Some of the new tests need the `nix` executable on `PATH`.
One also needs `hpack`, but we really don’t want to embiggen (heh) the closure
because of that.
The function isn’t used anywhere and `addErrorContext` is an undocumented
builtin.
The builtin is explicitely qualified at its two uses in the module system.
`attrNamesToStr` is very specific (and pretty trivial), so it doesn’t make sense
to have it in the library.
`traceXMLVal(Marked)` are just a builtin and `trace` and not very useful in
general (trace output should not be parsed anyway).
So far, `mkValueString` defaulted to `toString`,
which is a bad match for most configuration file formats,
especially because how booleans are formatted.
This also improves error messages for unsupported types.
Add a test to codify the formatting.
I’d like to be notified of changes so I can ensure the API always follows the same patterns.
Later (more specific) files replace more general codeowners [according to the documentation](https://help.github.com/articles/about-codeowners/), so I also listed `edolstra` and `nbp` again.
Based on a request by @oxij:
“Can we also rename this file to `maintainers/maintainers-list.nix` while we at
this? Motivation: much saner `git log ./lib`.”
Corrected every handle that had no commits to nixpkgs, manually researched the
correct handles by looking at maintained packages & blames/history on Github.
Based on https://github.com/NixOS/nixpkgs/pull/34842, the
nix-instantiate output was pretty-printed and the validity of the github handles
manually verified, by automatically checking whether the user handles exist on
github (https://github.com/userhandle, status 200 or 404).
Each handle under 5 characters was manually checked (because the collision
probability with non-maintainer accounts is high), each missing entry was
manually researched.
The script used is kept in `maintainers/scripts` as an example of how to work
with the mainainers list through nix’ JSON interface.
- split outputs (doc, man, out)
- a new runtime dependency on getconf
- fix up SCRIPTS_DIR in wrapper
- clean output folder of stray doc files
- add license & maintainer
Tested with `nixos/tests/rabbitmq.nix`, inside a docker image and manually.
This means the generation of values can now be influenced, even down from e.g.
an INI generator.
Breaks the interface of `mkKeyValueDefault` to match its interface to other
generator functions. It might me sensible to rename `mkKeyValue` and
`mkKeyValueSet` to conform to the `toX`-style of generator functions.
Previously only the line numbers of a giant, internally generated XML file were
printed, without any kind of debuggability.
Now at least the mentioned lines are printed with a little bit of surrounding
context (to have something to grep for).
```
manual-combined.xml:4863: element para: Relax-NG validity error : Did not expect element para there
4859 <chapter xmlns="http://docbook.org/ns/docbook" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xi="http://www.w3.org/2001/XInclude" version="5.0" xml:id="sec-writing-modules">
4860
4861 <title>Writing NixOS Modules</title>
4862
4863 <para>NixOS has a modular system for declarative configuration. This
4864 system combines multiple <emphasis>modules</emphasis> to produce the
4865 full system configuration. One of the modules that constitute the
```
Before:
<x> is not a integer between 0 and 100 (inclusively).
(notice that “a” is wrong, it should be “an”)
Now:
<x> is not of type `integer between 0 and 100 (inclusively)'.
This sounds a bit more formal, but circumvents the grammatical problems.
Multi-word type descriptions are also easier to see.
Previously a few clumsy shell globs were used to find haddock output, which
broke every time something in `generic-builder.nix` changed.
Use the new `haddockDir` identifier exported by haskell packages.
WIP
If the `hoogle.nix` file wants to have any sane chance of finding haddock
outputs, the packages need to export the haddock folders as an identifier.
A few TODOs still stand, like passing self to the package instead of needing to
pass it to `haddockDir`. Maybe the exported identifier should be integrated into
the fixpoint somehow instead of using `passthru`?
With `builtins.functionArgs` we can get some information if the first argument
is an attrset and whether the contained fields have default values. Encode that
into the pretty-printed lambda.
* ultrastardx-beta: init at 1.3.5
* libbass, libbass_fx: init at 24
* ultrastar-creator: init at 2017-04-12
* buildSupport/plugins.nix: add diffPlugins
Helper function to compare expected plugin lists to the found plugins.
* ultrastar-manager: init at 2017-05-24
The plugins are built in their own derivations, speeding up (re-)compilation.
The `diffPlugins` function from `beets` is reused to test for changes in the
plugin list on updates.
* beets: switch to diffPlugins
The function is basically just extracted for better reusability.
`unrar` is a non-free package, so it should only be used on request.
rarfile can use the rar-mode of `libarchive` instead, which should work well
enough for most cases.
Since they are executables they should be at top-level.
We build statically to shorten invocation time (at the moment haskell packages
have the problem of too large RPATHs).
[1.1] - 2017-01-11
Added
- Autodetect the number of CPUs available <esmil@mailme.dk>
- Option for source string added to torrent info (`-s`), included in infohash.
Often used by private trackers to create a unique infohash to prevent
peer-leak and the possibility to track the trackers that do use leaked
torrents. Having this option in mktorrent make it possible to create a
infohash accurate torrent to the tracker you want to upload it to
Changed
- Make`-a` (announce list) optional
- Optional announce URL even for private torrents. No need to require announce
for private torrents, they are added by most private trackers anyway and they
modify the infohash so you'd have to redownload their modified torrent with
injected unique announce URL anyway. (@mathieui, fix from @jrwren)
Fixed
- DHT makes trackers optional, so remove the warning
In order to better distinguish foldr from foldl the default name is changed to
foldr, but fold is still a synonym.
Additionally the docs are improved and examples added.
After the change of the bonding options, the examples were not quite correct.
The diff is over-the top because the new `let` needs everything indented.
Also add a small docstring to the `networkd` attr in the networking test.
Profpatsch