>
<para> <para>
The attribute path of any BEAM package corresponds to the name of that The attribute path of any BEAM package corresponds to the name of that
particular package in <link xlink:href="https://hex.pm">Hex</link> or its particular package in <link xlink:href="https://hex.pm">Hex</link> or its
OTP Application/Release name. OTP Application/Release name.
</para> </para>
</section> </section>
<section xml:id="packaging-beam-applications">
<section xml:id="packaging-beam-applications">
<title>Packaging BEAM Applications</title> <title>Packaging BEAM Applications</title>
<section xml:id="packaging-erlang-applications"> <section xml:id="packaging-erlang-applications">
<title>Erlang Applications</title> <title>Erlang Applications</title>
<section xml:id="rebar3-packages">
<title>Rebar3 Packages</title> <section xml:id="rebar3-packages">
<para> <title>Rebar3 Packages</title>
The Nix function, <literal>buildRebar3</literal>, defined in
<literal>beam.packages.erlang.buildRebar3</literal> and aliased at the <para>
top level, can be used to build a derivation that understands how to The Nix function, <literal>buildRebar3</literal>, defined in
build a Rebar3 project. For example, we can build <link <literal>beam.packages.erlang.buildRebar3</literal> and aliased at the top
xlink:href="https://github.com/erlang-nix/hex2nix">hex2nix</link> as level, can be used to build a derivation that understands how to build a
follows: Rebar3 project. For example, we can build
</para> <link
<programlisting> xlink:href="https://github.com/erlang-nix/hex2nix">hex2nix</link>
as follows:
</para>
<programlisting>
{ stdenv, fetchFromGitHub, buildRebar3, ibrowse, jsx, erlware_commons }: { stdenv, fetchFromGitHub, buildRebar3, ibrowse, jsx, erlware_commons }:
buildRebar3 rec { buildRebar3 rec {
@ -172,33 +194,40 @@ $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse
beamDeps = [ ibrowse jsx erlware_commons ]; beamDeps = [ ibrowse jsx erlware_commons ];
} }
</programlisting> </programlisting>
<para>
Such derivations are callable with <para>
<literal>beam.packages.erlang.callPackage</literal> (see <xref Such derivations are callable with
linkend="erlang-call-package"/>). To call this package using the normal <literal>beam.packages.erlang.callPackage</literal> (see
<literal>callPackage</literal>, refer to dependency packages via <xref
<literal>beamPackages</literal>, e.g. linkend="erlang-call-package"/>). To call this package using
<literal>beamPackages.ibrowse</literal>. the normal <literal>callPackage</literal>, refer to dependency packages
</para> via <literal>beamPackages</literal>, e.g.
<para> <literal>beamPackages.ibrowse</literal>.
Notably, <literal>buildRebar3</literal> includes </para>
<literal>beamDeps</literal>, while
<literal>stdenv.mkDerivation</literal> does not. BEAM dependencies added <para>
there will be correctly handled by the system. Notably, <literal>buildRebar3</literal> includes
</para> <literal>beamDeps</literal>, while <literal>stdenv.mkDerivation</literal>
<para> does not. BEAM dependencies added there will be correctly handled by the
If a package needs to compile native code via Rebar3's port compilation system.
mechanism, add <literal>compilePort = true;</literal> to the derivation. </para>
</para>
</section> <para>
<section xml:id="erlang-mk-packages"> If a package needs to compile native code via Rebar3's port compilation
<title>Erlang.mk Packages</title> mechanism, add <literal>compilePort = true;</literal> to the derivation.
<para> </para>
Erlang.mk functions similarly to Rebar3, except we use </section>
<literal>buildErlangMk</literal> instead of
<literal>buildRebar3</literal>. <section xml:id="erlang-mk-packages">
</para> <title>Erlang.mk Packages</title>
<programlisting>
<para>
Erlang.mk functions similarly to Rebar3, except we use
<literal>buildErlangMk</literal> instead of
<literal>buildRebar3</literal>.
</para>
<programlisting>
{ buildErlangMk, fetchHex, cowlib, ranch }: { buildErlangMk, fetchHex, cowlib, ranch }:
buildErlangMk { buildErlangMk {
@ -222,14 +251,17 @@ $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse
}; };
} }
</programlisting> </programlisting>
</section> </section>
<section xml:id="mix-packages">
<title>Mix Packages</title> <section xml:id="mix-packages">
<para> <title>Mix Packages</title>
Mix functions similarly to Rebar3, except we use
<literal>buildMix</literal> instead of <literal>buildRebar3</literal>. <para>
</para> Mix functions similarly to Rebar3, except we use
<programlisting> <literal>buildMix</literal> instead of <literal>buildRebar3</literal>.
</para>
<programlisting>
{ buildMix, fetchHex, plug, absinthe }: { buildMix, fetchHex, plug, absinthe }:
buildMix { buildMix {
@ -253,10 +285,12 @@ $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse
}; };
} }
</programlisting> </programlisting>
<para>
Alternatively, we can use <literal>buildHex</literal> as a shortcut: <para>
</para> Alternatively, we can use <literal>buildHex</literal> as a shortcut:
<programlisting> </para>
<programlisting>
{ buildHex, buildMix, plug, absinthe }: { buildHex, buildMix, plug, absinthe }:
buildHex { buildHex {
@ -278,21 +312,25 @@ $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse
}; };
} }
</programlisting> </programlisting>
</section> </section>
</section> </section>
</section> </section>
<section xml:id="how-to-develop">
<section xml:id="how-to-develop">
<title>How to Develop</title> <title>How to Develop</title>
<section xml:id="accessing-an-environment"> <section xml:id="accessing-an-environment">
<title>Accessing an Environment</title> <title>Accessing an Environment</title>
<para>
Often, we simply want to access a valid environment that contains a <para>
specific package and its dependencies. We can accomplish that with the Often, we simply want to access a valid environment that contains a
<literal>env</literal> attribute of a derivation. For example, let's say specific package and its dependencies. We can accomplish that with the
we want to access an Erlang REPL with <literal>ibrowse</literal> loaded <literal>env</literal> attribute of a derivation. For example, let's say we
up. We could do the following: want to access an Erlang REPL with <literal>ibrowse</literal> loaded up. We
</para> could do the following:
<programlisting> </para>
<programlisting>
$ nix-shell -A beamPackages.ibrowse.env --run "erl" $ nix-shell -A beamPackages.ibrowse.env --run "erl"
Erlang/OTP 18 [erts-7.0] [source] [64-bit] [smp:4:4] [async-threads:10] [hipe] [kernel-poll:false] Erlang/OTP 18 [erts-7.0] [source] [64-bit] [smp:4:4] [async-threads:10] [hipe] [kernel-poll:false]
@ -333,22 +371,25 @@ $ nix-env -f &quot;&lt;nixpkgs&gt;&quot; -iA beamPackages.ibrowse
ok ok
2> 2>
</programlisting> </programlisting>
<para>
Notice the <literal>-A beamPackages.ibrowse.env</literal>. That is the key <para>
to this functionality. Notice the <literal>-A beamPackages.ibrowse.env</literal>. That is the key
</para> to this functionality.
</para>
</section> </section>
<section xml:id="creating-a-shell"> <section xml:id="creating-a-shell">
<title>Creating a Shell</title> <title>Creating a Shell</title>
<para>
Getting access to an environment often isn't enough to do real <para>
development. Usually, we need to create a <literal>shell.nix</literal> Getting access to an environment often isn't enough to do real development.
file and do our development inside of the environment specified therein. Usually, we need to create a <literal>shell.nix</literal> file and do our
This file looks a lot like the packaging described above, except that development inside of the environment specified therein. This file looks a
<literal>src</literal> points to the project root and we call the package lot like the packaging described above, except that <literal>src</literal>
directly. points to the project root and we call the package directly.
</para> </para>
<programlisting>
<programlisting>
{ pkgs ? import &quot;&lt;nixpkgs&quot;&gt; {} }: { pkgs ? import &quot;&lt;nixpkgs&quot;&gt; {} }:
with pkgs; with pkgs;
@ -368,13 +409,16 @@ in
drv drv
</programlisting> </programlisting>
<section xml:id="building-in-a-shell">
<section xml:id="building-in-a-shell">
<title>Building in a Shell (for Mix Projects)</title> <title>Building in a Shell (for Mix Projects)</title>
<para> <para>
We can leverage the support of the derivation, irrespective of the build We can leverage the support of the derivation, irrespective of the build
derivation, by calling the commands themselves. derivation, by calling the commands themselves.
</para> </para>
<programlisting>
<programlisting>
# ============================================================================= # =============================================================================
# Variables # Variables
# ============================================================================= # =============================================================================
@ -431,44 +475,54 @@ analyze: build plt
$(NIX_SHELL) --run "mix dialyzer --no-compile" $(NIX_SHELL) --run "mix dialyzer --no-compile"
</programlisting> </programlisting>
<para> <para>
Using a <literal>shell.nix</literal> as described (see <xref Using a <literal>shell.nix</literal> as described (see
<xref
linkend="creating-a-shell"/>) should just work. Aside from linkend="creating-a-shell"/>) should just work. Aside from
<literal>test</literal>, <literal>plt</literal>, and <literal>test</literal>, <literal>plt</literal>, and
<literal>analyze</literal>, the Make targets work just fine for all of the <literal>analyze</literal>, the Make targets work just fine for all of the
build derivations. build derivations.
</para> </para>
</section>
</section> </section>
</section> </section>
</section>
<section xml:id="generating-packages-from-hex-with-hex2nix"> <section xml:id="generating-packages-from-hex-with-hex2nix">
<title>Generating Packages from Hex with <literal>hex2nix</literal></title> <title>Generating Packages from Hex with <literal>hex2nix</literal></title>
<para> <para>
Updating the <link xlink:href="https://hex.pm">Hex</link> package set Updating the <link xlink:href="https://hex.pm">Hex</link> package set
requires <link requires
xlink:href="https://github.com/erlang-nix/hex2nix">hex2nix</link>. Given the <link
path to the Erlang modules (usually xlink:href="https://github.com/erlang-nix/hex2nix">hex2nix</link>.
<literal>pkgs/development/erlang-modules</literal>), it will dump a file Given the path to the Erlang modules (usually
called <literal>hex-packages.nix</literal>, containing all the packages that <literal>pkgs/development/erlang-modules</literal>), it will dump a file
use a recognized build system in <link called <literal>hex-packages.nix</literal>, containing all the packages that
xlink:href="https://hex.pm">Hex</link>. It can't be determined, however, use a recognized build system in
whether every package is buildable. <link
</para> xlink:href="https://hex.pm">Hex</link>. It can't be determined,
<para> however, whether every package is buildable.
To make life easier for our users, try to build every <link </para>
xlink:href="https://hex.pm">Hex</link> package and remove those that fail.
To do that, simply run the following command in the root of your <para>
<literal>nixpkgs</literal> repository: To make life easier for our users, try to build every
</para> <link
<programlisting> xlink:href="https://hex.pm">Hex</link> package and remove those
that fail. To do that, simply run the following command in the root of your
<literal>nixpkgs</literal> repository:
</para>
<programlisting>
$ nix-build -A beamPackages $ nix-build -A beamPackages
</programlisting> </programlisting>
<para>
That will attempt to build every package in <para>
<literal>beamPackages</literal>. Then manually remove those that fail. That will attempt to build every package in <literal>beamPackages</literal>.
Hopefully, someone will improve <link Then manually remove those that fail. Hopefully, someone will improve
xlink:href="https://github.com/erlang-nix/hex2nix">hex2nix</link> in the <link
future to automate the process. xlink:href="https://github.com/erlang-nix/hex2nix">hex2nix</link>
</para> in the future to automate the process.
</section> </para>
</section>
</section> </section>

279
doc/languages-frameworks/bower.xml
View File

@ -1,40 +1,37 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-bower"> xml:id="sec-bower">
<title>Bower</title>
<title>Bower</title> <para>
<link xlink:href="http://bower.io">Bower</link> is a package manager for web
site front-end components. Bower packages (comprising of build artefacts and
sometimes sources) are stored in <command>git</command> repositories,
typically on Github. The package registry is run by the Bower team with
package metadata coming from the <filename>bower.json</filename> file within
each package.
</para>
<para> <para>
<link xlink:href="http://bower.io">Bower</link> is a package manager The end result of running Bower is a <filename>bower_components</filename>
for web site front-end components. Bower packages (comprising of directory which can be included in the web app's build process.
build artefacts and sometimes sources) are stored in </para>
<command>git</command> repositories, typically on Github. The
package registry is run by the Bower team with package metadata
coming from the <filename>bower.json</filename> file within each
package.
</para>
<para> <para>
The end result of running Bower is a
<filename>bower_components</filename> directory which can be included
in the web app's build process.
</para>
<para>
Bower can be run interactively, by installing Bower can be run interactively, by installing
<varname>nodePackages.bower</varname>. More interestingly, the Bower <varname>nodePackages.bower</varname>. More interestingly, the Bower
components can be declared in a Nix derivation, with the help of components can be declared in a Nix derivation, with the help of
<varname>nodePackages.bower2nix</varname>. <varname>nodePackages.bower2nix</varname>.
</para> </para>
<section xml:id="ssec-bower2nix-usage"> <section xml:id="ssec-bower2nix-usage">
<title><command>bower2nix</command> usage</title> <title><command>bower2nix</command> usage</title>
<para> <para>
Suppose you have a <filename>bower.json</filename> with the following contents: Suppose you have a <filename>bower.json</filename> with the following
contents:
<example xml:id="ex-bowerJson">
<example xml:id="ex-bowerJson"><title><filename>bower.json</filename></title> <title><filename>bower.json</filename></title>
<programlisting language="json"> <programlisting language="json">
<![CDATA[{ <![CDATA[{
"name": "my-web-app", "name": "my-web-app",
@ -44,14 +41,12 @@
} }
}]]> }]]>
</programlisting> </programlisting>
</example> </example>
</para> </para>
<para>
Running <command>bower2nix</command> will produce something like the
following output:
<para>
Running <command>bower2nix</command> will produce something like the
following output:
<programlisting language="nix"> <programlisting language="nix">
<![CDATA[{ fetchbower, buildEnv }: <![CDATA[{ fetchbower, buildEnv }:
buildEnv { name = "bower-env"; ignoreCollisions = true; paths = [ buildEnv { name = "bower-env"; ignoreCollisions = true; paths = [
@ -60,31 +55,31 @@ buildEnv { name = "bower-env"; ignoreCollisions = true; paths = [
(fetchbower "jquery" "2.2.2" "1.9.1 - 2" "10sp5h98sqwk90y4k6hbdviwqzvzwqf47r3r51pakch5ii2y7js1") (fetchbower "jquery" "2.2.2" "1.9.1 - 2" "10sp5h98sqwk90y4k6hbdviwqzvzwqf47r3r51pakch5ii2y7js1")
]; }]]> ]; }]]>
</programlisting> </programlisting>
</para> </para>
<para>
Using the <command>bower2nix</command> command line arguments, the
output can be redirected to a file. A name like
<filename>bower-packages.nix</filename> would be fine.
</para>
<para>
The resulting derivation is a union of all the downloaded Bower
packages (and their dependencies). To use it, they still need to be
linked together by Bower, which is where
<varname>buildBowerComponents</varname> is useful.
</para>
</section>
<section xml:id="ssec-build-bower-components"><title><varname>buildBowerComponents</varname> function</title>
<para> <para>
The function is implemented in <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/bower-modules/generic/default.nix"> Using the <command>bower2nix</command> command line arguments, the output
<filename>pkgs/development/bower-modules/generic/default.nix</filename></link>. can be redirected to a file. A name like
Example usage: <filename>bower-packages.nix</filename> would be fine.
</para>
<example xml:id="ex-buildBowerComponents"><title>buildBowerComponents</title> <para>
The resulting derivation is a union of all the downloaded Bower packages
(and their dependencies). To use it, they still need to be linked together
by Bower, which is where <varname>buildBowerComponents</varname> is useful.
</para>
</section>
<section xml:id="ssec-build-bower-components">
<title><varname>buildBowerComponents</varname> function</title>
<para>
The function is implemented in
<link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/bower-modules/generic/default.nix">
<filename>pkgs/development/bower-modules/generic/default.nix</filename></link>.
Example usage:
<example xml:id="ex-buildBowerComponents">
<title>buildBowerComponents</title>
<programlisting language="nix"> <programlisting language="nix">
bowerComponents = buildBowerComponents { bowerComponents = buildBowerComponents {
name = "my-web-app"; name = "my-web-app";
@ -92,42 +87,42 @@ bowerComponents = buildBowerComponents {
src = myWebApp; <co xml:id="ex-buildBowerComponents-2" /> src = myWebApp; <co xml:id="ex-buildBowerComponents-2" />
}; };
</programlisting> </programlisting>
</example> </example>
</para> </para>
<para> <para>
In <xref linkend="ex-buildBowerComponents" />, the following arguments In <xref linkend="ex-buildBowerComponents" />, the following arguments are
are of special significance to the function: of special significance to the function:
<calloutlist>
<callout arearefs="ex-buildBowerComponents-1">
<para>
<varname>generated</varname> specifies the file which was created by
<command>bower2nix</command>.
</para>
</callout>
<callout arearefs="ex-buildBowerComponents-2">
<para>
<varname>src</varname> is your project's sources. It needs to contain a
<filename>bower.json</filename> file.
</para>
</callout>
</calloutlist>
</para>
<calloutlist> <para>
<callout arearefs="ex-buildBowerComponents-1"> <varname>buildBowerComponents</varname> will run Bower to link together the
<para> output of <command>bower2nix</command>, resulting in a
<varname>generated</varname> specifies the file which was created by <command>bower2nix</command>. <filename>bower_components</filename> directory which can be used.
</para> </para>
</callout>
<callout arearefs="ex-buildBowerComponents-2"> <para>
<para> Here is an example of a web frontend build process using
<varname>src</varname> is your project's sources. It needs to <command>gulp</command>. You might use <command>grunt</command>, or anything
contain a <filename>bower.json</filename> file. else.
</para> </para>
</callout>
</calloutlist>
</para>
<para> <example xml:id="ex-bowerGulpFile">
<varname>buildBowerComponents</varname> will run Bower to link <title>Example build script (<filename>gulpfile.js</filename>)</title>
together the output of <command>bower2nix</command>, resulting in a
<filename>bower_components</filename> directory which can be used.
</para>
<para>
Here is an example of a web frontend build process using
<command>gulp</command>. You might use <command>grunt</command>, or
anything else.
</para>
<example xml:id="ex-bowerGulpFile"><title>Example build script (<filename>gulpfile.js</filename>)</title>
<programlisting language="javascript"> <programlisting language="javascript">
<![CDATA[var gulp = require('gulp'); <![CDATA[var gulp = require('gulp');
@ -142,10 +137,10 @@ gulp.task('build', [], function () {
.pipe(gulp.dest("./gulpdist/")); .pipe(gulp.dest("./gulpdist/"));
});]]> });]]>
</programlisting> </programlisting>
</example> </example>
<example xml:id="ex-buildBowerComponentsDefaultNix"> <example xml:id="ex-buildBowerComponentsDefaultNix">
<title>Full example — <filename>default.nix</filename></title> <title>Full example — <filename>default.nix</filename></title>
<programlisting language="nix"> <programlisting language="nix">
{ myWebApp ? { outPath = ./.; name = "myWebApp"; } { myWebApp ? { outPath = ./.; name = "myWebApp"; }
, pkgs ? import &lt;nixpkgs&gt; {} , pkgs ? import &lt;nixpkgs&gt; {}
@ -172,73 +167,63 @@ pkgs.stdenv.mkDerivation {
installPhase = "mv gulpdist $out"; installPhase = "mv gulpdist $out";
} }
</programlisting> </programlisting>
</example> </example>
<para> <para>
A few notes about <xref linkend="ex-buildBowerComponentsDefaultNix" />: A few notes about <xref linkend="ex-buildBowerComponentsDefaultNix" />:
<calloutlist>
<calloutlist> <callout arearefs="ex-buildBowerComponentsDefault-1">
<callout arearefs="ex-buildBowerComponentsDefault-1"> <para>
<para> The result of <varname>buildBowerComponents</varname> is an input to the
The result of <varname>buildBowerComponents</varname> is an frontend build.
input to the frontend build. </para>
</para> </callout>
</callout> <callout arearefs="ex-buildBowerComponentsDefault-2">
<para>
<callout arearefs="ex-buildBowerComponentsDefault-2"> Whether to symlink or copy the <filename>bower_components</filename>
<para> directory depends on the build tool in use. In this case a copy is used
Whether to symlink or copy the to avoid <command>gulp</command> silliness with permissions.
<filename>bower_components</filename> directory depends on the </para>
build tool in use. In this case a copy is used to avoid </callout>
<command>gulp</command> silliness with permissions. <callout arearefs="ex-buildBowerComponentsDefault-3">
</para> <para>
</callout> <command>gulp</command> requires <varname>HOME</varname> to refer to a
writeable directory.
<callout arearefs="ex-buildBowerComponentsDefault-3"> </para>
<para> </callout>
<command>gulp</command> requires <varname>HOME</varname> to <callout arearefs="ex-buildBowerComponentsDefault-4">
refer to a writeable directory. <para>
</para>
</callout>
<callout arearefs="ex-buildBowerComponentsDefault-4">
<para>
The actual build command. Other tools could be used. The actual build command. Other tools could be used.
</para> </para>
</callout> </callout>
</calloutlist> </calloutlist>
</para> </para>
</section> </section>
<section xml:id="ssec-bower2nix-troubleshooting"> <section xml:id="ssec-bower2nix-troubleshooting">
<title>Troubleshooting</title> <title>Troubleshooting</title>
<variablelist> <variablelist>
<varlistentry>
<varlistentry> <term><literal>ENOCACHE</literal> errors from
<term>
<literal>ENOCACHE</literal> errors from
<varname>buildBowerComponents</varname> <varname>buildBowerComponents</varname>
</term> </term>
<listitem> <listitem>
<para> <para>
This means that Bower was looking for a package version which This means that Bower was looking for a package version which doesn't
doesn't exist in the generated exist in the generated <filename>bower-packages.nix</filename>.
<filename>bower-packages.nix</filename>. </para>
</para> <para>
<para> If <filename>bower.json</filename> has been updated, then run
If <filename>bower.json</filename> has been updated, then run <command>bower2nix</command> again.
<command>bower2nix</command> again. </para>
</para> <para>
<para> It could also be a bug in <command>bower2nix</command> or
It could also be a bug in <command>bower2nix</command> or <command>fetchbower</command>. If possible, try reformulating the version
<command>fetchbower</command>. If possible, try reformulating specification in <filename>bower.json</filename>.
the version specification in <filename>bower.json</filename>. </para>
</para>
</listitem> </listitem>
</varlistentry> </varlistentry>
</variablelist> </variablelist>
</section>
</section>
</section> </section>

60
doc/languages-frameworks/coq.xml
View File

@ -1,36 +1,38 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-coq"> xml:id="sec-language-coq">
<title>Coq</title>
<title>Coq</title> <para>
<para> Coq libraries should be installed in
Coq libraries should be installed in <literal>$(out)/lib/coq/${coq.coq-version}/user-contrib/</literal>. Such
<literal>$(out)/lib/coq/${coq.coq-version}/user-contrib/</literal>. directories are automatically added to the <literal>$COQPATH</literal>
Such directories are automatically added to the environment variable by the hook defined in the Coq derivation.
<literal>$COQPATH</literal> environment variable by the hook defined </para>
in the Coq derivation.
</para> <para>
<para> Some libraries require OCaml and sometimes also Camlp5 or findlib. The exact
Some libraries require OCaml and sometimes also Camlp5 or findlib. versions that were used to build Coq are saved in the
The exact versions that were used to build Coq are saved in the <literal>coq.ocaml</literal> and <literal>coq.camlp5</literal> and
<literal>coq.ocaml</literal> and <literal>coq.camlp5</literal> <literal>coq.findlib</literal> attributes.
and <literal>coq.findlib</literal> attributes. </para>
</para>
<para> <para>
Coq libraries may be compatible with some specific versions of Coq only. Coq libraries may be compatible with some specific versions of Coq only. The
The <literal>compatibleCoqVersions</literal> attribute is used to <literal>compatibleCoqVersions</literal> attribute is used to precisely
precisely select those versions of Coq that are compatible with this select those versions of Coq that are compatible with this derivation.
derivation. </para>
</para>
<para> <para>
Here is a simple package example. It is a pure Coq library, thus it Here is a simple package example. It is a pure Coq library, thus it depends
depends on Coq. It builds on the Mathematical Components library, thus it on Coq. It builds on the Mathematical Components library, thus it also takes
also takes <literal>mathcomp</literal> as <literal>buildInputs</literal>. <literal>mathcomp</literal> as <literal>buildInputs</literal>. Its
Its <literal>Makefile</literal> has been generated using <literal>Makefile</literal> has been generated using
<literal>coq_makefile</literal> so we only have to <literal>coq_makefile</literal> so we only have to set the
set the <literal>$COQLIB</literal> variable at install time. <literal>$COQLIB</literal> variable at install time.
</para> </para>
<programlisting>
<programlisting>
{ stdenv, fetchFromGitHub, coq, mathcomp }: { stdenv, fetchFromGitHub, coq, mathcomp }:
stdenv.mkDerivation rec { stdenv.mkDerivation rec {

164
doc/languages-frameworks/go.xml
View File

@ -1,14 +1,14 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-go"> xml:id="sec-language-go">
<title>Go</title>
<title>Go</title> <para>
The function <varname>buildGoPackage</varname> builds standard Go programs.
</para>
<para>The function <varname>buildGoPackage</varname> builds <example xml:id='ex-buildGoPackage'>
standard Go programs. <title>buildGoPackage</title>
</para>
<example xml:id='ex-buildGoPackage'><title>buildGoPackage</title>
<programlisting> <programlisting>
deis = buildGoPackage rec { deis = buildGoPackage rec {
name = "deis-${version}"; name = "deis-${version}";
@ -29,55 +29,56 @@ deis = buildGoPackage rec {
buildFlags = "--tags release"; <co xml:id='ex-buildGoPackage-4' /> buildFlags = "--tags release"; <co xml:id='ex-buildGoPackage-4' />
} }
</programlisting> </programlisting>
</example> </example>
<para><xref linkend='ex-buildGoPackage'/> is an example expression using buildGoPackage, <para>
the following arguments are of special significance to the function: <xref linkend='ex-buildGoPackage'/> is an example expression using
buildGoPackage, the following arguments are of special significance to the
<calloutlist> function:
<calloutlist>
<callout arearefs='ex-buildGoPackage-1'> <callout arearefs='ex-buildGoPackage-1'>
<para> <para>
<varname>goPackagePath</varname> specifies the package's canonical Go import path. <varname>goPackagePath</varname> specifies the package's canonical Go
import path.
</para> </para>
</callout> </callout>
<callout arearefs='ex-buildGoPackage-2'>
<callout arearefs='ex-buildGoPackage-2'>
<para> <para>
<varname>subPackages</varname> limits the builder from building child packages that <varname>subPackages</varname> limits the builder from building child
have not been listed. If <varname>subPackages</varname> is not specified, all child packages that have not been listed. If <varname>subPackages</varname> is
packages will be built. not specified, all child packages will be built.
</para> </para>
<para> <para>
In this example only <literal>github.com/deis/deis/client</literal> will be built. In this example only <literal>github.com/deis/deis/client</literal> will
be built.
</para> </para>
</callout> </callout>
<callout arearefs='ex-buildGoPackage-3'>
<callout arearefs='ex-buildGoPackage-3'>
<para> <para>
<varname>goDeps</varname> is where the Go dependencies of a Go program are listed <varname>goDeps</varname> is where the Go dependencies of a Go program are
as a list of package source identified by Go import path. listed as a list of package source identified by Go import path. It could
It could be imported as a separate <varname>deps.nix</varname> file for be imported as a separate <varname>deps.nix</varname> file for
readability. The dependency data structure is described below. readability. The dependency data structure is described below.
</para> </para>
</callout> </callout>
<callout arearefs='ex-buildGoPackage-4'>
<callout arearefs='ex-buildGoPackage-4'>
<para> <para>
<varname>buildFlags</varname> is a list of flags passed to the go build command. <varname>buildFlags</varname> is a list of flags passed to the go build
command.
</para> </para>
</callout> </callout>
</calloutlist>
</para>
</calloutlist> <para>
The <varname>goDeps</varname> attribute can be imported from a separate
<varname>nix</varname> file that defines which Go libraries are needed and
should be included in <varname>GOPATH</varname> for
<varname>buildPhase</varname>.
</para>
</para> <example xml:id='ex-goDeps'>
<title>deps.nix</title>
<para>The <varname>goDeps</varname> attribute can be imported from a separate
<varname>nix</varname> file that defines which Go libraries are needed and should
be included in <varname>GOPATH</varname> for <varname>buildPhase</varname>.
</para>
<example xml:id='ex-goDeps'><title>deps.nix</title>
<programlisting> <programlisting>
[ <co xml:id='ex-goDeps-1' /> [ <co xml:id='ex-goDeps-1' />
{ {
@ -100,67 +101,60 @@ the following arguments are of special significance to the function:
} }
] ]
</programlisting> </programlisting>
</example> </example>
<para> <para>
<calloutlist>
<calloutlist> <callout arearefs='ex-goDeps-1'>
<callout arearefs='ex-goDeps-1'>
<para> <para>
<varname>goDeps</varname> is a list of Go dependencies. <varname>goDeps</varname> is a list of Go dependencies.
</para> </para>
</callout> </callout>
<callout arearefs='ex-goDeps-2'>
<callout arearefs='ex-goDeps-2'>
<para> <para>
<varname>goPackagePath</varname> specifies Go package import path. <varname>goPackagePath</varname> specifies Go package import path.
</para> </para>
</callout> </callout>
<callout arearefs='ex-goDeps-3'>
<callout arearefs='ex-goDeps-3'>
<para> <para>
<varname>fetch type</varname> that needs to be used to get package source. If <varname>git</varname> <varname>fetch type</varname> that needs to be used to get package source.
is used there should be <varname>url</varname>, <varname>rev</varname> and <varname>sha256</varname> If <varname>git</varname> is used there should be <varname>url</varname>,
defined next to it. <varname>rev</varname> and <varname>sha256</varname> defined next to it.
</para> </para>
</callout> </callout>
</calloutlist>
</para>
</calloutlist> <para>
To extract dependency information from a Go package in automated way use
<link xlink:href="https://github.com/kamilchm/go2nix">go2nix</link>. It can
produce complete derivation and <varname>goDeps</varname> file for Go
programs.
</para>
</para> <para>
<varname>buildGoPackage</varname> produces
<para>To extract dependency information from a Go package in automated way use <link xlink:href="https://github.com/kamilchm/go2nix">go2nix</link>. <xref linkend='chap-multiple-output' xrefstyle="select: title" /> where
It can produce complete derivation and <varname>goDeps</varname> file for Go programs.</para> <varname>bin</varname> includes program binaries. You can test build a Go
binary as follows:
<para> <screen>
<varname>buildGoPackage</varname> produces <xref linkend='chap-multiple-output' xrefstyle="select: title" />
where <varname>bin</varname> includes program binaries. You can test build a Go binary as follows:
<screen>
$ nix-build -A deis.bin $ nix-build -A deis.bin
</screen> </screen>
or build all outputs with: or build all outputs with:
<screen>
<screen>
$ nix-build -A deis.all $ nix-build -A deis.all
</screen> </screen>
<varname>bin</varname> output will be installed by default with
<varname>nix-env -i</varname> or <varname>systemPackages</varname>.
</para>
<varname>bin</varname> output will be installed by default with <varname>nix-env -i</varname> <para>
or <varname>systemPackages</varname>. You may use Go packages installed into the active Nix profiles by adding the
following to your ~/.bashrc:
</para>
<para>
You may use Go packages installed into the active Nix profiles by adding
the following to your ~/.bashrc:
<screen> <screen>
for p in $NIX_PROFILES; do for p in $NIX_PROFILES; do
GOPATH="$p/share/go:$GOPATH" GOPATH="$p/share/go:$GOPATH"
done done
</screen> </screen>
</para> </para>
</section> </section>

59
doc/languages-frameworks/index.xml
View File

@ -1,36 +1,31 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xi="http://www.w3.org/2001/XInclude"
xml:id="chap-language-support"> xml:id="chap-language-support">
<title>Support for specific programming languages and frameworks</title>
<title>Support for specific programming languages and frameworks</title> <para>
The <link linkend="chap-stdenv">standard build environment</link> makes it
<para>The <link linkend="chap-stdenv">standard build easy to build typical Autotools-based packages with very little code. Any
environment</link> makes it easy to build typical Autotools-based other kind of package can be accomodated by overriding the appropriate phases
packages with very little code. Any other kind of package can be of <literal>stdenv</literal>. However, there are specialised functions in
accomodated by overriding the appropriate phases of Nixpkgs to easily build packages for other programming languages, such as
<literal>stdenv</literal>. However, there are specialised functions Perl or Haskell. These are described in this chapter.
in Nixpkgs to easily build packages for other programming languages, </para>
such as Perl or Haskell. These are described in this chapter.</para> <xi:include href="beam.xml" />
<xi:include href="bower.xml" />
<xi:include href="coq.xml" />
<xi:include href="beam.xml" /> <xi:include href="go.xml" />
<xi:include href="bower.xml" /> <xi:include href="haskell.section.xml" />
<xi:include href="coq.xml" /> <xi:include href="idris.section.xml" />
<xi:include href="go.xml" /> <xi:include href="java.xml" />
<xi:include href="haskell.section.xml" /> <xi:include href="lua.xml" />
<xi:include href="idris.section.xml" /> <xi:include href="node.section.xml" />
<xi:include href="java.xml" /> <xi:include href="perl.xml" />
<xi:include href="lua.xml" /> <xi:include href="python.section.xml" />
<xi:include href="node.section.xml" /> <xi:include href="qt.xml" />
<xi:include href="perl.xml" /> <xi:include href="r.section.xml" />
<xi:include href="python.section.xml" /> <xi:include href="ruby.xml" />
<xi:include href="qt.xml" /> <xi:include href="rust.section.xml" />
<xi:include href="r.section.xml" /> <xi:include href="texlive.xml" />
<xi:include href="ruby.xml" /> <xi:include href="vim.section.xml" />
<xi:include href="rust.section.xml" /> <xi:include href="emscripten.section.xml" />
<xi:include href="texlive.xml" />
<xi:include href="vim.section.xml" />
<xi:include href="emscripten.section.xml" />
</chapter> </chapter>

76
doc/languages-frameworks/java.xml
View File

@ -1,11 +1,10 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-java"> xml:id="sec-language-java">
<title>Java</title>
<title>Java</title> <para>
Ant-based Java packages are typically built from source as follows:
<para>Ant-based Java packages are typically built from source as follows:
<programlisting> <programlisting>
stdenv.mkDerivation { stdenv.mkDerivation {
name = "..."; name = "...";
@ -16,33 +15,33 @@ stdenv.mkDerivation {
buildPhase = "ant"; buildPhase = "ant";
} }
</programlisting> </programlisting>
Note that <varname>jdk</varname> is an alias for the OpenJDK.
</para>
Note that <varname>jdk</varname> is an alias for the OpenJDK.</para> <para>
JAR files that are intended to be used by other packages should be installed
<para>JAR files that are intended to be used by other packages should in <filename>$out/share/java</filename>. The OpenJDK has a stdenv setup hook
be installed in <filename>$out/share/java</filename>. The OpenJDK has that adds any JARs in the <filename>share/java</filename> directories of the
a stdenv setup hook that adds any JARs in the build inputs to the <envar>CLASSPATH</envar> environment variable. For
<filename>share/java</filename> directories of the build inputs to the instance, if the package <literal>libfoo</literal> installs a JAR named
<envar>CLASSPATH</envar> environment variable. For instance, if the <filename>foo.jar</filename> in its <filename>share/java</filename>
package <literal>libfoo</literal> installs a JAR named directory, and another package declares the attribute
<filename>foo.jar</filename> in its <filename>share/java</filename>
directory, and another package declares the attribute
<programlisting> <programlisting>
buildInputs = [ jdk libfoo ]; buildInputs = [ jdk libfoo ];
</programlisting> </programlisting>
then <envar>CLASSPATH</envar> will be set to
<filename>/nix/store/...-libfoo/share/java/foo.jar</filename>.
</para>
then <envar>CLASSPATH</envar> will be set to <para>
<filename>/nix/store/...-libfoo/share/java/foo.jar</filename>.</para> Private JARs should be installed in a location like
<filename>$out/share/<replaceable>package-name</replaceable></filename>.
<para>Private JARs </para>
should be installed in a location like
<filename>$out/share/<replaceable>package-name</replaceable></filename>.</para>
<para>If your Java package provides a program, you need to generate a
wrapper script to run it using the OpenJRE. You can use
<literal>makeWrapper</literal> for this:
<para>
If your Java package provides a program, you need to generate a wrapper
script to run it using the OpenJRE. You can use
<literal>makeWrapper</literal> for this:
<programlisting> <programlisting>
buildInputs = [ makeWrapper ]; buildInputs = [ makeWrapper ];
@ -53,23 +52,20 @@ installPhase =
--add-flags "-cp $out/share/java/foo.jar org.foo.Main" --add-flags "-cp $out/share/java/foo.jar org.foo.Main"
''; '';
</programlisting> </programlisting>
Note the use of <literal>jre</literal>, which is the part of the OpenJDK
package that contains the Java Runtime Environment. By using
<literal>${jre}/bin/java</literal> instead of
<literal>${jdk}/bin/java</literal>, you prevent your package from depending
on the JDK at runtime.
</para>
Note the use of <literal>jre</literal>, which is the part of the <para>
OpenJDK package that contains the Java Runtime Environment. By using It is possible to use a different Java compiler than <command>javac</command>
<literal>${jre}/bin/java</literal> instead of from the OpenJDK. For instance, to use the GNU Java Compiler:
<literal>${jdk}/bin/java</literal>, you prevent your package from
depending on the JDK at runtime.</para>
<para>It is possible to use a different Java compiler than
<command>javac</command> from the OpenJDK. For instance, to use the
GNU Java Compiler:
<programlisting> <programlisting>
buildInputs = [ gcj ant ]; buildInputs = [ gcj ant ];
</programlisting> </programlisting>
Here, Ant will automatically use <command>gij</command> (the GNU Java
Here, Ant will automatically use <command>gij</command> (the GNU Java Runtime) instead of the OpenJRE.
Runtime) instead of the OpenJRE.</para> </para>
</section> </section>

43
doc/languages-frameworks/lua.xml
View File

@ -1,24 +1,22 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-lua"> xml:id="sec-language-lua">
<title>Lua</title>
<title>Lua</title> <para>
Lua packages are built by the <varname>buildLuaPackage</varname> function.
<para> This function is implemented in
Lua packages are built by the <varname>buildLuaPackage</varname> function. This function is <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/lua-modules/generic/default.nix">
implemented
in <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/lua-modules/generic/default.nix">
<filename>pkgs/development/lua-modules/generic/default.nix</filename></link> <filename>pkgs/development/lua-modules/generic/default.nix</filename></link>
and works similarly to <varname>buildPerlPackage</varname>. (See and works similarly to <varname>buildPerlPackage</varname>. (See
<xref linkend="sec-language-perl"/> for details.) <xref linkend="sec-language-perl"/> for details.)
</para> </para>
<para> <para>
Lua packages are defined Lua packages are defined in
in <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/lua-packages.nix"><filename>pkgs/top-level/lua-packages.nix</filename></link>. <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/lua-packages.nix"><filename>pkgs/top-level/lua-packages.nix</filename></link>.
Most of them are simple. For example: Most of them are simple. For example:
<programlisting>
<programlisting>
fileSystem = buildLuaPackage { fileSystem = buildLuaPackage {
name = "filesystem-1.6.2"; name = "filesystem-1.6.2";
src = fetchurl { src = fetchurl {
@ -32,20 +30,19 @@ fileSystem = buildLuaPackage {
}; };
}; };
</programlisting> </programlisting>
</para> </para>
<para> <para>
Though, more complicated package should be placed in a seperate file in Though, more complicated package should be placed in a seperate file in
<link <link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/lua-modules"><filename>pkgs/development/lua-modules</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/lua-modules"><filename>pkgs/development/lua-modules</filename></link>.
</para> </para>
<para>
Lua packages accept additional parameter <varname>disabled</varname>, which defines
the condition of disabling package from luaPackages. For example, if package has
<varname>disabled</varname> assigned to <literal>lua.luaversion != "5.1"</literal>,
it will not be included in any luaPackages except lua51Packages, making it
only be built for lua 5.1.
</para>
<para>
Lua packages accept additional parameter <varname>disabled</varname>, which
defines the condition of disabling package from luaPackages. For example, if
package has <varname>disabled</varname> assigned to <literal>lua.luaversion
!= "5.1"</literal>, it will not be included in any luaPackages except
lua51Packages, making it only be built for lua 5.1.
</para>
</section> </section>

219
doc/languages-frameworks/perl.xml
View File

@ -1,24 +1,27 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-perl"> xml:id="sec-language-perl">
<title>Perl</title>
<title>Perl</title> <para>
Nixpkgs provides a function <varname>buildPerlPackage</varname>, a generic
package builder function for any Perl package that has a standard
<varname>Makefile.PL</varname>. It’s implemented in
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/perl-modules/generic"><filename>pkgs/development/perl-modules/generic</filename></link>.
</para>
<para>Nixpkgs provides a function <varname>buildPerlPackage</varname>, <para>
a generic package builder function for any Perl package that has a Perl packages from CPAN are defined in
standard <varname>Makefile.PL</varname>. It’s implemented in <link <link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/perl-modules/generic"><filename>pkgs/development/perl-modules/generic</filename></link>.</para>
<para>Perl packages from CPAN are defined in <link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/perl-packages.nix"><filename>pkgs/top-level/perl-packages.nix</filename></link>, xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/perl-packages.nix"><filename>pkgs/top-level/perl-packages.nix</filename></link>,
rather than <filename>pkgs/all-packages.nix</filename>. Most Perl rather than <filename>pkgs/all-packages.nix</filename>. Most Perl packages
packages are so straight-forward to build that they are defined here are so straight-forward to build that they are defined here directly, rather
directly, rather than having a separate function for each package than having a separate function for each package called from
called from <filename>perl-packages.nix</filename>. However, more <filename>perl-packages.nix</filename>. However, more complicated packages
complicated packages should be put in a separate file, typically in should be put in a separate file, typically in
<filename>pkgs/development/perl-modules</filename>. Here is an <filename>pkgs/development/perl-modules</filename>. Here is an example of the
example of the former: former:
<programlisting> <programlisting>
ClassC3 = buildPerlPackage rec { ClassC3 = buildPerlPackage rec {
name = "Class-C3-0.21"; name = "Class-C3-0.21";
@ -28,74 +31,72 @@ ClassC3 = buildPerlPackage rec {
}; };
}; };
</programlisting> </programlisting>
Note the use of <literal>mirror://cpan/</literal>, and the
Note the use of <literal>mirror://cpan/</literal>, and the <literal>${name}</literal> in the URL definition to ensure that the name
<literal>${name}</literal> in the URL definition to ensure that the attribute is consistent with the source that we’re actually downloading.
name attribute is consistent with the source that we’re actually Perl packages are made available in <filename>all-packages.nix</filename>
downloading. Perl packages are made available in through the variable <varname>perlPackages</varname>. For instance, if you
<filename>all-packages.nix</filename> through the variable have a package that needs <varname>ClassC3</varname>, you would typically
<varname>perlPackages</varname>. For instance, if you have a package write
that needs <varname>ClassC3</varname>, you would typically write
<programlisting> <programlisting>
foo = import ../path/to/foo.nix { foo = import ../path/to/foo.nix {
inherit stdenv fetchurl ...; inherit stdenv fetchurl ...;
inherit (perlPackages) ClassC3; inherit (perlPackages) ClassC3;
}; };
</programlisting> </programlisting>
in <filename>all-packages.nix</filename>. You can test building a Perl
in <filename>all-packages.nix</filename>. You can test building a package as follows:
Perl package as follows:
<screen> <screen>
$ nix-build -A perlPackages.ClassC3 $ nix-build -A perlPackages.ClassC3
</screen> </screen>
<varname>buildPerlPackage</varname> adds <literal>perl-</literal> to the
<varname>buildPerlPackage</varname> adds <literal>perl-</literal> to start of the name attribute, so the package above is actually called
the start of the name attribute, so the package above is actually <literal>perl-Class-C3-0.21</literal>. So to install it, you can say:
called <literal>perl-Class-C3-0.21</literal>. So to install it, you
can say:
<screen> <screen>
$ nix-env -i perl-Class-C3 $ nix-env -i perl-Class-C3
</screen> </screen>
(Of course you can also install using the attribute name: <literal>nix-env -i
-A perlPackages.ClassC3</literal>.)
</para>
(Of course you can also install using the attribute name: <para>
<literal>nix-env -i -A perlPackages.ClassC3</literal>.)</para> So what does <varname>buildPerlPackage</varname> do? It does the following:
<orderedlist>
<para>So what does <varname>buildPerlPackage</varname> do? It does <listitem>
the following: <para>
In the configure phase, it calls <literal>perl Makefile.PL</literal> to
<orderedlist> generate a Makefile. You can set the variable
<varname>makeMakerFlags</varname> to pass flags to
<listitem><para>In the configure phase, it calls <literal>perl <filename>Makefile.PL</filename>
Makefile.PL</literal> to generate a Makefile. You can set the </para>
variable <varname>makeMakerFlags</varname> to pass flags to </listitem>
<filename>Makefile.PL</filename></para></listitem> <listitem>
<para>
<listitem><para>It adds the contents of the <envar>PERL5LIB</envar> It adds the contents of the <envar>PERL5LIB</envar> environment variable
environment variable to <literal>#! .../bin/perl</literal> line of to <literal>#! .../bin/perl</literal> line of Perl scripts as
Perl scripts as <literal>-I<replaceable>dir</replaceable></literal> <literal>-I<replaceable>dir</replaceable></literal> flags. This ensures
flags. This ensures that a script can find its that a script can find its dependencies.
dependencies.</para></listitem> </para>
</listitem>
<listitem><para>In the fixup phase, it writes the propagated build <listitem>
inputs (<varname>propagatedBuildInputs</varname>) to the file <para>
<filename>$out/nix-support/propagated-user-env-packages</filename>. In the fixup phase, it writes the propagated build inputs
<command>nix-env</command> recursively installs all packages listed (<varname>propagatedBuildInputs</varname>) to the file
in this file when you install a package that has it. This ensures <filename>$out/nix-support/propagated-user-env-packages</filename>.
that a Perl package can find its dependencies.</para></listitem> <command>nix-env</command> recursively installs all packages listed in
this file when you install a package that has it. This ensures that a Perl
</orderedlist> package can find its dependencies.
</para>
</para> </listitem>
</orderedlist>
<para><varname>buildPerlPackage</varname> is built on top of </para>
<varname>stdenv</varname>, so everything can be customised in the
usual way. For instance, the <literal>BerkeleyDB</literal> module has
a <varname>preConfigure</varname> hook to generate a configuration
file used by <filename>Makefile.PL</filename>:
<para>
<varname>buildPerlPackage</varname> is built on top of
<varname>stdenv</varname>, so everything can be customised in the usual way.
For instance, the <literal>BerkeleyDB</literal> module has a
<varname>preConfigure</varname> hook to generate a configuration file used by
<filename>Makefile.PL</filename>:
<programlisting> <programlisting>
{ buildPerlPackage, fetchurl, db }: { buildPerlPackage, fetchurl, db }:
@ -113,18 +114,15 @@ buildPerlPackage rec {
''; '';
} }
</programlisting> </programlisting>
</para>
</para> <para>
Dependencies on other Perl packages can be specified in the
<para>Dependencies on other Perl packages can be specified in the <varname>buildInputs</varname> and <varname>propagatedBuildInputs</varname>
<varname>buildInputs</varname> and attributes. If something is exclusively a build-time dependency, use
<varname>propagatedBuildInputs</varname> attributes. If something is <varname>buildInputs</varname>; if it’s (also) a runtime dependency, use
exclusively a build-time dependency, use <varname>propagatedBuildInputs</varname>. For instance, this builds a Perl
<varname>buildInputs</varname>; if it’s (also) a runtime dependency, module that has runtime dependencies on a bunch of other modules:
use <varname>propagatedBuildInputs</varname>. For instance, this
builds a Perl module that has runtime dependencies on a bunch of other
modules:
<programlisting> <programlisting>
ClassC3Componentised = buildPerlPackage rec { ClassC3Componentised = buildPerlPackage rec {
name = "Class-C3-Componentised-1.0004"; name = "Class-C3-Componentised-1.0004";
@ -137,24 +135,26 @@ ClassC3Componentised = buildPerlPackage rec {
]; ];
}; };
</programlisting> </programlisting>
</para>
</para> <section xml:id="ssec-generation-from-CPAN">
<title>Generation from CPAN</title>
<section xml:id="ssec-generation-from-CPAN"><title>Generation from CPAN</title> <para>
Nix expressions for Perl packages can be generated (almost) automatically
<para>Nix expressions for Perl packages can be generated (almost) from CPAN. This is done by the program
automatically from CPAN. This is done by the program <command>nix-generate-from-cpan</command>, which can be installed as
<command>nix-generate-from-cpan</command>, which can be installed follows:
as follows:</para> </para>
<screen> <screen>
$ nix-env -i nix-generate-from-cpan $ nix-env -i nix-generate-from-cpan
</screen> </screen>
<para>This program takes a Perl module name, looks it up on CPAN, <para>
fetches and unpacks the corresponding package, and prints a Nix This program takes a Perl module name, looks it up on CPAN, fetches and
expression on standard output. For example: unpacks the corresponding package, and prints a Nix expression on standard
output. For example:
<screen> <screen>
$ nix-generate-from-cpan XML::Simple $ nix-generate-from-cpan XML::Simple
XMLSimple = buildPerlPackage rec { XMLSimple = buildPerlPackage rec {
@ -170,26 +170,23 @@ $ nix-generate-from-cpan XML::Simple
}; };
}; };
</screen> </screen>
The output can be pasted into
<filename>pkgs/top-level/perl-packages.nix</filename> or wherever else you
need it.
</para>
</section>
The output can be pasted into <section xml:id="ssec-perl-cross-compilation">
<filename>pkgs/top-level/perl-packages.nix</filename> or wherever else <title>Cross-compiling modules</title>
you need it.</para>
<para>
Nixpkgs has experimental support for cross-compiling Perl modules. In many
cases, it will just work out of the box, even for modules with native
extensions. Sometimes, however, the Makefile.PL for a module may
(indirectly) import a native module. In that case, you will need to make a
stub for that module that will satisfy the Makefile.PL and install it into
<filename>lib/perl5/site_perl/cross_perl/${perl.version}</filename>. See the
<varname>postInstall</varname> for <varname>DBI</varname> for an example.
</para>
</section>
</section> </section>
<section xml:id="ssec-perl-cross-compilation"><title>Cross-compiling modules</title>
<para>Nixpkgs has experimental support for cross-compiling Perl
modules. In many cases, it will just work out of the box, even for
modules with native extensions. Sometimes, however, the Makefile.PL
for a module may (indirectly) import a native module. In that case,
you will need to make a stub for that module that will satisfy the
Makefile.PL and install it into
<filename>lib/perl5/site_perl/cross_perl/${perl.version}</filename>.
See the <varname>postInstall</varname> for <varname>DBI</varname> for
an example.</para>
</section>
</section>

112
doc/languages-frameworks/qt.xml
View File

@ -1,58 +1,74 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-qt"> xml:id="sec-language-qt">
<title>Qt</title>
<title>Qt</title> <para>
Qt is a comprehensive desktop and mobile application development toolkit for
C++. Legacy support is available for Qt 3 and Qt 4, but all current
development uses Qt 5. The Qt 5 packages in Nixpkgs are updated frequently to
take advantage of new features, but older versions are typically retained
until their support window ends. The most important consideration in
packaging Qt-based software is ensuring that each package and all its
dependencies use the same version of Qt 5; this consideration motivates most
of the tools described below.
</para>
<para> <section xml:id="ssec-qt-libraries">
Qt is a comprehensive desktop and mobile application development toolkit for C++. <title>Packaging Libraries for Nixpkgs</title>
Legacy support is available for Qt 3 and Qt 4, but all current development uses Qt 5.
The Qt 5 packages in Nixpkgs are updated frequently to take advantage of new features,
but older versions are typically retained until their support window ends.
The most important consideration in packaging Qt-based software is ensuring that each package and all its dependencies use the same version of Qt 5;
this consideration motivates most of the tools described below.
</para>
<section xml:id="ssec-qt-libraries"><title>Packaging Libraries for Nixpkgs</title> <para>
Whenever possible, libraries that use Qt 5 should be built with each
available version. Packages providing libraries should be added to the
top-level function <varname>mkLibsForQt5</varname>, which is used to build a
set of libraries for every Qt 5 version. A special
<varname>callPackage</varname> function is used in this scope to ensure that
the entire dependency tree uses the same Qt 5 version. Import dependencies
unqualified, i.e., <literal>qtbase</literal> not
<literal>qt5.qtbase</literal>. <emphasis>Do not</emphasis> import a package
set such as <literal>qt5</literal> or <literal>libsForQt5</literal>.
</para>
<para> <para>
Whenever possible, libraries that use Qt 5 should be built with each available version. If a library does not support a particular version of Qt 5, it is best to
Packages providing libraries should be added to the top-level function <varname>mkLibsForQt5</varname>, mark it as broken by setting its <literal>meta.broken</literal> attribute. A
which is used to build a set of libraries for every Qt 5 version. package may be marked broken for certain versions by testing the
A special <varname>callPackage</varname> function is used in this scope to ensure that the entire dependency tree uses the same Qt 5 version. <literal>qtbase.version</literal> attribute, which will always give the
Import dependencies unqualified, i.e., <literal>qtbase</literal> not <literal>qt5.qtbase</literal>. current Qt 5 version.
<emphasis>Do not</emphasis> import a package set such as <literal>qt5</literal> or <literal>libsForQt5</literal>. </para>
</para> </section>
<para> <section xml:id="ssec-qt-applications">
If a library does not support a particular version of Qt 5, it is best to mark it as broken by setting its <literal>meta.broken</literal> attribute. <title>Packaging Applications for Nixpkgs</title>
A package may be marked broken for certain versions by testing the <literal>qtbase.version</literal> attribute, which will always give the current Qt 5 version.
</para>
<para>
Call your application expression using
<literal>libsForQt5.callPackage</literal> instead of
<literal>callPackage</literal>. Import dependencies unqualified, i.e.,
<literal>qtbase</literal> not <literal>qt5.qtbase</literal>. <emphasis>Do
not</emphasis> import a package set such as <literal>qt5</literal> or
<literal>libsForQt5</literal>.
</para>
<para>
Qt 5 maintains strict backward compatibility, so it is generally best to
build an application package against the latest version using the
<varname>libsForQt5</varname> library set. In case a package does not build
with the latest Qt version, it is possible to pick a set pinned to a
particular version, e.g. <varname>libsForQt55</varname> for Qt 5.5, if that
is the latest version the package supports. If a package must be pinned to
an older Qt version, be sure to file a bug upstream; because Qt is strictly
backwards-compatible, any incompatibility is by definition a bug in the
application.
</para>
<para>
When testing applications in Nixpkgs, it is a common practice to build the
package with <literal>nix-build</literal> and run it using the created
symbolic link. This will not work with Qt applications, however, because
they have many hard runtime requirements that can only be guaranteed if the
package is actually installed. To test a Qt application, install it with
<literal>nix-env</literal> or run it inside <literal>nix-shell</literal>.
</para>
</section>
</section> </section>
<section xml:id="ssec-qt-applications"><title>Packaging Applications for Nixpkgs</title>
<para>
Call your application expression using <literal>libsForQt5.callPackage</literal> instead of <literal>callPackage</literal>.
Import dependencies unqualified, i.e., <literal>qtbase</literal> not <literal>qt5.qtbase</literal>.
<emphasis>Do not</emphasis> import a package set such as <literal>qt5</literal> or <literal>libsForQt5</literal>.
</para>
<para>
Qt 5 maintains strict backward compatibility, so it is generally best to build an application package against the latest version using the <varname>libsForQt5</varname> library set.
In case a package does not build with the latest Qt version, it is possible to pick a set pinned to a particular version, e.g. <varname>libsForQt55</varname> for Qt 5.5, if that is the latest version the package supports.
If a package must be pinned to an older Qt version, be sure to file a bug upstream;
because Qt is strictly backwards-compatible, any incompatibility is by definition a bug in the application.
</para>
<para>
When testing applications in Nixpkgs, it is a common practice to build the package with <literal>nix-build</literal> and run it using the created symbolic link.
This will not work with Qt applications, however, because they have many hard runtime requirements that can only be guaranteed if the package is actually installed.
To test a Qt application, install it with <literal>nix-env</literal> or run it inside <literal>nix-shell</literal>.
</para>
</section>
</section>

90
doc/languages-frameworks/ruby.xml
View File

@ -1,17 +1,19 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-ruby"> xml:id="sec-language-ruby">
<title>Ruby</title>
<title>Ruby</title> <para>
There currently is support to bundle applications that are packaged as Ruby
gems. The utility "bundix" allows you to write a
<filename>Gemfile</filename>, let bundler create a
<filename>Gemfile.lock</filename>, and then convert this into a nix
expression that contains all Gem dependencies automatically.
</para>
<para>There currently is support to bundle applications that are packaged as <para>
Ruby gems. The utility "bundix" allows you to write a For example, to package sensu, we did:
<filename>Gemfile</filename>, let bundler create a </para>
<filename>Gemfile.lock</filename>, and then convert this into a nix
expression that contains all Gem dependencies automatically.
</para>
<para>For example, to package sensu, we did:</para>
<screen> <screen>
<![CDATA[$ cd pkgs/servers/monitoring <![CDATA[$ cd pkgs/servers/monitoring
@ -42,17 +44,18 @@ bundlerEnv rec {
}]]> }]]>
</screen> </screen>
<para>Please check in the <filename>Gemfile</filename>, <para>
<filename>Gemfile.lock</filename> and the Please check in the <filename>Gemfile</filename>,
<filename>gemset.nix</filename> so future updates can be run easily. <filename>Gemfile.lock</filename> and the <filename>gemset.nix</filename> so
</para> future updates can be run easily.
</para>
<para>For tools written in Ruby - i.e. where the desire is to install <para>
a package and then execute e.g. <command>rake</command> at the command For tools written in Ruby - i.e. where the desire is to install a package and
line, there is an alternative builder called <literal>bundlerApp</literal>. then execute e.g. <command>rake</command> at the command line, there is an
Set up the <filename>gemset.nix</filename> the same way, and then, for alternative builder called <literal>bundlerApp</literal>. Set up the
example: <filename>gemset.nix</filename> the same way, and then, for example:
</para> </para>
<screen> <screen>
<![CDATA[{ lib, bundlerApp }: <![CDATA[{ lib, bundlerApp }:
@ -72,31 +75,31 @@ bundlerApp {
}]]> }]]>
</screen> </screen>
<para>The chief advantage of <literal>bundlerApp</literal> over <para>
<literal>bundlerEnv</literal> is the executables introduced in the The chief advantage of <literal>bundlerApp</literal> over
environment are precisely those selected in the <literal>exes</literal> <literal>bundlerEnv</literal> is the executables introduced in the
list, as opposed to <literal>bundlerEnv</literal> which adds all the environment are precisely those selected in the <literal>exes</literal> list,
executables made available by gems in the gemset, which can mean e.g. as opposed to <literal>bundlerEnv</literal> which adds all the executables
<command>rspec</command> or <command>rake</command> in unpredictable made available by gems in the gemset, which can mean e.g.
versions available from various packages. <command>rspec</command> or <command>rake</command> in unpredictable versions
</para> available from various packages.
</para>
<para>Resulting derivations for both builders also have two helpful <para>
attributes, <literal>env</literal> and <literal>wrappedRuby</literal>. Resulting derivations for both builders also have two helpful attributes,
The first one allows one to quickly drop into <literal>env</literal> and <literal>wrappedRuby</literal>. The first one
<command>nix-shell</command> with the specified environment present. allows one to quickly drop into <command>nix-shell</command> with the
E.g. <command>nix-shell -A sensu.env</command> would give you an specified environment present. E.g. <command>nix-shell -A sensu.env</command>
environment with Ruby preset so it has all the libraries necessary would give you an environment with Ruby preset so it has all the libraries
for <literal>sensu</literal> in its paths. The second one can be necessary for <literal>sensu</literal> in its paths. The second one can be
used to make derivations from custom Ruby scripts which have used to make derivations from custom Ruby scripts which have
<filename>Gemfile</filename>s with their dependencies specified. It is <filename>Gemfile</filename>s with their dependencies specified. It is a
a derivation with <command>ruby</command> wrapped so it can find all derivation with <command>ruby</command> wrapped so it can find all the needed
the needed dependencies. For example, to make a derivation dependencies. For example, to make a derivation <literal>my-script</literal>
<literal>my-script</literal> for a <filename>my-script.rb</filename> for a <filename>my-script.rb</filename> (which should be placed in
(which should be placed in <filename>bin</filename>) you should run <filename>bin</filename>) you should run <command>bundix</command> as
<command>bundix</command> as specified above and then use specified above and then use <literal>bundlerEnv</literal> like this:
<literal>bundlerEnv</literal> like this: </para>
</para>
<programlisting> <programlisting>
<![CDATA[let env = bundlerEnv { <![CDATA[let env = bundlerEnv {
@ -118,5 +121,4 @@ in stdenv.mkDerivation {
''; '';
}]]> }]]>
</programlisting> </programlisting>
</section> </section>

108
doc/languages-frameworks/texlive.xml
View File

@ -1,27 +1,42 @@
<section xmlns="http://docbook.org/ns/docbook" <section xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="sec-language-texlive"> xml:id="sec-language-texlive">
<title>TeX Live</title>
<title>TeX Live</title> <para>
Since release 15.09 there is a new TeX Live packaging that lives entirely
under attribute <varname>texlive</varname>.
</para>
<section>
<title>User's guide</title>
<para>Since release 15.09 there is a new TeX Live packaging that lives entirely under attribute <varname>texlive</varname>.</para>
<section><title>User's guide</title>
<itemizedlist> <itemizedlist>
<listitem><para> <listitem>
For basic usage just pull <varname>texlive.combined.scheme-basic</varname> for an environment with basic LaTeX support.</para></listitem> <para>
<listitem><para> For basic usage just pull <varname>texlive.combined.scheme-basic</varname>
It typically won't work to use separately installed packages together. for an environment with basic LaTeX support.
Instead, you can build a custom set of packages like this: </para>
<programlisting> </listitem>
<listitem>
<para>
It typically won't work to use separately installed packages together.
Instead, you can build a custom set of packages like this:
<programlisting>
texlive.combine { texlive.combine {
inherit (texlive) scheme-small collection-langkorean algorithms cm-super; inherit (texlive) scheme-small collection-langkorean algorithms cm-super;
} }
</programlisting> </programlisting>
There are all the schemes, collections and a few thousand packages, as defined upstream (perhaps with tiny differences). There are all the schemes, collections and a few thousand packages, as
</para></listitem> defined upstream (perhaps with tiny differences).
<listitem><para> </para>
By default you only get executables and files needed during runtime, and a little documentation for the core packages. To change that, you need to add <varname>pkgFilter</varname> function to <varname>combine</varname>. </listitem>
<programlisting> <listitem>
<para>
By default you only get executables and files needed during runtime, and a
little documentation for the core packages. To change that, you need to
add <varname>pkgFilter</varname> function to <varname>combine</varname>.
<programlisting>
texlive.combine { texlive.combine {
# inherit (texlive) whatever-you-want; # inherit (texlive) whatever-you-want;
pkgFilter = pkg: pkgFilter = pkg:
@ -30,34 +45,55 @@ texlive.combine {
# there are also other attributes: version, name # there are also other attributes: version, name
} }
</programlisting> </programlisting>
</para></listitem> </para>
<listitem><para> </listitem>
You can list packages e.g. by <command>nix-repl</command>. <listitem>
<programlisting> <para>
You can list packages e.g. by <command>nix-repl</command>.
<programlisting>
$ nix-repl $ nix-repl
nix-repl> :l &lt;nixpkgs> nix-repl> :l &lt;nixpkgs>
nix-repl> texlive.collection-&lt;TAB> nix-repl> texlive.collection-&lt;TAB>
</programlisting> </programlisting>
</para></listitem> </para>
<listitem><para> </listitem>
Note that the wrapper assumes that the result has a chance to be useful. For example, the core executables should be present, as well as some core data files. The supported way of ensuring this is by including some scheme, for example <varname>scheme-basic</varname>, into the combination. <listitem>
</para></listitem> <para>
Note that the wrapper assumes that the result has a chance to be useful.
For example, the core executables should be present, as well as some core
data files. The supported way of ensuring this is by including some
scheme, for example <varname>scheme-basic</varname>, into the combination.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</section> </section>
<section>
<title>Known problems</title>
<section><title>Known problems</title>
<itemizedlist> <itemizedlist>
<listitem><para> <listitem>
Some tools are still missing, e.g. luajittex;</para></listitem> <para>
<listitem><para> Some tools are still missing, e.g. luajittex;
some apps aren't packaged/tested yet (asymptote, biber, etc.);</para></listitem> </para>
<listitem><para> </listitem>
feature/bug: when a package is rejected by <varname>pkgFilter</varname>, its dependencies are still propagated;</para></listitem> <listitem>
<listitem><para> <para>
in case of any bugs or feature requests, file a github issue or better a pull request and /cc @vcunat.</para></listitem> some apps aren't packaged/tested yet (asymptote, biber, etc.);
</para>
</listitem>
<listitem>
<para>
feature/bug: when a package is rejected by <varname>pkgFilter</varname>,
its dependencies are still propagated;
</para>
</listitem>
<listitem>
<para>
in case of any bugs or feature requests, file a github issue or better a
pull request and /cc @vcunat.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</section>
</section> </section>
</section>

47
doc/manual.xml
View File

@ -1,29 +1,24 @@
<book xmlns="http://docbook.org/ns/docbook" <book xmlns="http://docbook.org/ns/docbook"
xmlns:xi="http://www.w3.org/2001/XInclude"> xmlns:xi="http://www.w3.org/2001/XInclude">
<info>
<info> <title>Nixpkgs Contributors Guide</title>
<subtitle>Version <xi:include href=".version" parse="text" />
<title>Nixpkgs Contributors Guide</title> </subtitle>
</info>
<subtitle>Version <xi:include href=".version" parse="text" /></subtitle> <xi:include href="introduction.chapter.xml" />
<xi:include href="quick-start.xml" />
</info> <xi:include href="stdenv.xml" />
<xi:include href="multiple-output.xml" />
<xi:include href="introduction.chapter.xml" /> <xi:include href="cross-compilation.xml" />
<xi:include href="quick-start.xml" /> <xi:include href="configuration.xml" />
<xi:include href="stdenv.xml" /> <xi:include href="functions.xml" />
<xi:include href="multiple-output.xml" /> <xi:include href="meta.xml" />
<xi:include href="cross-compilation.xml" /> <xi:include href="languages-frameworks/index.xml" />
<xi:include href="configuration.xml" /> <xi:include href="platform-notes.xml" />
<xi:include href="functions.xml" /> <xi:include href="package-notes.xml" />
<xi:include href="meta.xml" /> <xi:include href="overlays.xml" />
<xi:include href="languages-frameworks/index.xml" /> <xi:include href="coding-conventions.xml" />
<xi:include href="platform-notes.xml" /> <xi:include href="submitting-changes.xml" />
<xi:include href="package-notes.xml" /> <xi:include href="reviewing-contributions.xml" />
<xi:include href="overlays.xml" /> <xi:include href="contributing.xml" />
<xi:include href="coding-conventions.xml" />
<xi:include href="submitting-changes.xml" />
<xi:include href="reviewing-contributions.xml" />
<xi:include href="contributing.xml" />
</book> </book>

540
doc/meta.xml
View File

@ -1,14 +1,12 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-meta"> xml:id="chap-meta">
<title>Meta-attributes</title>
<title>Meta-attributes</title> <para>
Nix packages can declare <emphasis>meta-attributes</emphasis> that contain
<para>Nix packages can declare <emphasis>meta-attributes</emphasis> information about a package such as a description, its homepage, its license,
that contain information about a package such as a description, its and so on. For instance, the GNU Hello package has a <varname>meta</varname>
homepage, its license, and so on. For instance, the GNU Hello package declaration like this:
has a <varname>meta</varname> declaration like this:
<programlisting> <programlisting>
meta = { meta = {
description = "A program that produces a familiar, friendly greeting"; description = "A program that produces a familiar, friendly greeting";
@ -22,16 +20,15 @@ meta = {
platforms = stdenv.lib.platforms.all; platforms = stdenv.lib.platforms.all;
}; };
</programlisting> </programlisting>
</para>
</para> <para>
Meta-attributes are not passed to the builder of the package. Thus, a change
<para>Meta-attributes are not passed to the builder of the package. to a meta-attribute doesn’t trigger a recompilation of the package. The
Thus, a change to a meta-attribute doesn’t trigger a recompilation of value of a meta-attribute must be a string.
the package. The value of a meta-attribute must be a string.</para> </para>
<para>
<para>The meta-attributes of a package can be queried from the The meta-attributes of a package can be queried from the command-line using
command-line using <command>nix-env</command>: <command>nix-env</command>:
<screen> <screen>
$ nix-env -qa hello --json $ nix-env -qa hello --json
{ {
@ -70,252 +67,299 @@ $ nix-env -qa hello --json
</screen> </screen>
<command>nix-env</command> knows about the <varname>description</varname>
<command>nix-env</command> knows about the field specifically:
<varname>description</varname> field specifically:
<screen> <screen>
$ nix-env -qa hello --description $ nix-env -qa hello --description
hello-2.3 A program that produces a familiar, friendly greeting hello-2.3 A program that produces a familiar, friendly greeting
</screen> </screen>
</para>
<section xml:id="sec-standard-meta-attributes">
<title>Standard meta-attributes</title>
</para> <para>
It is expected that each meta-attribute is one of the following:
</para>
<variablelist>
<section xml:id="sec-standard-meta-attributes"><title>Standard <varlistentry>
meta-attributes</title> <term><varname>description</varname>
</term>
<para>It is expected that each meta-attribute is one of the following:</para>
<variablelist>
<varlistentry>
<term><varname>description</varname></term>
<listitem><para>A short (one-line) description of the package.
This is shown by <command>nix-env -q --description</command> and
also on the Nixpkgs release pages.</para>
<para>Don’t include a period at the end. Don’t include newline
characters. Capitalise the first character. For brevity, don’t
repeat the name of package — just describe what it does.</para>
<para>Wrong: <literal>"libpng is a library that allows you to decode PNG images."</literal></para>
<para>Right: <literal>"A library for decoding PNG images"</literal></para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>longDescription</varname></term>
<listitem><para>An arbitrarily long description of the
package.</para></listitem>
</varlistentry>
<varlistentry>
<term><varname>branch</varname></term>
<listitem><para>Release branch. Used to specify that a package is not
going to receive updates that are not in this branch; for example, Linux
kernel 3.0 is supposed to be updated to 3.0.X, not 3.1.</para></listitem>
</varlistentry>
<varlistentry>
<term><varname>homepage</varname></term>
<listitem><para>The package’s homepage. Example:
<literal>http://www.gnu.org/software/hello/manual/</literal></para></listitem>
</varlistentry>
<varlistentry>
<term><varname>downloadPage</varname></term>
<listitem><para>The page where a link to the current version can be found. Example:
<literal>http://ftp.gnu.org/gnu/hello/</literal></para></listitem>
</varlistentry>
<varlistentry>
<term><varname>license</varname></term>
<listitem> <listitem>
<para> <para>
The license, or licenses, for the package. One from the attribute set A short (one-line) description of the package. This is shown by
defined in <link <command>nix-env -q --description</command> and also on the Nixpkgs
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/lib/licenses.nix"> release pages.
<filename>nixpkgs/lib/licenses.nix</filename></link>. At this moment </para>
using both a list of licenses and a single license is valid. If the <para>
license field is in the form of a list representation, then it means Don’t include a period at the end. Don’t include newline characters.
that parts of the package are licensed differently. Each license Capitalise the first character. For brevity, don’t repeat the name of
should preferably be referenced by their attribute. The non-list package — just describe what it does.
attribute value can also be a space delimited string representation of </para>
the contained attribute shortNames or spdxIds. The following are all valid <para>
examples: Wrong: <literal>"libpng is a library that allows you to decode PNG
<itemizedlist> images."</literal>
<listitem><para>Single license referenced by attribute (preferred) </para>
<literal>stdenv.lib.licenses.gpl3</literal>. <para>
</para></listitem> Right: <literal>"A library for decoding PNG images"</literal>
<listitem><para>Single license referenced by its attribute shortName (frowned upon) </para>
<literal>"gpl3"</literal>.
</para></listitem>
<listitem><para>Single license referenced by its attribute spdxId (frowned upon)
<literal>"GPL-3.0"</literal>.
</para></listitem>
<listitem><para>Multiple licenses referenced by attribute (preferred)
<literal>with stdenv.lib.licenses; [ asl20 free ofl ]</literal>.
</para></listitem>
<listitem><para>Multiple licenses referenced as a space delimited string of attribute shortNames (frowned upon)
<literal>"asl20 free ofl"</literal>.
</para></listitem>
</itemizedlist>
For details, see <xref linkend='sec-meta-license'/>.
</para>
</listitem> </listitem>
</varlistentry> </varlistentry>
<varlistentry>
<varlistentry> <term><varname>longDescription</varname>
<term><varname>maintainers</varname></term> </term>
<listitem><para>A list of names and e-mail addresses of the <listitem>
maintainers of this Nix expression. If <para>
you would like to be a maintainer of a package, you may want to add An arbitrarily long description of the package.
yourself to <link </para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>branch</varname>
</term>
<listitem>
<para>
Release branch. Used to specify that a package is not going to receive
updates that are not in this branch; for example, Linux kernel 3.0 is
supposed to be updated to 3.0.X, not 3.1.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>homepage</varname>
</term>
<listitem>
<para>
The package’s homepage. Example:
<literal>http://www.gnu.org/software/hello/manual/</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>downloadPage</varname>
</term>
<listitem>
<para>
The page where a link to the current version can be found. Example:
<literal>http://ftp.gnu.org/gnu/hello/</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>license</varname>
</term>
<listitem>
<para>
The license, or licenses, for the package. One from the attribute set
defined in
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/lib/licenses.nix">
<filename>nixpkgs/lib/licenses.nix</filename></link>. At this moment
using both a list of licenses and a single license is valid. If the
license field is in the form of a list representation, then it means that
parts of the package are licensed differently. Each license should
preferably be referenced by their attribute. The non-list attribute value
can also be a space delimited string representation of the contained
attribute shortNames or spdxIds. The following are all valid examples:
<itemizedlist>
<listitem>
<para>
Single license referenced by attribute (preferred)
<literal>stdenv.lib.licenses.gpl3</literal>.
</para>
</listitem>
<listitem>
<para>
Single license referenced by its attribute shortName (frowned upon)
<literal>"gpl3"</literal>.
</para>
</listitem>
<listitem>
<para>
Single license referenced by its attribute spdxId (frowned upon)
<literal>"GPL-3.0"</literal>.
</para>
</listitem>
<listitem>
<para>
Multiple licenses referenced by attribute (preferred) <literal>with
stdenv.lib.licenses; [ asl20 free ofl ]</literal>.
</para>
</listitem>
<listitem>
<para>
Multiple licenses referenced as a space delimited string of attribute
shortNames (frowned upon) <literal>"asl20 free ofl"</literal>.
</para>
</listitem>
</itemizedlist>
For details, see <xref linkend='sec-meta-license'/>.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>maintainers</varname>
</term>
<listitem>
<para>
A list of names and e-mail addresses of the maintainers of this Nix
expression. If you would like to be a maintainer of a package, you may
want to add yourself to
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/maintainers/maintainer-list.nix"><filename>nixpkgs/maintainers/maintainer-list.nix</filename></link> xlink:href="https://github.com/NixOS/nixpkgs/blob/master/maintainers/maintainer-list.nix"><filename>nixpkgs/maintainers/maintainer-list.nix</filename></link>
and write something like <literal>[ stdenv.lib.maintainers.alice and write something like <literal>[ stdenv.lib.maintainers.alice
stdenv.lib.maintainers.bob ]</literal>.</para></listitem> stdenv.lib.maintainers.bob ]</literal>.
</varlistentry> </para>
</listitem>
<varlistentry> </varlistentry>
<term><varname>priority</varname></term> <varlistentry>
<listitem><para>The <emphasis>priority</emphasis> of the package, <term><varname>priority</varname>
used by <command>nix-env</command> to resolve file name conflicts </term>
between packages. See the Nix manual page for <listitem>
<command>nix-env</command> for details. Example: <para>
<literal>"10"</literal> (a low-priority The <emphasis>priority</emphasis> of the package, used by
package).</para></listitem> <command>nix-env</command> to resolve file name conflicts between
</varlistentry> packages. See the Nix manual page for <command>nix-env</command> for
details. Example: <literal>"10"</literal> (a low-priority package).
<varlistentry> </para>
<term><varname>platforms</varname></term> </listitem>
<listitem><para>The list of Nix platform types on which the </varlistentry>
package is supported. Hydra builds packages according to the <varlistentry>
platform specified. If no platform is specified, the package does <term><varname>platforms</varname>
not have prebuilt binaries. An example is: </term>
<listitem>
<para>
The list of Nix platform types on which the package is supported. Hydra
builds packages according to the platform specified. If no platform is
specified, the package does not have prebuilt binaries. An example is:
<programlisting> <programlisting>
meta.platforms = stdenv.lib.platforms.linux; meta.platforms = stdenv.lib.platforms.linux;
</programlisting> </programlisting>
Attribute Set <varname>stdenv.lib.platforms</varname> defines
Attribute Set <varname>stdenv.lib.platforms</varname> defines <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/lib/systems/doubles.nix">
<link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/lib/systems/doubles.nix"> various common lists</link> of platforms types.
various common lists</link> of platforms types.</para></listitem> </para>
</varlistentry> </listitem>
</varlistentry>
<varlistentry> <varlistentry>
<term><varname>hydraPlatforms</varname></term> <term><varname>hydraPlatforms</varname>
<listitem><para>The list of Nix platform types for which the Hydra </term>
instance at <literal>hydra.nixos.org</literal> will build the <listitem>
package. (Hydra is the Nix-based continuous build system.) It <para>
defaults to the value of <varname>meta.platforms</varname>. Thus, The list of Nix platform types for which the Hydra instance at
the only reason to set <varname>meta.hydraPlatforms</varname> is <literal>hydra.nixos.org</literal> will build the package. (Hydra is the
if you want <literal>hydra.nixos.org</literal> to build the Nix-based continuous build system.) It defaults to the value of
package on a subset of <varname>meta.platforms</varname>, or not <varname>meta.platforms</varname>. Thus, the only reason to set
at all, e.g. <varname>meta.hydraPlatforms</varname> is if you want
<literal>hydra.nixos.org</literal> to build the package on a subset of
<varname>meta.platforms</varname>, or not at all, e.g.
<programlisting> <programlisting>
meta.platforms = stdenv.lib.platforms.linux; meta.platforms = stdenv.lib.platforms.linux;
meta.hydraPlatforms = []; meta.hydraPlatforms = [];
</programlisting> </programlisting>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>broken</varname>
</term>
<listitem>
<para>
If set to <literal>true</literal>, the package is marked as “broken”,
meaning that it won’t show up in <literal>nix-env -qa</literal>, and
cannot be built or installed. Such packages should be removed from
Nixpkgs eventually unless they are fixed.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>updateWalker</varname>
</term>
<listitem>
<para>
If set to <literal>true</literal>, the package is tested to be updated
correctly by the <literal>update-walker.sh</literal> script without
additional settings. Such packages have <varname>meta.version</varname>
set and their homepage (or the page specified by
<varname>meta.downloadPage</varname>) contains a direct link to the
package tarball.
</para>
</listitem>
</varlistentry>
</variablelist>
</section>
<section xml:id="sec-meta-license">
<title>Licenses</title>
</para></listitem> <para>
</varlistentry> The <varname>meta.license</varname> attribute should preferrably contain a
value from <varname>stdenv.lib.licenses</varname> defined in
<varlistentry> <link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/lib/licenses.nix">
<term><varname>broken</varname></term> <filename>nixpkgs/lib/licenses.nix</filename></link>, or in-place license
<listitem><para>If set to <literal>true</literal>, the package is description of the same format if the license is unlikely to be useful in
marked as “broken”, meaning that it won’t show up in another expression.
<literal>nix-env -qa</literal>, and cannot be built or installed. </para>
Such packages should be removed from Nixpkgs eventually unless
they are fixed.</para></listitem>
</varlistentry>
<varlistentry>
<term><varname>updateWalker</varname></term>
<listitem><para>If set to <literal>true</literal>, the package is
tested to be updated correctly by the <literal>update-walker.sh</literal>
script without additional settings. Such packages have
<varname>meta.version</varname> set and their homepage (or
the page specified by <varname>meta.downloadPage</varname>) contains
a direct link to the package tarball.</para></listitem>
</varlistentry>
</variablelist>
</section>
<section xml:id="sec-meta-license"><title>Licenses</title>
<para>The <varname>meta.license</varname> attribute should preferrably contain
a value from <varname>stdenv.lib.licenses</varname> defined in
<link xlink:href="https://github.com/NixOS/nixpkgs/blob/master/lib/licenses.nix">
<filename>nixpkgs/lib/licenses.nix</filename></link>,
or in-place license description of the same format if the license is
unlikely to be useful in another expression.</para>
<para>Although it's typically better to indicate the specific license,
a few generic options are available:
<variablelist>
<varlistentry>
<term><varname>stdenv.lib.licenses.free</varname>,
<varname>"free"</varname></term>
<listitem><para>Catch-all for free software licenses not listed
above.</para></listitem>
</varlistentry>
<varlistentry>
<term><varname>stdenv.lib.licenses.unfreeRedistributable</varname>,
<varname>"unfree-redistributable"</varname></term>
<listitem><para>Unfree package that can be redistributed in binary
form. That is, it’s legal to redistribute the
<emphasis>output</emphasis> of the derivation. This means that
the package can be included in the Nixpkgs
channel.</para>
<para>Sometimes proprietary software can only be redistributed
unmodified. Make sure the builder doesn’t actually modify the
original binaries; otherwise we’re breaking the license. For
instance, the NVIDIA X11 drivers can be redistributed unmodified,
but our builder applies <command>patchelf</command> to make them
work. Thus, its license is <varname>"unfree"</varname> and it
cannot be included in the Nixpkgs channel.</para></listitem>
</varlistentry>
<varlistentry>
<term><varname>stdenv.lib.licenses.unfree</varname>,
<varname>"unfree"</varname></term>
<listitem><para>Unfree package that cannot be redistributed. You
can build it yourself, but you cannot redistribute the output of
the derivation. Thus it cannot be included in the Nixpkgs
channel.</para></listitem>
</varlistentry>
<varlistentry>
<term><varname>stdenv.lib.licenses.unfreeRedistributableFirmware</varname>,
<varname>"unfree-redistributable-firmware"</varname></term>
<listitem><para>This package supplies unfree, redistributable
firmware. This is a separate value from
<varname>unfree-redistributable</varname> because not everybody
cares whether firmware is free.</para></listitem>
</varlistentry>
</variablelist>
</para>
</section>
<para>
Although it's typically better to indicate the specific license, a few
generic options are available:
<variablelist>
<varlistentry>
<term><varname>stdenv.lib.licenses.free</varname>,
<varname>"free"</varname>
</term>
<listitem>
<para>
Catch-all for free software licenses not listed above.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>stdenv.lib.licenses.unfreeRedistributable</varname>,
<varname>"unfree-redistributable"</varname>
</term>
<listitem>
<para>
Unfree package that can be redistributed in binary form. That is, it’s
legal to redistribute the <emphasis>output</emphasis> of the derivation.
This means that the package can be included in the Nixpkgs channel.
</para>
<para>
Sometimes proprietary software can only be redistributed unmodified.
Make sure the builder doesn’t actually modify the original binaries;
otherwise we’re breaking the license. For instance, the NVIDIA X11
drivers can be redistributed unmodified, but our builder applies
<command>patchelf</command> to make them work. Thus, its license is
<varname>"unfree"</varname> and it cannot be included in the Nixpkgs
channel.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>stdenv.lib.licenses.unfree</varname>,
<varname>"unfree"</varname>
</term>
<listitem>
<para>
Unfree package that cannot be redistributed. You can build it yourself,
but you cannot redistribute the output of the derivation. Thus it cannot
be included in the Nixpkgs channel.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>stdenv.lib.licenses.unfreeRedistributableFirmware</varname>,
<varname>"unfree-redistributable-firmware"</varname>
</term>
<listitem>
<para>
This package supplies unfree, redistributable firmware. This is a
separate value from <varname>unfree-redistributable</varname> because
not everybody cares whether firmware is free.
</para>
</listitem>
</varlistentry>
</variablelist>
</para>
</section>
</chapter> </chapter>

380
doc/multiple-output.xml
View File

@ -5,105 +5,319 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-multiple-output"> xml:id="chap-multiple-output">
<title>Multiple-output packages</title>
<section>
<title>Introduction</title>
<title>Multiple-output packages</title> <para>
The Nix language allows a derivation to produce multiple outputs, which is
similar to what is utilized by other Linux distribution packaging systems.
The outputs reside in separate nix store paths, so they can be mostly
handled independently of each other, including passing to build inputs,
garbage collection or binary substitution. The exception is that building
from source always produces all the outputs.
</para>
<section><title>Introduction</title> <para>
<para>The Nix language allows a derivation to produce multiple outputs, which is similar to what is utilized by other Linux distribution packaging systems. The outputs reside in separate nix store paths, so they can be mostly handled independently of each other, including passing to build inputs, garbage collection or binary substitution. The exception is that building from source always produces all the outputs.</para> The main motivation is to save disk space by reducing runtime closure sizes;
<para>The main motivation is to save disk space by reducing runtime closure sizes; consequently also sizes of substituted binaries get reduced. Splitting can be used to have more granular runtime dependencies, for example the typical reduction is to split away development-only files, as those are typically not needed during runtime. As a result, closure sizes of many packages can get reduced to a half or even much less.</para> consequently also sizes of substituted binaries get reduced. Splitting can
<note><para>The reduction effects could be instead achieved by building the parts in completely separate derivations. That would often additionally reduce build-time closures, but it tends to be much harder to write such derivations, as build systems typically assume all parts are being built at once. This compromise approach of single source package producing multiple binary packages is also utilized often by rpm and deb.</para></note> be used to have more granular runtime dependencies, for example the typical
</section> reduction is to split away development-only files, as those are typically
not needed during runtime. As a result, closure sizes of many packages can
get reduced to a half or even much less.
</para>
<note>
<para>
The reduction effects could be instead achieved by building the parts in
completely separate derivations. That would often additionally reduce
build-time closures, but it tends to be much harder to write such
derivations, as build systems typically assume all parts are being built at
once. This compromise approach of single source package producing multiple
binary packages is also utilized often by rpm and deb.
</para>
</note>
</section>
<section>
<title>Installing a split package</title>
<para>
When installing a package via <varname>systemPackages</varname> or
<command>nix-env</command> you have several options:
</para>
<section><title>Installing a split package</title>
<para>When installing a package via <varname>systemPackages</varname> or <command>nix-env</command> you have several options:</para>
<itemizedlist> <itemizedlist>
<listitem><para>You can install particular outputs explicitly, as each is available in the Nix language as an attribute of the package. The <varname>outputs</varname> attribute contains a list of output names.</para></listitem> <listitem>
<listitem><para>You can let it use the default outputs. These are handled by <varname>meta.outputsToInstall</varname> attribute that contains a list of output names.</para> <para>
<para>TODO: more about tweaking the attribute, etc.</para></listitem> You can install particular outputs explicitly, as each is available in the
<listitem><para>NixOS provides configuration option <varname>environment.extraOutputsToInstall</varname> that allows adding extra outputs of <varname>environment.systemPackages</varname> atop the default ones. It's mainly meant for documentation and debug symbols, and it's also modified by specific options.</para> Nix language as an attribute of the package. The
<note><para>At this moment there is no similar configurability for packages installed by <command>nix-env</command>. You can still use approach from <xref linkend="sec-modify-via-packageOverrides" /> to override <varname>meta.outputsToInstall</varname> attributes, but that's a rather inconvenient way.</para></note> <varname>outputs</varname> attribute contains a list of output names.
</listitem> </para>
</listitem>
<listitem>
<para>
You can let it use the default outputs. These are handled by
<varname>meta.outputsToInstall</varname> attribute that contains a list of
output names.
</para>
<para>
TODO: more about tweaking the attribute, etc.
</para>
</listitem>
<listitem>
<para>
NixOS provides configuration option
<varname>environment.extraOutputsToInstall</varname> that allows adding
extra outputs of <varname>environment.systemPackages</varname> atop the
default ones. It's mainly meant for documentation and debug symbols, and
it's also modified by specific options.
</para>
<note>
<para>
At this moment there is no similar configurability for packages installed
by <command>nix-env</command>. You can still use approach from
<xref linkend="sec-modify-via-packageOverrides" /> to override
<varname>meta.outputsToInstall</varname> attributes, but that's a rather
inconvenient way.
</para>
</note>
</listitem>
</itemizedlist> </itemizedlist>
</section> </section>
<section>
<title>Using a split package</title>
<section><title>Using a split package</title> <para>
<para>In the Nix language the individual outputs can be reached explicitly as attributes, e.g. <varname>coreutils.info</varname>, but the typical case is just using packages as build inputs.</para> In the Nix language the individual outputs can be reached explicitly as
<para>When a multiple-output derivation gets into a build input of another derivation, the <varname>dev</varname> output is added if it exists, otherwise the first output is added. In addition to that, <varname>propagatedBuildOutputs</varname> of that package which by default contain <varname>$outputBin</varname> and <varname>$outputLib</varname> are also added. (See <xref linkend="multiple-output-file-type-groups" />.)</para> attributes, e.g. <varname>coreutils.info</varname>, but the typical case is
</section> just using packages as build inputs.
</para>
<para>
When a multiple-output derivation gets into a build input of another
derivation, the <varname>dev</varname> output is added if it exists,
otherwise the first output is added. In addition to that,
<varname>propagatedBuildOutputs</varname> of that package which by default
contain <varname>$outputBin</varname> and <varname>$outputLib</varname> are
also added. (See <xref linkend="multiple-output-file-type-groups" />.)
</para>
</section>
<section>
<title>Writing a split derivation</title>
<section><title>Writing a split derivation</title> <para>
<para>Here you find how to write a derivation that produces multiple outputs.</para> Here you find how to write a derivation that produces multiple outputs.
<para>In nixpkgs there is a framework supporting multiple-output derivations. It tries to cover most cases by default behavior. You can find the source separated in &lt;<filename>nixpkgs/pkgs/build-support/setup-hooks/multiple-outputs.sh</filename>&gt;; it's relatively well-readable. The whole machinery is triggered by defining the <varname>outputs</varname> attribute to contain the list of desired output names (strings).</para> </para>
<programlisting>outputs = [ "bin" "dev" "out" "doc" ];</programlisting>
<para>Often such a single line is enough. For each output an equally named environment variable is passed to the builder and contains the path in nix store for that output. Typically you also want to have the main <varname>out</varname> output, as it catches any files that didn't get elsewhere.</para> <para>
<note><para>There is a special handling of the <varname>debug</varname> output, described at <xref linkend="stdenv-separateDebugInfo" />.</para></note> In nixpkgs there is a framework supporting multiple-output derivations. It
tries to cover most cases by default behavior. You can find the source
separated in
&lt;<filename>nixpkgs/pkgs/build-support/setup-hooks/multiple-outputs.sh</filename>&gt;;
it's relatively well-readable. The whole machinery is triggered by defining
the <varname>outputs</varname> attribute to contain the list of desired
output names (strings).
</para>
<programlisting>outputs = [ "bin" "dev" "out" "doc" ];</programlisting>
<para>
Often such a single line is enough. For each output an equally named
environment variable is passed to the builder and contains the path in nix
store for that output. Typically you also want to have the main
<varname>out</varname> output, as it catches any files that didn't get
elsewhere.
</para>
<note>
<para>
There is a special handling of the <varname>debug</varname> output,
described at <xref linkend="stdenv-separateDebugInfo" />.
</para>
</note>
<section xml:id="multiple-output-file-binaries-first-convention"> <section xml:id="multiple-output-file-binaries-first-convention">
<title><quote>Binaries first</quote></title> <title><quote>Binaries first</quote></title>
<para>A commonly adopted convention in <literal>nixpkgs</literal> is that executables provided by the package are contained within its first output. This convention allows the dependent packages to reference the executables provided by packages in a uniform manner. For instance, provided with the knowledge that the <literal>perl</literal> package contains a <literal>perl</literal> executable it can be referenced as <literal>${pkgs.perl}/bin/perl</literal> within a Nix derivation that needs to execute a Perl script.</para>
<para>The <literal>glibc</literal> package is a deliberate single exception to the <quote>binaries first</quote> convention. The <literal>glibc</literal> has <literal>libs</literal> as its first output allowing the libraries provided by <literal>glibc</literal> to be referenced directly (e.g. <literal>${stdenv.glibc}/lib/ld-linux-x86-64.so.2</literal>). The executables provided by <literal>glibc</literal> can be accessed via its <literal>bin</literal> attribute (e.g. <literal>${stdenv.glibc.bin}/bin/ldd</literal>).</para> <para>
<para>The reason for why <literal>glibc</literal> deviates from the convention is because referencing a library provided by <literal>glibc</literal> is a very common operation among Nix packages. For instance, third-party executables packaged by Nix are typically patched and relinked with the relevant version of <literal>glibc</literal> libraries from Nix packages (please see the documentation on <link xlink:href="https://nixos.org/patchelf.html">patchelf</link> for more details).</para> A commonly adopted convention in <literal>nixpkgs</literal> is that
executables provided by the package are contained within its first output.
This convention allows the dependent packages to reference the executables
provided by packages in a uniform manner. For instance, provided with the
knowledge that the <literal>perl</literal> package contains a
<literal>perl</literal> executable it can be referenced as
<literal>${pkgs.perl}/bin/perl</literal> within a Nix derivation that needs
to execute a Perl script.
</para>
<para>
The <literal>glibc</literal> package is a deliberate single exception to
the <quote>binaries first</quote> convention. The <literal>glibc</literal>
has <literal>libs</literal> as its first output allowing the libraries
provided by <literal>glibc</literal> to be referenced directly (e.g.
<literal>${stdenv.glibc}/lib/ld-linux-x86-64.so.2</literal>). The
executables provided by <literal>glibc</literal> can be accessed via its
<literal>bin</literal> attribute (e.g.
<literal>${stdenv.glibc.bin}/bin/ldd</literal>).
</para>
<para>
The reason for why <literal>glibc</literal> deviates from the convention is
because referencing a library provided by <literal>glibc</literal> is a
very common operation among Nix packages. For instance, third-party
executables packaged by Nix are typically patched and relinked with the
relevant version of <literal>glibc</literal> libraries from Nix packages
(please see the documentation on
<link xlink:href="https://nixos.org/patchelf.html">patchelf</link> for more
details).
</para>
</section> </section>
<section xml:id="multiple-output-file-type-groups"> <section xml:id="multiple-output-file-type-groups">
<title>File type groups</title> <title>File type groups</title>
<para>The support code currently recognizes some particular kinds of outputs and either instructs the build system of the package to put files into their desired outputs or it moves the files during the fixup phase. Each group of file types has an <varname>outputFoo</varname> variable specifying the output name where they should go. If that variable isn't defined by the derivation writer, it is guessed &ndash; a default output name is defined, falling back to other possibilities if the output isn't defined.</para>
<variablelist>
<varlistentry><term><varname> <para>
$outputDev</varname></term><listitem><para> The support code currently recognizes some particular kinds of outputs and
is for development-only files. These include C(++) headers, pkg-config, cmake and aclocal files. They go to <varname>dev</varname> or <varname>out</varname> by default. either instructs the build system of the package to put files into their
</para></listitem> desired outputs or it moves the files during the fixup phase. Each group of
</varlistentry> file types has an <varname>outputFoo</varname> variable specifying the
output name where they should go. If that variable isn't defined by the
derivation writer, it is guessed &ndash; a default output name is defined,
falling back to other possibilities if the output isn't defined.
</para>
<varlistentry><term><varname> <variablelist>
$outputBin</varname></term><listitem><para> <varlistentry>
is meant for user-facing binaries, typically residing in bin/. They go to <varname>bin</varname> or <varname>out</varname> by default. <term><varname>
</para></listitem></varlistentry> $outputDev</varname>
</term>
<varlistentry><term><varname> <listitem>
$outputLib</varname></term><listitem><para> <para>
is meant for libraries, typically residing in <filename>lib/</filename> and <filename>libexec/</filename>. They go to <varname>lib</varname> or <varname>out</varname> by default. is for development-only files. These include C(++) headers, pkg-config,
</para></listitem></varlistentry> cmake and aclocal files. They go to <varname>dev</varname> or
<varname>out</varname> by default.
<varlistentry><term><varname> </para>
$outputDoc</varname></term><listitem><para> </listitem>
is for user documentation, typically residing in <filename>share/doc/</filename>. It goes to <varname>doc</varname> or <varname>out</varname> by default. </varlistentry>
</para></listitem></varlistentry> <varlistentry>
<term><varname>
<varlistentry><term><varname> $outputBin</varname>
$outputDevdoc</varname></term><listitem><para> </term>
is for <emphasis>developer</emphasis> documentation. Currently we count gtk-doc and devhelp books in there. It goes to <varname>devdoc</varname> or is removed (!) by default. This is because e.g. gtk-doc tends to be rather large and completely unused by nixpkgs users. <listitem>
</para></listitem></varlistentry> <para>
is meant for user-facing binaries, typically residing in bin/. They go
<varlistentry><term><varname> to <varname>bin</varname> or <varname>out</varname> by default.
$outputMan</varname></term><listitem><para> </para>
is for man pages (except for section 3). They go to <varname>man</varname> or <varname>$outputBin</varname> by default. </listitem>
</para></listitem></varlistentry> </varlistentry>
<varlistentry>
<varlistentry><term><varname> <term><varname>
$outputDevman</varname></term><listitem><para> $outputLib</varname>
is for section 3 man pages. They go to <varname>devman</varname> or <varname>$outputMan</varname> by default. </term>
</para></listitem></varlistentry> <listitem>
<para>
<varlistentry><term><varname> is meant for libraries, typically residing in <filename>lib/</filename>
$outputInfo</varname></term><listitem><para> and <filename>libexec/</filename>. They go to <varname>lib</varname> or
is for info pages. They go to <varname>info</varname> or <varname>$outputBin</varname> by default. <varname>out</varname> by default.
</para></listitem></varlistentry> </para>
</listitem>
</variablelist> </varlistentry>
<varlistentry>
<term><varname>
$outputDoc</varname>
</term>
<listitem>
<para>
is for user documentation, typically residing in
<filename>share/doc/</filename>. It goes to <varname>doc</varname> or
<varname>out</varname> by default.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>
$outputDevdoc</varname>
</term>
<listitem>
<para>
is for <emphasis>developer</emphasis> documentation. Currently we count
gtk-doc and devhelp books in there. It goes to <varname>devdoc</varname>
or is removed (!) by default. This is because e.g. gtk-doc tends to be
rather large and completely unused by nixpkgs users.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>
$outputMan</varname>
</term>
<listitem>
<para>
is for man pages (except for section 3). They go to
<varname>man</varname> or <varname>$outputBin</varname> by default.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>
$outputDevman</varname>
</term>
<listitem>
<para>
is for section 3 man pages. They go to <varname>devman</varname> or
<varname>$outputMan</varname> by default.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><varname>
$outputInfo</varname>
</term>
<listitem>
<para>
is for info pages. They go to <varname>info</varname> or
<varname>$outputBin</varname> by default.
</para>
</listitem>
</varlistentry>
</variablelist>
</section> </section>
<section><title>Common caveats</title> <section>
<itemizedlist> <title>Common caveats</title>
<listitem><para>Some configure scripts don't like some of the parameters passed by default by the framework, e.g. <literal>--docdir=/foo/bar</literal>. You can disable this by setting <literal>setOutputFlags = false;</literal>.</para></listitem>
<listitem><para>The outputs of a single derivation can retain references to each other, but note that circular references are not allowed. (And each strongly-connected component would act as a single output anyway.)</para></listitem> <itemizedlist>
<listitem><para>Most of split packages contain their core functionality in libraries. These libraries tend to refer to various kind of data that typically gets into <varname>out</varname>, e.g. locale strings, so there is often no advantage in separating the libraries into <varname>lib</varname>, as keeping them in <varname>out</varname> is easier.</para></listitem> <listitem>
<listitem><para>Some packages have hidden assumptions on install paths, which complicates splitting.</para></listitem> <para>
</itemizedlist> Some configure scripts don't like some of the parameters passed by
default by the framework, e.g. <literal>--docdir=/foo/bar</literal>. You
can disable this by setting <literal>setOutputFlags = false;</literal>.
</para>
</listitem>
<listitem>
<para>
The outputs of a single derivation can retain references to each other,
but note that circular references are not allowed. (And each
strongly-connected component would act as a single output anyway.)
</para>
</listitem>
<listitem>
<para>
Most of split packages contain their core functionality in libraries.
These libraries tend to refer to various kind of data that typically gets
into <varname>out</varname>, e.g. locale strings, so there is often no
advantage in separating the libraries into <varname>lib</varname>, as
keeping them in <varname>out</varname> is easier.
</para>
</listitem>
<listitem>
<para>
Some packages have hidden assumptions on install paths, which complicates
splitting.
</para>
</listitem>
</itemizedlist>
</section> </section>
</section>
</section><!--Writing a split derivation--> <!--Writing a split derivation-->
</chapter> </chapter>

240
doc/overlays.xml
View File

@ -1,95 +1,117 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-overlays"> xml:id="chap-overlays">
<title>Overlays</title>
<title>Overlays</title> <para>
This chapter describes how to extend and change Nixpkgs packages using
<para>This chapter describes how to extend and change Nixpkgs packages using overlays. Overlays are used to add layers in the fix-point used by Nixpkgs to
overlays. Overlays are used to add layers in the fix-point used by Nixpkgs compose the set of all packages.
to compose the set of all packages.</para> </para>
<para>
<para>Nixpkgs can be configured with a list of overlays, which are Nixpkgs can be configured with a list of overlays, which are applied in
applied in order. This means that the order of the overlays can be significant order. This means that the order of the overlays can be significant if
if multiple layers override the same package.</para> multiple layers override the same package.
</para>
<!--============================================================--> <!--============================================================-->
<section xml:id="sec-overlays-install">
<title>Installing overlays</title>
<section xml:id="sec-overlays-install"> <para>
<title>Installing overlays</title> The list of overlays is determined as follows.
</para>
<para>The list of overlays is determined as follows.</para> <para>
If the <varname>overlays</varname> argument is not provided explicitly, we
look for overlays in a path. The path is determined as follows:
<orderedlist>
<listitem>
<para>
First, if an <varname>overlays</varname> argument to the nixpkgs function
itself is given, then that is used.
</para>
<para>
This can be passed explicitly when importing nipxkgs, for example
<literal>import &lt;nixpkgs> { overlays = [ overlay1 overlay2 ];
}</literal>.
</para>
</listitem>
<listitem>
<para>
Otherwise, if the Nix path entry <literal>&lt;nixpkgs-overlays></literal>
exists, we look for overlays at that path, as described below.
</para>
<para>
See the section on <literal>NIX_PATH</literal> in the Nix manual for more
details on how to set a value for
<literal>&lt;nixpkgs-overlays>.</literal>
</para>
</listitem>
<listitem>
<para>
If one of <filename>~/.config/nixpkgs/overlays.nix</filename> and
<filename>~/.config/nixpkgs/overlays/</filename> exists, then we look for
overlays at that path, as described below. It is an error if both exist.
</para>
</listitem>
</orderedlist>
</para>
<para>If the <varname>overlays</varname> argument is not provided explicitly, we look for overlays in a path. The path <para>
is determined as follows: If we are looking for overlays at a path, then there are two cases:
<itemizedlist>
<listitem>
<para>
If the path is a file, then the file is imported as a Nix expression and
used as the list of overlays.
</para>
</listitem>
<listitem>
<para>
If the path is a directory, then we take the content of the directory,
order it lexicographically, and attempt to interpret each as an overlay
by:
<itemizedlist>
<listitem>
<para>
Importing the file, if it is a <literal>.nix</literal> file.
</para>
</listitem>
<listitem>
<para>
Importing a top-level <filename>default.nix</filename> file, if it is
a directory.
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</para>
<orderedlist> <para>
On a NixOS system the value of the <literal>nixpkgs.overlays</literal>
<listitem> option, if present, is passed to the system Nixpkgs directly as an argument.
<para>First, if an <varname>overlays</varname> argument to the nixpkgs function itself is given, Note that this does not affect the overlays for non-NixOS operations (e.g.
then that is used.</para> <literal>nix-env</literal>), which are looked up independently.
</para>
<para>This can be passed explicitly when importing nipxkgs, for example
<literal>import &lt;nixpkgs> { overlays = [ overlay1 overlay2 ]; }</literal>.</para>
</listitem>
<listitem>
<para>Otherwise, if the Nix path entry <literal>&lt;nixpkgs-overlays></literal> exists, we look for overlays
at that path, as described below.</para>
<para>See the section on <literal>NIX_PATH</literal> in the Nix manual for more details on how to
set a value for <literal>&lt;nixpkgs-overlays>.</literal></para>
</listitem>
<listitem>
<para>If one of <filename>~/.config/nixpkgs/overlays.nix</filename> and
<filename>~/.config/nixpkgs/overlays/</filename> exists, then we look for overlays at that path, as
described below. It is an error if both exist.</para>
</listitem>
</orderedlist>
</para>
<para>If we are looking for overlays at a path, then there are two cases:
<itemizedlist>
<listitem>
<para>If the path is a file, then the file is imported as a Nix expression and used as the list of
overlays.</para>
</listitem>
<listitem>
<para>If the path is a directory, then we take the content of the directory, order it
lexicographically, and attempt to interpret each as an overlay by:
<itemizedlist>
<listitem>
<para>Importing the file, if it is a <literal>.nix</literal> file.</para>
</listitem>
<listitem>
<para>Importing a top-level <filename>default.nix</filename> file, if it is a directory.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</para>
<para>On a NixOS system the value of the <literal>nixpkgs.overlays</literal> option, if present,
is passed to the system Nixpkgs directly as an argument. Note that this does not affect the overlays for
non-NixOS operations (e.g. <literal>nix-env</literal>), which are looked up independently.</para>
<para>The <filename>overlays.nix</filename> option therefore provides a convenient way to use the same
overlays for a NixOS system configuration and user configuration: the same file can be used
as <filename>overlays.nix</filename> and imported as the value of <literal>nixpkgs.overlays</literal>.</para>
</section>
<para>
The <filename>overlays.nix</filename> option therefore provides a convenient
way to use the same overlays for a NixOS system configuration and user
configuration: the same file can be used as
<filename>overlays.nix</filename> and imported as the value of
<literal>nixpkgs.overlays</literal>.
</para>
</section>
<!--============================================================--> <!--============================================================-->
<section xml:id="sec-overlays-definition">
<title>Defining overlays</title>
<section xml:id="sec-overlays-definition"> <para>
<title>Defining overlays</title> Overlays are Nix functions which accept two arguments, conventionally called
<varname>self</varname> and <varname>super</varname>, and return a set of
<para>Overlays are Nix functions which accept two arguments, packages. For example, the following is a valid overlay.
conventionally called <varname>self</varname> and <varname>super</varname>, </para>
and return a set of packages. For example, the following is a valid overlay.</para>
<programlisting> <programlisting>
self: super: self: super:
@ -104,31 +126,39 @@ self: super:
} }
</programlisting> </programlisting>
<para>The first argument (<varname>self</varname>) corresponds to the final package <para>
set. You should use this set for the dependencies of all packages specified in your The first argument (<varname>self</varname>) corresponds to the final
overlay. For example, all the dependencies of <varname>rr</varname> in the example above come package set. You should use this set for the dependencies of all packages
from <varname>self</varname>, as well as the overridden dependencies used in the specified in your overlay. For example, all the dependencies of
<varname>boost</varname> override.</para> <varname>rr</varname> in the example above come from
<varname>self</varname>, as well as the overridden dependencies used in the
<varname>boost</varname> override.
</para>
<para>The second argument (<varname>super</varname>) <para>
corresponds to the result of the evaluation of the previous stages of The second argument (<varname>super</varname>) corresponds to the result of
Nixpkgs. It does not contain any of the packages added by the current the evaluation of the previous stages of Nixpkgs. It does not contain any of
overlay, nor any of the following overlays. This set should be used either the packages added by the current overlay, nor any of the following
to refer to packages you wish to override, or to access functions defined overlays. This set should be used either to refer to packages you wish to
in Nixpkgs. For example, the original recipe of <varname>boost</varname> override, or to access functions defined in Nixpkgs. For example, the
in the above example, comes from <varname>super</varname>, as well as the original recipe of <varname>boost</varname> in the above example, comes from
<varname>callPackage</varname> function.</para> <varname>super</varname>, as well as the <varname>callPackage</varname>
function.
</para>
<para>The value returned by this function should be a set similar to <para>
<filename>pkgs/top-level/all-packages.nix</filename>, containing The value returned by this function should be a set similar to
overridden and/or new packages.</para> <filename>pkgs/top-level/all-packages.nix</filename>, containing overridden
and/or new packages.
<para>Overlays are similar to other methods for customizing Nixpkgs, in particular </para>
the <literal>packageOverrides</literal> attribute described in <xref linkend="sec-modify-via-packageOverrides"/>.
Indeed, <literal>packageOverrides</literal> acts as an overlay with only the
<varname>super</varname> argument. It is therefore appropriate for basic use,
but overlays are more powerful and easier to distribute.</para>
</section>
<para>
Overlays are similar to other methods for customizing Nixpkgs, in particular
the <literal>packageOverrides</literal> attribute described in
<xref linkend="sec-modify-via-packageOverrides"/>. Indeed,
<literal>packageOverrides</literal> acts as an overlay with only the
<varname>super</varname> argument. It is therefore appropriate for basic
use, but overlays are more powerful and easier to distribute.
</para>
</section>
</chapter> </chapter>

901
doc/package-notes.xml
View File

File diff suppressed because it is too large Load Diff

84
doc/platform-notes.xml
View File

@ -1,27 +1,25 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-platform-nodes"> xml:id="chap-platform-nodes">
<title>Platform Notes</title>
<section xml:id="sec-darwin">
<title>Darwin (macOS)</title>
<title>Platform Notes</title> <para>
Some common issues when packaging software for darwin:
</para>
<section xml:id="sec-darwin"> <itemizedlist>
<listitem>
<title>Darwin (macOS)</title>
<para>Some common issues when packaging software for darwin:</para>
<itemizedlist>
<listitem>
<para> <para>
The darwin <literal>stdenv</literal> uses clang instead of gcc. The darwin <literal>stdenv</literal> uses clang instead of gcc. When
When referring to the compiler <varname>$CC</varname> or <command>cc</command> referring to the compiler <varname>$CC</varname> or <command>cc</command>
will work in both cases. Some builds hardcode gcc/g++ in their will work in both cases. Some builds hardcode gcc/g++ in their build
build scripts, that can usually be fixed with using something scripts, that can usually be fixed with using something like
like <literal>makeFlags = [ "CC=cc" ];</literal> or by patching <literal>makeFlags = [ "CC=cc" ];</literal> or by patching the build
the build scripts. scripts.
</para> </para>
<programlisting>
<programlisting>
stdenv.mkDerivation { stdenv.mkDerivation {
name = "libfoo-1.2.3"; name = "libfoo-1.2.3";
# ... # ...
@ -30,36 +28,33 @@
''; '';
} }
</programlisting> </programlisting>
</listitem> </listitem>
<listitem>
<listitem>
<para> <para>
On darwin libraries are linked using absolute paths, libraries On darwin libraries are linked using absolute paths, libraries are
are resolved by their <literal>install_name</literal> at link resolved by their <literal>install_name</literal> at link time. Sometimes
time. Sometimes packages won't set this correctly causing the packages won't set this correctly causing the library lookups to fail at
library lookups to fail at runtime. This can be fixed by adding runtime. This can be fixed by adding extra linker flags or by running
extra linker flags or by running <command>install_name_tool -id</command> <command>install_name_tool -id</command> during the
during the <function>fixupPhase</function>. <function>fixupPhase</function>.
</para> </para>
<programlisting>
<programlisting>
stdenv.mkDerivation { stdenv.mkDerivation {
name = "libfoo-1.2.3"; name = "libfoo-1.2.3";
# ... # ...
makeFlags = stdenv.lib.optional stdenv.isDarwin "LDFLAGS=-Wl,-install_name,$(out)/lib/libfoo.dylib"; makeFlags = stdenv.lib.optional stdenv.isDarwin "LDFLAGS=-Wl,-install_name,$(out)/lib/libfoo.dylib";
} }
</programlisting> </programlisting>
</listitem> </listitem>
<listitem>
<listitem>
<para> <para>
Some packages assume xcode is available and use <command>xcrun</command> Some packages assume xcode is available and use <command>xcrun</command>
to resolve build tools like <command>clang</command>, etc. to resolve build tools like <command>clang</command>, etc. This causes
This causes errors like <code>xcode-select: error: no developer tools were found at '/Applications/Xcode.app'</code> errors like <code>xcode-select: error: no developer tools were found at
while the build doesn't actually depend on xcode. '/Applications/Xcode.app'</code> while the build doesn't actually depend
on xcode.
</para> </para>
<programlisting>
<programlisting>
stdenv.mkDerivation { stdenv.mkDerivation {
name = "libfoo-1.2.3"; name = "libfoo-1.2.3";
# ... # ...
@ -69,15 +64,12 @@
''; '';
} }
</programlisting> </programlisting>
<para> <para>
The package <literal>xcbuild</literal> can be used to build projects The package <literal>xcbuild</literal> can be used to build projects that
that really depend on Xcode, however projects that build some kind of really depend on Xcode, however projects that build some kind of graphical
graphical interface won't work without using Xcode in an impure way. interface won't work without using Xcode in an impure way.
</para> </para>
</listitem> </listitem>
</itemizedlist>
</itemizedlist> </section>
</section>
</chapter> </chapter>

374
doc/quick-start.xml
View File

@ -1,223 +1,219 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-quick-start"> xml:id="chap-quick-start">
<title>Quick Start to Adding a Package</title>
<title>Quick Start to Adding a Package</title> <para>
To add a package to Nixpkgs:
<para>To add a package to Nixpkgs: <orderedlist>
<listitem>
<orderedlist> <para>
Checkout the Nixpkgs source tree:
<listitem>
<para>Checkout the Nixpkgs source tree:
<screen> <screen>
$ git clone git://github.com/NixOS/nixpkgs.git $ git clone git://github.com/NixOS/nixpkgs.git
$ cd nixpkgs</screen> $ cd nixpkgs</screen>
</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Find a good place in the Nixpkgs tree to add the Nix Find a good place in the Nixpkgs tree to add the Nix expression for your
expression for your package. For instance, a library package package. For instance, a library package typically goes into
typically goes into <filename>pkgs/development/libraries/<replaceable>pkgname</replaceable></filename>,
<filename>pkgs/development/libraries/<replaceable>pkgname</replaceable></filename>, while a web browser goes into
while a web browser goes into <filename>pkgs/applications/networking/browsers/<replaceable>pkgname</replaceable></filename>.
<filename>pkgs/applications/networking/browsers/<replaceable>pkgname</replaceable></filename>. See <xref linkend="sec-organisation" /> for some hints on the tree
See <xref linkend="sec-organisation" /> for some hints on the tree organisation. Create a directory for your package, e.g.
organisation. Create a directory for your package, e.g.
<screen> <screen>
$ mkdir pkgs/development/libraries/libfoo</screen> $ mkdir pkgs/development/libraries/libfoo</screen>
</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>In the package directory, create a Nix expression — a piece In the package directory, create a Nix expression — a piece of code that
of code that describes how to build the package. In this case, it describes how to build the package. In this case, it should be a
should be a <emphasis>function</emphasis> that is called with the <emphasis>function</emphasis> that is called with the package dependencies
package dependencies as arguments, and returns a build of the as arguments, and returns a build of the package in the Nix store. The
package in the Nix store. The expression should usually be called expression should usually be called <filename>default.nix</filename>.
<filename>default.nix</filename>.
<screen> <screen>
$ emacs pkgs/development/libraries/libfoo/default.nix $ emacs pkgs/development/libraries/libfoo/default.nix
$ git add pkgs/development/libraries/libfoo/default.nix</screen> $ git add pkgs/development/libraries/libfoo/default.nix</screen>
</para> </para>
<para>
<para>You can have a look at the existing Nix expressions under You can have a look at the existing Nix expressions under
<filename>pkgs/</filename> to see how it’s done. Here are some <filename>pkgs/</filename> to see how it’s done. Here are some good
good ones: ones:
<itemizedlist>
<itemizedlist> <listitem>
<para>
<listitem> GNU Hello:
<para>GNU Hello: <link <link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/hello/default.nix"><filename>pkgs/applications/misc/hello/default.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/hello/default.nix"><filename>pkgs/applications/misc/hello/default.nix</filename></link>.
Trivial package, which specifies some <varname>meta</varname> Trivial package, which specifies some <varname>meta</varname>
attributes which is good practice.</para> attributes which is good practice.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>GNU cpio: <link <para>
GNU cpio:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/tools/archivers/cpio/default.nix"><filename>pkgs/tools/archivers/cpio/default.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/tools/archivers/cpio/default.nix"><filename>pkgs/tools/archivers/cpio/default.nix</filename></link>.
Also a simple package. The generic builder in Also a simple package. The generic builder in <varname>stdenv</varname>
<varname>stdenv</varname> does everything for you. It has does everything for you. It has no dependencies beyond
no dependencies beyond <varname>stdenv</varname>.</para> <varname>stdenv</varname>.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>GNU Multiple Precision arithmetic library (GMP): <link <para>
GNU Multiple Precision arithmetic library (GMP):
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/libraries/gmp/5.1.x.nix"><filename>pkgs/development/libraries/gmp/5.1.x.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/libraries/gmp/5.1.x.nix"><filename>pkgs/development/libraries/gmp/5.1.x.nix</filename></link>.
Also done by the generic builder, but has a dependency on Also done by the generic builder, but has a dependency on
<varname>m4</varname>.</para> <varname>m4</varname>.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>Pan, a GTK-based newsreader: <link <para>
Pan, a GTK-based newsreader:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/networking/newsreaders/pan/default.nix"><filename>pkgs/applications/networking/newsreaders/pan/default.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/networking/newsreaders/pan/default.nix"><filename>pkgs/applications/networking/newsreaders/pan/default.nix</filename></link>.
Has an optional dependency on <varname>gtkspell</varname>, Has an optional dependency on <varname>gtkspell</varname>, which is
which is only built if <varname>spellCheck</varname> is only built if <varname>spellCheck</varname> is <literal>true</literal>.
<literal>true</literal>.</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Apache HTTPD: <link Apache HTTPD:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/servers/http/apache-httpd/2.4.nix"><filename>pkgs/servers/http/apache-httpd/2.4.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/servers/http/apache-httpd/2.4.nix"><filename>pkgs/servers/http/apache-httpd/2.4.nix</filename></link>.
A bunch of optional features, variable substitutions in the A bunch of optional features, variable substitutions in the configure
configure flags, a post-install hook, and miscellaneous flags, a post-install hook, and miscellaneous hackery.
hackery.</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Thunderbird: <link Thunderbird:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/networking/mailreaders/thunderbird/default.nix"><filename>pkgs/applications/networking/mailreaders/thunderbird/default.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/networking/mailreaders/thunderbird/default.nix"><filename>pkgs/applications/networking/mailreaders/thunderbird/default.nix</filename></link>.
Lots of dependencies.</para> Lots of dependencies.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>JDiskReport, a Java utility: <link <para>
JDiskReport, a Java utility:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/tools/misc/jdiskreport/default.nix"><filename>pkgs/tools/misc/jdiskreport/default.nix</filename></link> xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/tools/misc/jdiskreport/default.nix"><filename>pkgs/tools/misc/jdiskreport/default.nix</filename></link>
(and the <link (and the
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/tools/misc/jdiskreport/builder.sh">builder</link>). xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/tools/misc/jdiskreport/builder.sh">builder</link>).
Nixpkgs doesn’t have a decent <varname>stdenv</varname> for Nixpkgs doesn’t have a decent <varname>stdenv</varname> for Java yet
Java yet so this is pretty ad-hoc.</para> so this is pretty ad-hoc.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>XML::Simple, a Perl module: <link <para>
XML::Simple, a Perl module:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/perl-packages.nix"><filename>pkgs/top-level/perl-packages.nix</filename></link> xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/perl-packages.nix"><filename>pkgs/top-level/perl-packages.nix</filename></link>
(search for the <varname>XMLSimple</varname> attribute). (search for the <varname>XMLSimple</varname> attribute). Most Perl
Most Perl modules are so simple to build that they are modules are so simple to build that they are defined directly in
defined directly in <filename>perl-packages.nix</filename>; <filename>perl-packages.nix</filename>; no need to make a separate file
no need to make a separate file for them.</para> for them.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>Adobe Reader: <link <para>
Adobe Reader:
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/adobe-reader/default.nix"><filename>pkgs/applications/misc/adobe-reader/default.nix</filename></link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/adobe-reader/default.nix"><filename>pkgs/applications/misc/adobe-reader/default.nix</filename></link>.
Shows how binary-only packages can be supported. In Shows how binary-only packages can be supported. In particular the
particular the <link <link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/adobe-reader/builder.sh">builder</link> xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/adobe-reader/builder.sh">builder</link>
uses <command>patchelf</command> to set the RUNPATH and ELF uses <command>patchelf</command> to set the RUNPATH and ELF interpreter
interpreter of the executables so that the right libraries of the executables so that the right libraries are found at runtime.
are found at runtime.</para> </para>
</listitem> </listitem>
</itemizedlist>
</itemizedlist>
</para> </para>
<para>
<para>Some notes: Some notes:
<itemizedlist>
<itemizedlist> <listitem>
<para>
<listitem> All <varname linkend="chap-meta">meta</varname> attributes are
<para>All <varname linkend="chap-meta">meta</varname> optional, but it’s still a good idea to provide at least the
attributes are optional, but it’s still a good idea to <varname>description</varname>, <varname>homepage</varname> and
provide at least the <varname>description</varname>, <varname
<varname>homepage</varname> and <varname linkend="sec-meta-license">license</varname>.
linkend="sec-meta-license">license</varname>.</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>You can use <command>nix-prefetch-url</command> (or similar nix-prefetch-git, etc) You can use <command>nix-prefetch-url</command> (or similar
<replaceable>url</replaceable> to get the SHA-256 hash of nix-prefetch-git, etc) <replaceable>url</replaceable> to get the
source distributions. There are similar commands as <command>nix-prefetch-git</command> and SHA-256 hash of source distributions. There are similar commands as
<command>nix-prefetch-hg</command> available in <literal>nix-prefetch-scripts</literal> package.</para> <command>nix-prefetch-git</command> and
</listitem> <command>nix-prefetch-hg</command> available in
<literal>nix-prefetch-scripts</literal> package.
<listitem> </para>
<para>A list of schemes for <literal>mirror://</literal> </listitem>
URLs can be found in <link <listitem>
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/build-support/fetchurl/mirrors.nix"><filename>pkgs/build-support/fetchurl/mirrors.nix</filename></link>.</para> <para>
</listitem> A list of schemes for <literal>mirror://</literal> URLs can be found in
<link
</itemizedlist> xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/build-support/fetchurl/mirrors.nix"><filename>pkgs/build-support/fetchurl/mirrors.nix</filename></link>.
</para>
</listitem>
</itemizedlist>
</para> </para>
<para>
<para>The exact syntax and semantics of the Nix expression The exact syntax and semantics of the Nix expression language, including
language, including the built-in function, are described in the the built-in function, are described in the Nix manual in the
Nix manual in the <link <link
xlink:href="http://hydra.nixos.org/job/nix/trunk/tarball/latest/download-by-type/doc/manual/#chap-writing-nix-expressions">chapter xlink:href="http://hydra.nixos.org/job/nix/trunk/tarball/latest/download-by-type/doc/manual/#chap-writing-nix-expressions">chapter
on writing Nix expressions</link>.</para> on writing Nix expressions</link>.
</para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Add a call to the function defined in the previous step to Add a call to the function defined in the previous step to
<link <link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/all-packages.nix"><filename>pkgs/top-level/all-packages.nix</filename></link> xlink:href="https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/all-packages.nix"><filename>pkgs/top-level/all-packages.nix</filename></link>
with some descriptive name for the variable, with some descriptive name for the variable, e.g.
e.g. <varname>libfoo</varname>. <varname>libfoo</varname>.
<screen>
<screen>
$ emacs pkgs/top-level/all-packages.nix</screen> $ emacs pkgs/top-level/all-packages.nix</screen>
</para> </para>
<para>
<para>The attributes in that file are sorted by category (like The attributes in that file are sorted by category (like “Development /
“Development / Libraries”) that more-or-less correspond to the Libraries”) that more-or-less correspond to the directory structure of
directory structure of Nixpkgs, and then by attribute name.</para> Nixpkgs, and then by attribute name.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>To test whether the package builds, run the following command <para>
from the root of the nixpkgs source tree: To test whether the package builds, run the following command from the
root of the nixpkgs source tree:
<screen> <screen>
$ nix-build -A libfoo</screen> $ nix-build -A libfoo</screen>
where <varname>libfoo</varname> should be the variable name defined in the
where <varname>libfoo</varname> should be the variable name previous step. You may want to add the flag <option>-K</option> to keep
defined in the previous step. You may want to add the flag the temporary build directory in case something fails. If the build
<option>-K</option> to keep the temporary build directory in case succeeds, a symlink <filename>./result</filename> to the package in the
something fails. If the build succeeds, a symlink Nix store is created.
<filename>./result</filename> to the package in the Nix store is
created.</para>
</listitem>
<listitem>
<para>If you want to install the package into your profile
(optional), do
<screen>
$ nix-env -f . -iA libfoo</screen>
</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Optionally commit the new package and open a pull request, or send a patch to If you want to install the package into your profile (optional), do
<literal>https://groups.google.com/forum/#!forum/nix-devel</literal>.</para> <screen>
</listitem> $ nix-env -f . -iA libfoo</screen>
</para>
</listitem>
</orderedlist> <listitem>
<para>
</para> Optionally commit the new package and open a pull request, or send a patch
to <literal>https://groups.google.com/forum/#!forum/nix-devel</literal>.
</para>
</listitem>
</orderedlist>
</para>
</chapter> </chapter>

1341
doc/release-notes.xml
View File

File diff suppressed because it is too large Load Diff

715
doc/reviewing-contributions.xml
View File

@ -3,95 +3,148 @@
xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xi="http://www.w3.org/2001/XInclude"
version="5.0" version="5.0"
xml:id="sec-reviewing-contributions"> xml:id="sec-reviewing-contributions">
<title>Reviewing contributions</title>
<title>Reviewing contributions</title> <warning>
<para>
<warning> The following section is a draft and reviewing policy is still being
<para>The following section is a draft and reviewing policy is still being discussed.
discussed.</para> </para>
</warning> </warning>
<para>
<para>The nixpkgs projects receives a fairly high number of contributions via The nixpkgs projects receives a fairly high number of contributions via
GitHub pull-requests. Reviewing and approving these is an important task and a GitHub pull-requests. Reviewing and approving these is an important task and
way to contribute to the project.</para> a way to contribute to the project.
</para>
<para>The high change rate of nixpkgs make any pull request that is open for <para>
long enough subject to conflicts that will require extra work from the The high change rate of nixpkgs make any pull request that is open for long
submitter or the merger. Reviewing pull requests in a timely manner and being enough subject to conflicts that will require extra work from the submitter
or the merger. Reviewing pull requests in a timely manner and being
responsive to the comments is the key to avoid these. GitHub provides sort responsive to the comments is the key to avoid these. GitHub provides sort
filters that can be used to see the <link filters that can be used to see the
<link
xlink:href="https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-desc">most xlink:href="https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-desc">most
recently</link> and the <link recently</link> and the
<link
xlink:href="https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-asc">least xlink:href="https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-asc">least
recently</link> updated pull-requests. recently</link> updated pull-requests. We highly encourage looking at
We highly encourage looking at <link xlink:href="https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+review%3Anone+status%3Asuccess+-label%3A%222.status%3A+work-in-progress%22+no%3Aproject+no%3Aassignee+no%3Amilestone"> <link xlink:href="https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+review%3Anone+status%3Asuccess+-label%3A%222.status%3A+work-in-progress%22+no%3Aproject+no%3Aassignee+no%3Amilestone">
this list of ready to merge, unreviewed pull requests</link>.</para> this list of ready to merge, unreviewed pull requests</link>.
</para>
<para>When reviewing a pull request, please always be nice and polite. <para>
When reviewing a pull request, please always be nice and polite.
Controversial changes can lead to controversial opinions, but it is important Controversial changes can lead to controversial opinions, but it is important
to respect every community members and their work.</para> to respect every community members and their work.
</para>
<para>GitHub provides reactions, they are a simple and quick way to provide <para>
GitHub provides reactions, they are a simple and quick way to provide
feedback to pull-requests or any comments. The thumb-down reaction should be feedback to pull-requests or any comments. The thumb-down reaction should be
used with care and if possible accompanied with some explanations so the used with care and if possible accompanied with some explanations so the
submitter has directions to improve his contribution.</para> submitter has directions to improve his contribution.
</para>
<para>Pull-requests reviews should include a list of what has been reviewed in a <para>
comment, so other reviewers and mergers can know the state of the Pull-requests reviews should include a list of what has been reviewed in a
review.</para> comment, so other reviewers and mergers can know the state of the review.
</para>
<para>All the review template samples provided in this section are generic and <para>
All the review template samples provided in this section are generic and
meant as examples. Their usage is optional and the reviewer is free to adapt meant as examples. Their usage is optional and the reviewer is free to adapt
them to his liking.</para> them to his liking.
</para>
<section>
<title>Package updates</title>
<section><title>Package updates</title> <para>
A package update is the most trivial and common type of pull-request. These
pull-requests mainly consist in updating the version part of the package
name and the source hash.
</para>
<para>A package update is the most trivial and common type of pull-request. <para>
These pull-requests mainly consist in updating the version part of the package It can happen that non trivial updates include patches or more complex
name and the source hash.</para> changes.
<para>It can happen that non trivial updates include patches or more complex </para>
changes.</para>
<para>Reviewing process:</para> <para>
Reviewing process:
</para>
<itemizedlist> <itemizedlist>
<listitem><para>Add labels to the pull-request. (Requires commit <listitem>
rights)</para> <para>
Add labels to the pull-request. (Requires commit rights)
</para>
<itemizedlist> <itemizedlist>
<listitem><para><literal>8.has: package (update)</literal> and any topic <listitem>
label that fit the updated package.</para></listitem> <para>
<literal>8.has: package (update)</literal> and any topic label that fit
the updated package.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the package versioning is fitting the <listitem>
guidelines.</para></listitem> <para>
<listitem><para>Ensure that the commit text is fitting the Ensure that the package versioning is fitting the guidelines.
guidelines.</para></listitem> </para>
<listitem><para>Ensure that the package maintainers are notified.</para> </listitem>
<listitem>
<para>
Ensure that the commit text is fitting the guidelines.
</para>
</listitem>
<listitem>
<para>
Ensure that the package maintainers are notified.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>mention-bot usually notify GitHub users based on the <listitem>
submitted changes, but it can happen that it misses some of the <para>
package maintainers.</para></listitem> mention-bot usually notify GitHub users based on the submitted changes,
but it can happen that it misses some of the package maintainers.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the meta field contains correct <listitem>
information.</para> <para>
Ensure that the meta field contains correct information.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>License can change with version updates, so it should be <listitem>
checked to be fitting upstream license.</para></listitem> <para>
<listitem><para>If the package has no maintainer, a maintainer must be License can change with version updates, so it should be checked to be
set. This can be the update submitter or a community member that fitting upstream license.
accepts to take maintainership of the package.</para></listitem> </para>
</listitem>
<listitem>
<para>
If the package has no maintainer, a maintainer must be set. This can be
the update submitter or a community member that accepts to take
maintainership of the package.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the code contains no typos.</para></listitem> <listitem>
<listitem><para>Building the package locally.</para> <para>
Ensure that the code contains no typos.
</para>
</listitem>
<listitem>
<para>
Building the package locally.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>Pull-requests are often targeted to the master or staging <listitem>
branch so building the pull-request locally as it is submitted can <para>
trigger a large amount of source builds.</para> Pull-requests are often targeted to the master or staging branch so
<para>It is possible to rebase the changes on nixos-unstable or building the pull-request locally as it is submitted can trigger a large
nixpkgs-unstable for easier review by running the following commands amount of source builds.
from a nixpkgs clone. </para>
<para>
It is possible to rebase the changes on nixos-unstable or
nixpkgs-unstable for easier review by running the following commands
from a nixpkgs clone.
<screen> <screen>
$ git remote add channels https://github.com/NixOS/nixpkgs-channels.git <co $ git remote add channels https://github.com/NixOS/nixpkgs-channels.git <co
xml:id='reviewing-rebase-1' /> xml:id='reviewing-rebase-1' />
@ -100,43 +153,56 @@ $ git fetch origin pull/PRNUMBER/head <co xml:id='reviewing-rebase-3' />
$ git rebase --onto nixos-unstable BASEBRANCH FETCH_HEAD <co $ git rebase --onto nixos-unstable BASEBRANCH FETCH_HEAD <co
xml:id='reviewing-rebase-4' /> xml:id='reviewing-rebase-4' />
</screen> </screen>
<calloutlist> <calloutlist>
<callout arearefs='reviewing-rebase-1'> <callout arearefs='reviewing-rebase-1'>
<para>This should be done only once to be able to fetch channel <para>
branches from the nixpkgs-channels repository.</para> This should be done only once to be able to fetch channel branches
</callout> from the nixpkgs-channels repository.
<callout arearefs='reviewing-rebase-2'> </para>
<para>Fetching the nixos-unstable branch.</para> </callout>
</callout> <callout arearefs='reviewing-rebase-2'>
<callout arearefs='reviewing-rebase-3'> <para>
<para>Fetching the pull-request changes, <varname>PRNUMBER</varname> Fetching the nixos-unstable branch.
is the number at the end of the pull-request title and </para>
<varname>BASEBRANCH</varname> the base branch of the </callout>
pull-request.</para> <callout arearefs='reviewing-rebase-3'>
</callout> <para>
<callout arearefs='reviewing-rebase-4'> Fetching the pull-request changes, <varname>PRNUMBER</varname> is the
<para>Rebasing the pull-request changes to the nixos-unstable number at the end of the pull-request title and
branch.</para> <varname>BASEBRANCH</varname> the base branch of the pull-request.
</callout> </para>
</calloutlist> </callout>
</para> <callout arearefs='reviewing-rebase-4'>
</listitem> <para>
<listitem> Rebasing the pull-request changes to the nixos-unstable branch.
<para>The <link xlink:href="https://github.com/madjar/nox">nox</link> </para>
tool can be used to review a pull-request content in a single command. </callout>
It doesn't rebase on a channel branch so it might trigger multiple </calloutlist>
source builds. <varname>PRNUMBER</varname> should be replaced by the </para>
number at the end of the pull-request title.</para> </listitem>
<listitem>
<para>
The <link xlink:href="https://github.com/madjar/nox">nox</link> tool can
be used to review a pull-request content in a single command. It doesn't
rebase on a channel branch so it might trigger multiple source builds.
<varname>PRNUMBER</varname> should be replaced by the number at the end
of the pull-request title.
</para>
<screen> <screen>
$ nix-shell -p nox --run "nox-review -k pr PRNUMBER" $ nix-shell -p nox --run "nox-review -k pr PRNUMBER"
</screen> </screen>
</listitem> </listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Running every binary.</para></listitem> <listitem>
</itemizedlist> <para>
Running every binary.
</para>
</listitem>
</itemizedlist>
<example><title>Sample template for a package update review</title> <example>
<title>Sample template for a package update review</title>
<screen> <screen>
##### Reviewed points ##### Reviewed points
@ -150,55 +216,105 @@ $ nix-shell -p nox --run "nox-review -k pr PRNUMBER"
##### Comments ##### Comments
</screen></example> </screen>
</section> </example>
</section>
<section>
<title>New packages</title>
<section><title>New packages</title> <para>
New packages are a common type of pull-requests. These pull requests
consists in adding a new nix-expression for a package.
</para>
<para>New packages are a common type of pull-requests. These pull requests <para>
consists in adding a new nix-expression for a package.</para> Reviewing process:
</para>
<para>Reviewing process:</para> <itemizedlist>
<listitem>
<itemizedlist> <para>
<listitem><para>Add labels to the pull-request. (Requires commit Add labels to the pull-request. (Requires commit rights)
rights)</para> </para>
<itemizedlist> <itemizedlist>
<listitem><para><literal>8.has: package (new)</literal> and any topic <listitem>
label that fit the new package.</para></listitem> <para>
<literal>8.has: package (new)</literal> and any topic label that fit the
new package.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the package versioning is fitting the <listitem>
guidelines.</para></listitem> <para>
<listitem><para>Ensure that the commit name is fitting the Ensure that the package versioning is fitting the guidelines.
guidelines.</para></listitem> </para>
<listitem><para>Ensure that the meta field contains correct </listitem>
information.</para> <listitem>
<para>
Ensure that the commit name is fitting the guidelines.
</para>
</listitem>
<listitem>
<para>
Ensure that the meta field contains correct information.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>License must be checked to be fitting upstream <listitem>
license.</para></listitem> <para>
<listitem><para>Platforms should be set or the package will not get binary License must be checked to be fitting upstream license.
substitutes.</para></listitem> </para>
<listitem><para>A maintainer must be set, this can be the package </listitem>
submitter or a community member that accepts to take maintainership of <listitem>
the package.</para></listitem> <para>
Platforms should be set or the package will not get binary substitutes.
</para>
</listitem>
<listitem>
<para>
A maintainer must be set, this can be the package submitter or a
community member that accepts to take maintainership of the package.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the code contains no typos.</para></listitem> <listitem>
<listitem><para>Ensure the package source.</para> <para>
Ensure that the code contains no typos.
</para>
</listitem>
<listitem>
<para>
Ensure the package source.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>Mirrors urls should be used when <listitem>
available.</para></listitem> <para>
<listitem><para>The most appropriate function should be used (e.g. Mirrors urls should be used when available.
packages from GitHub should use </para>
<literal>fetchFromGitHub</literal>).</para></listitem> </listitem>
<listitem>
<para>
The most appropriate function should be used (e.g. packages from GitHub
should use <literal>fetchFromGitHub</literal>).
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Building the package locally.</para></listitem> <listitem>
<listitem><para>Running every binary.</para></listitem> <para>
</itemizedlist> Building the package locally.
</para>
</listitem>
<listitem>
<para>
Running every binary.
</para>
</listitem>
</itemizedlist>
<example><title>Sample template for a new package review</title> <example>
<title>Sample template for a new package review</title>
<screen> <screen>
##### Reviewed points ##### Reviewed points
@ -220,58 +336,107 @@ $ nix-shell -p nox --run "nox-review -k pr PRNUMBER"
##### Comments ##### Comments
</screen></example> </screen>
</section> </example>
</section>
<section>
<title>Module updates</title>
<section><title>Module updates</title> <para>
Module updates are submissions changing modules in some ways. These often
contains changes to the options or introduce new options.
</para>
<para>Module updates are submissions changing modules in some ways. These often <para>
contains changes to the options or introduce new options.</para> Reviewing process
</para>
<para>Reviewing process</para> <itemizedlist>
<listitem>
<itemizedlist> <para>
<listitem><para>Add labels to the pull-request. (Requires commit Add labels to the pull-request. (Requires commit rights)
rights)</para> </para>
<itemizedlist> <itemizedlist>
<listitem><para><literal>8.has: module (update)</literal> and any topic <listitem>
label that fit the module.</para></listitem> <para>
<literal>8.has: module (update)</literal> and any topic label that fit
the module.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the module maintainers are notified.</para> <listitem>
<para>
Ensure that the module maintainers are notified.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>Mention-bot notify GitHub users based on the submitted <listitem>
changes, but it can happen that it miss some of the package <para>
maintainers.</para></listitem> Mention-bot notify GitHub users based on the submitted changes, but it
can happen that it miss some of the package maintainers.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that the module tests, if any, are <listitem>
succeeding.</para></listitem> <para>
<listitem><para>Ensure that the introduced options are correct.</para> Ensure that the module tests, if any, are succeeding.
</para>
</listitem>
<listitem>
<para>
Ensure that the introduced options are correct.
</para>
<itemizedlist> <itemizedlist>
<listitem><para>Type should be appropriate (string related types differs <listitem>
in their merging capabilities, <literal>optionSet</literal> and <para>
<literal>string</literal> types are deprecated).</para></listitem> Type should be appropriate (string related types differs in their
<listitem><para>Description, default and example should be merging capabilities, <literal>optionSet</literal> and
provided.</para></listitem> <literal>string</literal> types are deprecated).
</para>
</listitem>
<listitem>
<para>
Description, default and example should be provided.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that option changes are backward compatible.</para> <listitem>
<para>
Ensure that option changes are backward compatible.
</para>
<itemizedlist> <itemizedlist>
<listitem><para><literal>mkRenamedOptionModule</literal> and <listitem>
<literal>mkAliasOptionModule</literal> functions provide way to make <para>
option changes backward compatible.</para></listitem> <literal>mkRenamedOptionModule</literal> and
<literal>mkAliasOptionModule</literal> functions provide way to make
option changes backward compatible.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</listitem> </listitem>
<listitem><para>Ensure that removed options are declared with <listitem>
<literal>mkRemovedOptionModule</literal></para></listitem> <para>
<listitem><para>Ensure that changes that are not backward compatible are Ensure that removed options are declared with
mentioned in release notes.</para></listitem> <literal>mkRemovedOptionModule</literal>
<listitem><para>Ensure that documentations affected by the change is </para>
updated.</para></listitem> </listitem>
</itemizedlist> <listitem>
<para>
Ensure that changes that are not backward compatible are mentioned in
release notes.
</para>
</listitem>
<listitem>
<para>
Ensure that documentations affected by the change is updated.
</para>
</listitem>
</itemizedlist>
<example><title>Sample template for a module update review</title> <example>
<title>Sample template for a module update review</title>
<screen> <screen>
##### Reviewed points ##### Reviewed points
@ -288,51 +453,89 @@ $ nix-shell -p nox --run "nox-review -k pr PRNUMBER"
##### Comments ##### Comments
</screen></example> </screen>
</section> </example>
</section>
<section>
<title>New modules</title>
<section><title>New modules</title> <para>
New modules submissions introduce a new module to NixOS.
</para>
<para>New modules submissions introduce a new module to NixOS.</para> <itemizedlist>
<listitem>
<para>
Add labels to the pull-request. (Requires commit rights)
</para>
<itemizedlist>
<listitem>
<para>
<literal>8.has: module (new)</literal> and any topic label that fit the
module.
</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
Ensure that the module tests, if any, are succeeding.
</para>
</listitem>
<listitem>
<para>
Ensure that the introduced options are correct.
</para>
<itemizedlist>
<listitem>
<para>
Type should be appropriate (string related types differs in their
merging capabilities, <literal>optionSet</literal> and
<literal>string</literal> types are deprecated).
</para>
</listitem>
<listitem>
<para>
Description, default and example should be provided.
</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
Ensure that module <literal>meta</literal> field is present
</para>
<itemizedlist>
<listitem>
<para>
Maintainers should be declared in <literal>meta.maintainers</literal>.
</para>
</listitem>
<listitem>
<para>
Module documentation should be declared with
<literal>meta.doc</literal>.
</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
Ensure that the module respect other modules functionality.
</para>
<itemizedlist>
<listitem>
<para>
For example, enabling a module should not open firewall ports by
default.
</para>
</listitem>
</itemizedlist>
</listitem>
</itemizedlist>
<itemizedlist> <example>
<listitem><para>Add labels to the pull-request. (Requires commit <title>Sample template for a new module review</title>
rights)</para>
<itemizedlist>
<listitem><para><literal>8.has: module (new)</literal> and any topic label
that fit the module.</para></listitem>
</itemizedlist>
</listitem>
<listitem><para>Ensure that the module tests, if any, are
succeeding.</para></listitem>
<listitem><para>Ensure that the introduced options are correct.</para>
<itemizedlist>
<listitem><para>Type should be appropriate (string related types differs
in their merging capabilities, <literal>optionSet</literal> and
<literal>string</literal> types are deprecated).</para></listitem>
<listitem><para>Description, default and example should be
provided.</para></listitem>
</itemizedlist>
</listitem>
<listitem><para>Ensure that module <literal>meta</literal> field is
present</para>
<itemizedlist>
<listitem><para>Maintainers should be declared in
<literal>meta.maintainers</literal>.</para></listitem>
<listitem><para>Module documentation should be declared with
<literal>meta.doc</literal>.</para></listitem>
</itemizedlist>
</listitem>
<listitem><para>Ensure that the module respect other modules
functionality.</para>
<itemizedlist>
<listitem><para>For example, enabling a module should not open firewall
ports by default.</para></listitem>
</itemizedlist>
</listitem>
</itemizedlist>
<example><title>Sample template for a new module review</title>
<screen> <screen>
##### Reviewed points ##### Reviewed points
@ -350,32 +553,41 @@ $ nix-shell -p nox --run "nox-review -k pr PRNUMBER"
##### Comments ##### Comments
</screen></example> </screen>
</section> </example>
</section>
<section>
<title>Other submissions</title>
<section><title>Other submissions</title> <para>
Other type of submissions requires different reviewing steps.
</para>
<para>Other type of submissions requires different reviewing steps.</para> <para>
If you consider having enough knowledge and experience in a topic and would
like to be a long-term reviewer for related submissions, please contact the
current reviewers for that topic. They will give you information about the
reviewing process. The main reviewers for a topic can be hard to find as
there is no list, but checking past pull-requests to see who reviewed or
git-blaming the code to see who committed to that topic can give some hints.
</para>
<para>If you consider having enough knowledge and experience in a topic and <para>
would like to be a long-term reviewer for related submissions, please contact Container system, boot system and library changes are some examples of the
the current reviewers for that topic. They will give you information about the pull requests fitting this category.
reviewing process. </para>
The main reviewers for a topic can be hard to find as there is no list, but </section>
checking past pull-requests to see who reviewed or git-blaming the code to see <section>
who committed to that topic can give some hints.</para> <title>Merging pull-requests</title>
<para>Container system, boot system and library changes are some examples of the <para>
pull requests fitting this category.</para> It is possible for community members that have enough knowledge and
experience on a special topic to contribute by merging pull requests.
</para>
</section> <para>
TODO: add the procedure to request merging rights.
<section><title>Merging pull-requests</title> </para>
<para>It is possible for community members that have enough knowledge and
experience on a special topic to contribute by merging pull requests.</para>
<para>TODO: add the procedure to request merging rights.</para>
<!-- <!--
The following paragraph about how to deal with unactive contributors is just a The following paragraph about how to deal with unactive contributors is just a
@ -386,10 +598,11 @@ policy.
three months will have their commit rights revoked.</para> three months will have their commit rights revoked.</para>
--> -->
<para>In a case a contributor leaves definitively the Nix community, he should <para>
create an issue or notify the mailing list with references of packages and In a case a contributor leaves definitively the Nix community, he should
modules he maintains so the maintainership can be taken over by other create an issue or notify the mailing list with references of packages and
contributors.</para> modules he maintains so the maintainership can be taken over by other
contributors.
</section> </para>
</section>
</chapter> </chapter>

4033
doc/stdenv.xml
View File

File diff suppressed because it is too large Load Diff

878
doc/submitting-changes.xml
View File

@ -1,447 +1,513 @@
<chapter xmlns="http://docbook.org/ns/docbook" <chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xlink="http://www.w3.org/1999/xlink"
xml:id="chap-submitting-changes"> xml:id="chap-submitting-changes">
<title>Submitting changes</title>
<section>
<title>Making patches</title>
<title>Submitting changes</title> <itemizedlist>
<listitem>
<section> <para>
<title>Making patches</title> Read <link xlink:href="https://nixos.org/nixpkgs/manual/">Manual (How to
write packages for Nix)</link>.
<itemizedlist> </para>
<listitem> </listitem>
<para>Read <link xlink:href="https://nixos.org/nixpkgs/manual/">Manual (How to write packages for Nix)</link>.</para> <listitem>
</listitem> <para>
Fork the repository on GitHub.
<listitem> </para>
<para>Fork the repository on GitHub.</para> </listitem>
</listitem> <listitem>
<para>
<listitem> Create a branch for your future fix.
<para>Create a branch for your future fix. <itemizedlist>
<listitem>
<itemizedlist> <para>
<listitem> You can make branch from a commit of your local
<para>You can make branch from a commit of your local <command>nixos-version</command>. That will help you to avoid additional local compilations. Because you will receive packages from binary cache. <command>nixos-version</command>. That will help you to avoid
additional local compilations. Because you will receive packages from
<itemizedlist> binary cache.
<listitem> <itemizedlist>
<para>For example: <command>nixos-version</command> returns <command>15.05.git.0998212 (Dingo)</command>. So you can do:</para> <listitem>
</listitem> <para>
</itemizedlist> For example: <command>nixos-version</command> returns
<command>15.05.git.0998212 (Dingo)</command>. So you can do:
</para>
</listitem>
</itemizedlist>
<screen> <screen>
$ git checkout 0998212 $ git checkout 0998212
$ git checkout -b 'fix/pkg-name-update' $ git checkout -b 'fix/pkg-name-update'
</screen> </screen>
</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Please avoid working directly on the <command>master</command> branch.</para> Please avoid working directly on the <command>master</command> branch.
</listitem> </para>
</itemizedlist> </listitem>
</para> </itemizedlist>
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>Make commits of logical units. <para>
Make commits of logical units.
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para>If you removed pkgs, made some major NixOS changes etc., write about them in <command>nixos/doc/manual/release-notes/rl-unstable.xml</command>.</para> <para>
</listitem> If you removed pkgs, made some major NixOS changes etc., write about
</itemizedlist> them in
</para> <command>nixos/doc/manual/release-notes/rl-unstable.xml</command>.
</listitem> </para>
</listitem>
<listitem> </itemizedlist>
<para>Check for unnecessary whitespace with <command>git diff --check</command> before committing.</para> </para>
</listitem> </listitem>
<listitem>
<listitem> <para>
<para>Format the commit in a following way:</para> Check for unnecessary whitespace with <command>git diff --check</command>
before committing.
</para>
</listitem>
<listitem>
<para>
Format the commit in a following way:
</para>
<programlisting> <programlisting>
(pkg-name | nixos/&lt;module>): (from -> to | init at version | refactor | etc) (pkg-name | nixos/&lt;module>): (from -> to | init at version | refactor | etc)
Additional information. Additional information.
</programlisting> </programlisting>
<itemizedlist>
<listitem>
<para>
Examples:
<itemizedlist>
<listitem>
<para>
<command>nginx: init at 2.0.1</command>
</para>
</listitem>
<listitem>
<para>
<command>firefox: 54.0.1 -> 55.0</command>
</para>
</listitem>
<listitem>
<para>
<command>nixos/hydra: add bazBaz option</command>
</para>
</listitem>
<listitem>
<para>
<command>nixos/nginx: refactor config generation</command>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
Test your changes. If you work with
<itemizedlist>
<listitem>
<para>
nixpkgs:
<itemizedlist>
<listitem>
<para>
update pkg ->
<itemizedlist>
<listitem>
<para>
<command>nix-env -i pkg-name -f &lt;path to your local nixpkgs
folder&gt;</command>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
add pkg ->
<itemizedlist>
<listitem>
<para>
Make sure it's in
<command>pkgs/top-level/all-packages.nix</command>
</para>
</listitem>
<listitem>
<para>
<command>nix-env -i pkg-name -f &lt;path to your local nixpkgs
folder&gt;</command>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
<emphasis>If you don't want to install pkg in you
profile</emphasis>.
<itemizedlist>
<listitem>
<para>
<command>nix-build -A pkg-attribute-name &lt;path to your local
nixpkgs folder&gt;/default.nix</command> and check results in the
folder <command>result</command>. It will appear in the same
directory where you did <command>nix-build</command>.
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
If you did <command>nix-env -i pkg-name</command> you can do
<command>nix-env -e pkg-name</command> to uninstall it from your
system.
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
NixOS and its modules:
<itemizedlist>
<listitem>
<para>
You can add new module to your NixOS configuration file (usually
it's <command>/etc/nixos/configuration.nix</command>). And do
<command>sudo nixos-rebuild test -I nixpkgs=&lt;path to your local
nixpkgs folder&gt; --fast</command>.
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
If you have commits <command>pkg-name: oh, forgot to insert
whitespace</command>: squash commits in this case. Use <command>git rebase
-i</command>.
</para>
</listitem>
<listitem>
<para>
Rebase you branch against current <command>master</command>.
</para>
</listitem>
</itemizedlist>
</section>
<section>
<title>Submitting changes</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para>Examples: <para>
Push your changes to your fork of nixpkgs.
<itemizedlist> </para>
<listitem> </listitem>
<para> <listitem>
<command>nginx: init at 2.0.1</command> <para>
</para> Create pull request:
</listitem> <itemizedlist>
<listitem>
<listitem> <para>
<para> Write the title in format <command>(pkg-name | nixos/&lt;module>):
<command>firefox: 54.0.1 -> 55.0</command> improvement</command>.
</para> <itemizedlist>
</listitem> <listitem>
<para>
<listitem> If you update the pkg, write versions <command>from -> to</command>.
<para> </para>
<command>nixos/hydra: add bazBaz option</command> </listitem>
</para> </itemizedlist>
</listitem> </para>
</listitem>
<listitem> <listitem>
<para> <para>
<command>nixos/nginx: refactor config generation</command> Write in comment if you have tested your patch. Do not rely much on
</para> <command>TravisCI</command>.
</listitem> </para>
</itemizedlist> </listitem>
</para> <listitem>
</listitem> <para>
</itemizedlist> If you make an improvement, write about your motivation.
</listitem> </para>
</listitem>
<listitem> <listitem>
<para>Test your changes. If you work with <para>
Notify maintainers of the package. For example add to the message:
<itemizedlist> <command>cc @jagajaga @domenkozar</command>.
<listitem> </para>
<para>nixpkgs: </listitem>
</itemizedlist>
<itemizedlist> </para>
<listitem> </listitem>
<para>update pkg -> </itemizedlist>
</section>
<itemizedlist> <section>
<listitem>
<para>
<command>nix-env -i pkg-name -f &lt;path to your local nixpkgs folder&gt;</command>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>add pkg ->
<itemizedlist>
<listitem>
<para>Make sure it's in <command>pkgs/top-level/all-packages.nix</command>
</para>
</listitem>
<listitem>
<para>
<command>nix-env -i pkg-name -f &lt;path to your local nixpkgs folder&gt;</command>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>
<emphasis>If you don't want to install pkg in you profile</emphasis>.
<itemizedlist>
<listitem>
<para>
<command>nix-build -A pkg-attribute-name &lt;path to your local nixpkgs folder&gt;/default.nix</command> and check results in the folder <command>result</command>. It will appear in the same directory where you did <command>nix-build</command>.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>If you did <command>nix-env -i pkg-name</command> you can do <command>nix-env -e pkg-name</command> to uninstall it from your system.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>NixOS and its modules:
<itemizedlist>
<listitem>
<para>You can add new module to your NixOS configuration file (usually it's <command>/etc/nixos/configuration.nix</command>).
And do <command>sudo nixos-rebuild test -I nixpkgs=&lt;path to your local nixpkgs folder&gt; --fast</command>.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>If you have commits <command>pkg-name: oh, forgot to insert whitespace</command>: squash commits in this case. Use <command>git rebase -i</command>.</para>
</listitem>
<listitem>
<para>Rebase you branch against current <command>master</command>.</para>
</listitem>
</itemizedlist>
</section>
<section>
<title>Submitting changes</title>
<itemizedlist>
<listitem>
<para>Push your changes to your fork of nixpkgs.</para>
</listitem>
<listitem>
<para>Create pull request:
<itemizedlist>
<listitem>
<para>Write the title in format <command>(pkg-name | nixos/&lt;module>): improvement</command>.
<itemizedlist>
<listitem>
<para>If you update the pkg, write versions <command>from -> to</command>.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>Write in comment if you have tested your patch. Do not rely much on <command>TravisCI</command>.</para>
</listitem>
<listitem>
<para>If you make an improvement, write about your motivation.</para>
</listitem>
<listitem>
<para>Notify maintainers of the package. For example add to the message: <command>cc @jagajaga @domenkozar</command>.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</section>
<section>
<title>Pull Request Template</title> <title>Pull Request Template</title>
<para> <para>
The pull request template helps determine what steps have been made for a The pull request template helps determine what steps have been made for a
contribution so far, and will help guide maintainers on the status of a contribution so far, and will help guide maintainers on the status of a
change. The motivation section of the PR should include any extra details change. The motivation section of the PR should include any extra details
the title does not address and link any existing issues related to the pull the title does not address and link any existing issues related to the pull
request. request.
</para> </para>
<para>When a PR is created, it will be pre-populated with some checkboxes detailed below:
<para>
When a PR is created, it will be pre-populated with some checkboxes detailed
below:
</para> </para>
<section> <section>
<title>Tested using sandboxing</title> <title>Tested using sandboxing</title>
<para>
When sandbox builds are enabled, Nix will setup an isolated environment <para>
for each build process. It is used to remove further hidden dependencies When sandbox builds are enabled, Nix will setup an isolated environment for
set by the build environment to improve reproducibility. This includes each build process. It is used to remove further hidden dependencies set by
access to the network during the build outside of the build environment to improve reproducibility. This includes access to
<function>fetch*</function> functions and files outside the Nix store. the network during the build outside of <function>fetch*</function>
Depending on the operating system access to other resources are blocked functions and files outside the Nix store. Depending on the operating
as well (ex. inter process communication is isolated on Linux); see <link system access to other resources are blocked as well (ex. inter process
communication is isolated on Linux); see
<link
xlink:href="https://nixos.org/nix/manual/#description-45">build-use-sandbox</link> xlink:href="https://nixos.org/nix/manual/#description-45">build-use-sandbox</link>
in Nix manual for details. in Nix manual for details.
</para> </para>
<para>
Sandboxing is not enabled by default in Nix due to a small performance <para>
hit on each build. In pull requests for <link Sandboxing is not enabled by default in Nix due to a small performance hit
xlink:href="https://github.com/NixOS/nixpkgs/">nixpkgs</link> people on each build. In pull requests for
are asked to test builds with sandboxing enabled (see <literal>Tested <link
using sandboxing</literal> in the pull request template) because xlink:href="https://github.com/NixOS/nixpkgs/">nixpkgs</link>
in<link people are asked to test builds with sandboxing enabled (see
<literal>Tested using sandboxing</literal> in the pull request template)
because
in<link
xlink:href="https://nixos.org/hydra/">https://nixos.org/hydra/</link> xlink:href="https://nixos.org/hydra/">https://nixos.org/hydra/</link>
sandboxing is also used. sandboxing is also used.
</para> </para>
<para>
Depending if you use NixOS or other platforms you can use one of the <para>
following methods to enable sandboxing <emphasis role="bold">before</emphasis> building the package: Depending if you use NixOS or other platforms you can use one of the
<itemizedlist> following methods to enable sandboxing
<listitem> <emphasis role="bold">before</emphasis> building the package:
<para> <itemizedlist>
<emphasis role="bold">Globally enable sandboxing on NixOS</emphasis>: <listitem>
add the following to <para>
<filename>configuration.nix</filename> <emphasis role="bold">Globally enable sandboxing on NixOS</emphasis>:
<screen>nix.useSandbox = true;</screen> add the following to <filename>configuration.nix</filename>
</para> <screen>nix.useSandbox = true;</screen>
</listitem> </para>
<listitem> </listitem>
<para> <listitem>
<emphasis role="bold">Globally enable sandboxing on non-NixOS platforms</emphasis>: <para>
add the following to: <filename>/etc/nix/nix.conf</filename> <emphasis role="bold">Globally enable sandboxing on non-NixOS
<screen>build-use-sandbox = true</screen> platforms</emphasis>: add the following to:
</para> <filename>/etc/nix/nix.conf</filename>
</listitem> <screen>build-use-sandbox = true</screen>
</itemizedlist> </para>
</para> </listitem>
</itemizedlist>
</para>
</section> </section>
<section> <section>
<title>Built on platform(s)</title> <title>Built on platform(s)</title>
<para>
Many Nix packages are designed to run on multiple <para>
platforms. As such, it's important to let the maintainer know which Many Nix packages are designed to run on multiple platforms. As such, it's
platforms your changes have been tested on. It's not always practical to important to let the maintainer know which platforms your changes have been
test a change on all platforms, and is not required for a pull request to tested on. It's not always practical to test a change on all platforms, and
be merged. Only check the systems you tested the build on in this is not required for a pull request to be merged. Only check the systems you
section. tested the build on in this section.
</para> </para>
</section> </section>
<section> <section>
<title>Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)</title> <title>Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)</title>
<para>
Packages with automated tests are much more likely to be merged in a <para>
timely fashion because it doesn't require as much manual testing by the Packages with automated tests are much more likely to be merged in a timely
maintainer to verify the functionality of the package. If there are fashion because it doesn't require as much manual testing by the maintainer
existing tests for the package, they should be run to verify your changes to verify the functionality of the package. If there are existing tests for
do not break the tests. Tests only apply to packages with NixOS modules the package, they should be run to verify your changes do not break the
defined and can only be run on Linux. For more details on writing and tests. Tests only apply to packages with NixOS modules defined and can only
running tests, see the <link be run on Linux. For more details on writing and running tests, see the
<link
xlink:href="https://nixos.org/nixos/manual/index.html#sec-nixos-tests">section xlink:href="https://nixos.org/nixos/manual/index.html#sec-nixos-tests">section
in the NixOS manual</link>. in the NixOS manual</link>.
</para> </para>
</section> </section>
<section> <section>
<title>Tested compilation of all pkgs that depend on this change using <command>nox-review</command></title> <title>Tested compilation of all pkgs that depend on this change using <command>nox-review</command></title>
<para>
If you are updating a package's version, you can use nox to make sure all <para>
packages that depend on the updated package still compile correctly. This If you are updating a package's version, you can use nox to make sure all
can be done using the nox utility. The <command>nox-review</command> packages that depend on the updated package still compile correctly. This
utility can look for and build all dependencies either based on can be done using the nox utility. The <command>nox-review</command>
uncommited changes with the <literal>wip</literal> option or specifying a utility can look for and build all dependencies either based on uncommited
github pull request number. changes with the <literal>wip</literal> option or specifying a github pull
</para> request number.
<para> </para>
review uncommitted changes:
<screen>nix-shell -p nox --run "nox-review wip"</screen> <para>
</para> review uncommitted changes:
<para> <screen>nix-shell -p nox --run "nox-review wip"</screen>
review changes from pull request number 12345: </para>
<screen>nix-shell -p nox --run "nox-review pr 12345"</screen>
</para> <para>
review changes from pull request number 12345:
<screen>nix-shell -p nox --run "nox-review pr 12345"</screen>
</para>
</section> </section>
<section> <section>
<title>Tested execution of all binary files (usually in <filename>./result/bin/</filename>)</title> <title>Tested execution of all binary files (usually in <filename>./result/bin/</filename>)</title>
<para>
It's important to test any executables generated by a build when you <para>
change or create a package in nixpkgs. This can be done by looking in It's important to test any executables generated by a build when you change
<filename>./result/bin</filename> and running any files in there, or at a or create a package in nixpkgs. This can be done by looking in
minimum, the main executable for the package. For example, if you make a change <filename>./result/bin</filename> and running any files in there, or at a
to <package>texlive</package>, you probably would only check the binaries minimum, the main executable for the package. For example, if you make a
associated with the change you made rather than testing all of them. change to <package>texlive</package>, you probably would only check the
</para> binaries associated with the change you made rather than testing all of
them.
</para>
</section> </section>
<section> <section>
<title>Meets nixpkgs contribution standards</title> <title>Meets nixpkgs contribution standards</title>
<para>
The last checkbox is fits <link <para>
The last checkbox is fits
<link
xlink:href="https://github.com/NixOS/nixpkgs/blob/master/.github/CONTRIBUTING.md">CONTRIBUTING.md</link>. xlink:href="https://github.com/NixOS/nixpkgs/blob/master/.github/CONTRIBUTING.md">CONTRIBUTING.md</link>.
The contributing document has detailed information on standards the Nix The contributing document has detailed information on standards the Nix
community has for commit messages, reviews, licensing of contributions community has for commit messages, reviews, licensing of contributions you
you make to the project, etc... Everyone should read and understand the make to the project, etc... Everyone should read and understand the
standards the community has for contributing before submitting a pull standards the community has for contributing before submitting a pull
request. request.
</para> </para>
</section> </section>
</section> </section>
<section>
<section> <title>Hotfixing pull requests</title>
<title>Hotfixing pull requests</title>
<itemizedlist>
<listitem>
<para>Make the appropriate changes in you branch.</para>
</listitem>
<listitem>
<para>Don't create additional commits, do
<itemizedlist>
<listitem>
<para><command>git rebase -i</command></para>
</listitem>
<listitem>
<para>
<command>git push --force</command> to your branch.</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</section>
<section>
<title>Commit policy</title>
<itemizedlist>
<listitem>
<para>Commits must be sufficiently tested before being merged, both for the master and staging branches.</para>
</listitem>
<listitem>
<para>Hydra builds for master and staging should not be used as testing platform, it's a build farm for changes that have been already tested.</para>
</listitem>
<listitem>
<para>When changing the bootloader installation process, extra care must be taken. Grub installations cannot be rolled back, hence changes may break people's installations forever. For any non-trivial change to the bootloader please file a PR asking for review, especially from @edolstra.</para>
</listitem>
</itemizedlist>
<section>
<title>Master branch</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
It should only see non-breaking commits that do not cause mass rebuilds. Make the appropriate changes in you branch.
</para> </para>
</listitem> </listitem>
<listitem>
<para>
Don't create additional commits, do
<itemizedlist>
<listitem>
<para>
<command>git rebase -i</command>
</para>
</listitem>
<listitem>
<para>
<command>git push --force</command> to your branch.
</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist> </itemizedlist>
</section> </section>
<section>
<section> <title>Commit policy</title>
<title>Staging branch</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
It's only for non-breaking mass-rebuild commits. That means it's not to Commits must be sufficiently tested before being merged, both for the
be used for testing, and changes must have been well tested already. master and staging branches.
<link xlink:href="http://comments.gmane.org/gmane.linux.distributions.nixos/13447">Read policy here</link>. </para>
</para> </listitem>
</listitem> <listitem>
<listitem> <para>
<para> Hydra builds for master and staging should not be used as testing
If the branch is already in a broken state, please refrain from adding platform, it's a build farm for changes that have been already tested.
extra new breakages. Stabilize it for a few days, merge into master, </para>
then resume development on staging. </listitem>
<link xlink:href="http://hydra.nixos.org/jobset/nixpkgs/staging#tabs-evaluations">Keep an eye on the staging evaluations here</link>. <listitem>
If any fixes for staging happen to be already in master, then master can <para>
be merged into staging. When changing the bootloader installation process, extra care must be
</para> taken. Grub installations cannot be rolled back, hence changes may break
</listitem> people's installations forever. For any non-trivial change to the
bootloader please file a PR asking for review, especially from @edolstra.
</para>
</listitem>
</itemizedlist> </itemizedlist>
</section>
<section> <section>
<title>Stable release branches</title> <title>Master branch</title>
<itemizedlist> <itemizedlist>
<listitem> <listitem>
<para> <para>
If you're cherry-picking a commit to a stable release branch, always use It should only see non-breaking commits that do not cause mass rebuilds.
<command>git cherry-pick -xe</command> and ensure the message contains a </para>
clear description about why this needs to be included in the stable </listitem>
branch. </itemizedlist>
</para> </section>
<para>An example of a cherry-picked commit would look like this:</para>
<screen> <section>
<title>Staging branch</title>
<itemizedlist>
<listitem>
<para>
It's only for non-breaking mass-rebuild commits. That means it's not to
be used for testing, and changes must have been well tested already.
<link xlink:href="http://comments.gmane.org/gmane.linux.distributions.nixos/13447">Read
policy here</link>.
</para>
</listitem>
<listitem>
<para>
If the branch is already in a broken state, please refrain from adding
extra new breakages. Stabilize it for a few days, merge into master, then
resume development on staging.
<link xlink:href="http://hydra.nixos.org/jobset/nixpkgs/staging#tabs-evaluations">Keep
an eye on the staging evaluations here</link>. If any fixes for staging
happen to be already in master, then master can be merged into staging.
</para>
</listitem>
</itemizedlist>
</section>
<section>
<title>Stable release branches</title>
<itemizedlist>
<listitem>
<para>
If you're cherry-picking a commit to a stable release branch, always use
<command>git cherry-pick -xe</command> and ensure the message contains a
clear description about why this needs to be included in the stable
branch.
</para>
<para>
An example of a cherry-picked commit would look like this:
</para>
<screen>
nixos: Refactor the world. nixos: Refactor the world.
The original commit message describing the reason why the world was torn apart. The original commit message describing the reason why the world was torn apart.
@ -451,9 +517,7 @@ Reason: I just had a gut feeling that this would also be wanted by people from
the stone age. the stone age.
</screen> </screen>
</listitem> </listitem>
</itemizedlist> </itemizedlist>
</section> </section>
</section>
</section>
</chapter> </chapter>