If you happen to fetch a zip archive of the git repo and try to build
from that, you may, for example, ask erlc to build src/._rebar.erl.
._* are OS X resource forks and not real .erl files. This may also
happen with network filesystems on OS X. To fix that, limit the
files compiled by rebar to include only those which start with
a letter or a digit.
REBAR will be set to the rebar binary which was executed and runs the
builds. Enables the use of the same binary for rebar invocations as
part of a pre or post hook like so:
${REBAR} escriptize
Since the introduction of -r/--recursive, deps were not properly added
to the code path when running ct, eunit, etc.
To fix that, pass a flag down to process_dir1 and conditionalize
execution of the command. This moves the decision into process_dir1
where we can decide to invoke preprocess/2 and postprocess/2 but not
execute the command.
Without this fix, you'd have to, for example, invoke 'rebar -r ct
skip_deps=true', if you wanted to run base_dir's ct suites with deps on
the code path (while skipping all non-base_dir ct suites).
So, with this patch applied, if you run
$ rebar ct
deps will be on the code path, and only base_dir's ct suites will be
tested.
If you want to test ct suites in base_dir and sub_dirs, you have to run
$ rebar -r ct skip_deps=true
If you want to test ct suites in all dirs, you have to run
$ rebar -r ct
The fix is not specific to ct and applies to all commands.
To be able to add inttest/code_path_no_recurse/deps, I had to fix
.gitignore. While at it, I've updated and fixed all entries.
* Fix arg order:
The order of arguments got inconsistent over time. To fix that, use
the same consistent order in all functions.
* Avoid one erlang:'++'/2 call in process_dir/6.
* Avoid lists:prefix/2 and atom_to_list/1 calls:
We can easily avoid 2 lists:prefix/2 calls and one atom_to_list/1 call
in execute/5 by passing in whether the command is a hook or not. The
resulting code is simpler and easier to read.
If the directory we're about to process contains
reltool.config[.script] and the command to be applied is
'generate', then it's safe to process. We do this to retain the
behavior of specifying {sub_dirs, ["rel"]} and have "rebar generate"
pick up rel/reltool.config[.script]. Without this workaround you'd
have to run "rebar -r generate" (which you don't want to do if you
have deps or other sub_dirs) or "cd rel && rebar generate".
Always-on recursive application of all rebar commands causes too many
issues. Recursive application is required for:
1. dealing with dependencies: get-deps, update-deps, and compile of deps
right after get-deps or update-deps
2. projects with a riak-like apps/ project structure and dev process
The vast majority of projects are not structured like riak. Therefore,
moving forward it's best to (by default) restrict recursive behavior to
dealing with deps. This commit does that and also adds command line and
rebar.config options for controlling or configuring recursion. Also, we
introduce two meta commands: prepare-deps (equivalent to rebar -r
get-deps compile) and refresh-deps (equivalent to rebar -r update-deps
compile). riak-like projects can extend the list of recursive commands
(to include 'eunit' and 'compile') by adding
{recursive_cmds, [eunit, compile]} to rebar.config.
Rebar, when it encounters a lib_dir directive, caches the current code
path, adds the libdir(s) and returns the cached copy of the path. When
rebar has finished processing that directory, it restores the cached
path. This is problematic in the below scenario:
/(lib_dir)->G
A -> B -> C -> D -> E
\-> F -> D -> E
When rebar is finished processing B, it restores the code path to what
it was before it processed B, removing C, D, E and G from the code path.
This means when it comes to process F, neither D or E are in the code
path, so any header includes, rebar plugins or parse transforms will not
be in the code path. Without the lib_dir directive, rebar does no code
path cleanups, so everything works fine.
This change makes rebar only remove the explicit lib_dir code paths it
added and adds an inttest that replicates the above scenario.
Because rebar_core handles skipping apps, we had to specialcase the
handling in the case of update-deps because it has to do its own dep
handling. The way this was done is not particularly clean, but there
currently does not exist another way for a command to signal rebar_core
that it doesn't want rebar_core to pay attention to skip_apps.
With this change, however, you can update-deps even with local
conflicting changes/commits by simply skipping the deps you don't wish
to update, or whitelisting he ones you do wish to update.
This ensures that deps of deps are updated AFTER the dep listing them
is, so that a complicated project with many layers of deps will be
updated correctly. Any new deps encountered along the way are also
cloned, and THEIR deps are also evaluated.
Also added was conflict detection, if a dep has differing versions or
source information, inherited from different places, that will be logged
at the end of update-deps, along with the origin of each conflicting
dep.
- refactor plugin dirs code to be simpler and easier to read
- use erlang-mode's default (%%) comments for portability/consistency
- make sure erlang-mode's indenter is used so that a future whole
buffer indent doesn't get messed up
Tristan Sloughter
Tuncer Ayaz
Tino Breddin
Andrew Thompson
Juan Jose Comellas
Motiejus Jakštys
Dave Smith