Rebuild edocs when headers or config change #314

Closed
wants to merge 2 commits into
from

Projects

None yet

4 participants

@gregrperkins
  • Adds rebar_config:source_file/1 for retrieving config file location
  • Indicate on console when an app's edocs are generated
  • More debug logging for doc command
@dizzyd

In a project with multiple rebar.configs, it's unclear to me if this is going to do the Right Thing, since it will only check the top-most rebar.config file for change. Why does/should rebar.config be considered in this calculation?

@gregrperkins

The rationale was that the edoc_opts can change in the rebar config, which may change the desired output. What does a project with multiple rebar.configs look like; is there an example somewhere to play with?

@dizzyd

I was thinking specifically about situations where an app is a dependency of another app and thus inherits configuration info from the top-level config. I.e. in riak options can be defined at the top-level rebar.config and will be inherited by deps if they don't define their own options. There is some debate about this "feature", but it's something to consider.

Can you please rebase this on latest master? I think we'll probably merge it but want to pick through it a bit more.

@dizzyd

Re-up. I'd like to get this ticket closed.

gregrperkins added some commits Sep 18, 2012
@gregrperkins gregrperkins Rebuild edocs when headers or config change
* Adds rebar_config:source_file/1 for retrieving config file location
* Indicate on console when an app's edocs are generated
* More debug logging for doc command
46cee8f
@gregrperkins gregrperkins Allow doc generation as only target
* Fixes bug when calling ./rebar doc without an ebin directory
d2b29ff
@gregrperkins

Sorry to leave this lingering.

My make check is failing (both on this branch and current master) with:

...
make: [dialyzer_warnings] Error 1 (ignored)
--- dialyzer_reference  2012-11-26 02:56:45.000000000 -0800
+++ dialyzer_warnings   2012-11-26 03:00:53.000000000 -0800
@@ -2,2 +2,2 @@
-rebar_eunit.erl:351: Call to missing or unexported function eunit_test:function_wrapper/2
-rebar_utils.erl:163: Call to missing or unexported function escript:foldl/3
+dialyzer: Could not read PLT file /Users/gregp/.dialyzer_plt: no_such_file
+
make: *** [dialyzer] Error 1

But travis is giving a green light...? Seems like a misconfiguration on my end?

@tuncer

That means you don't have ~/.dialyzer_plt (default PLT used if none specified). You have to build one first. the travis-ci config doesn't run Dialyzer.

@gregrperkins

Ahh, thanks. https://gist.github.com/1139033 was helpful too. All tests passing now. :-)

@jaredmorrow

@gregrperkins ping. We abandoned basho/rebar when rebar/rebar was created. We (basho) are now back to using basho/rebar, is this PR something you'd still like to see? Has it already been merged into rebar/rebar?

@gregrperkins

Never merged to my knowledge, would need to be rebased, probably. I can't drive it.

@roowe roowe pushed a commit to roowe/rebar that referenced this pull request Jul 26, 2014
@talentdeficit talentdeficit update group leaders when restarting user process
ensure any processes with a reference to an old user process as their
group leader are updated to use the new user process. this introduces a
slight delay at startup as the system must wait for the new processes
to be registered. there is a max wait period of three seconds (before
the shell command gives up and throws a timeout error)

fixes #314 ("rebar shell" somehow blocks using io:format in gen_server
handle_call)
68089c6
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment