Using code:set_path/1 with very large paths is very slow on larger
projects. On my mid-sized project, it seems to take around .4s per
call. Emulating the call with direct path removal (using
code:del_path/1) seems to be quite a lot faster.
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.
The combination of changes to rebar_erlc_compiler, and the fact
that erl_first_files is inherited, caused a regression. To fix
that, ensure every project uses its own .rebar/erlcinfo. While at
it, fix the issue that erl_first_files entries were not included
when initializing the dep digraph.
Reported-by: Louis-Philippe Gauthier
Reported-by: Roland Karlsson
Thanks: Tuncer Ayaz
Using the filename as a prefix is less readable and inconsistent with
the other log messages.
Before:
DEBUG: src/foo.erl depends on...
After:
DEBUG: Dependencies of src/foo.erl ...
* update files
* fix Dialyzer warning
* unconditionally enable info fil
* clean-up inconsistencies
* use term_to_binary compression
* use try...catch instead of case...catch...of
* do not write build info file if the graph is unmodified
* store info file as <base_dir>/.rebarinfo
* properly support list of compile directives
* fix regressions:
- Fix a bug in handling of files to compile first.
- If a file that is depended upon itself depends on other files, make sure
those are compiled first. While at it, rename variables for correctness.
Reported-by: David Robakowski
- Make sure that FirstFiles has no dupes and preserves the proper order.
- headers referenced via -include_lib() were not properly resolved to absolute
filenames
- .erl files found in sub dirs of src_dirs were not properly resolved to
absolute filenames
* Do not parse source files twice while checking for relationship.
* Keep files relationships in a graph.
* The option 'keep_build_info' is introduced. When set to 'true'
the graph will be kept in ebin/.rebar.build.info and will be
used by further compiler calls. The default is 'false'.
Previously, the configuration setting 'mib_opts' in rebar.config
would affect the call to snmpc:compile/2, so that (for example)
verbosity could be controlled. However, the subsequent call to
snmpc:mib_to_hrl/1 did not include any of these options, so it
did not appear to be possible to control the verbosity of the
process of converting a MIB to a .hrl file. To make matters
worse, the default was to dump a full trace -- including debug
output and various logging -- so the act of compiling a large
number of MIBs could result in a huge amount of "noisy" output
that hid any signal (meaningful warnings, errors, etc.).
This commit addresses that issue by replacing the call to
snmpc:mib_to_hrl/1 with a call to snmpc:mib_to_hrl/3 instead,
which includes an "options" argument that, at present, is only
capable of setting verbosity. The verbosity setting is taken
from the 'mib_opts' setting in rebar_config, if present, and
the approriate kind of argument is passed to snmpc:mib_to_hrl/3.
It should be noted that snmpc:mib_to_hrl/3 is not listed in
Erlang's documentation, but does appear in the list of "API"
exports at the top of snmpc.erl in R15B01 (and remains that way
in R16B01), so this appears to be more of a documentation oversight
than the use of a deep, dark function call that was not intended
to be public. snmpc:mib_to_hrl/3 accepts an #options{} record
(defined in lib/srdlib/include/erl_compile.hrl within Erlang's
source distribution), though most of the fields in that record
are ignored by snmpc:mib_to_hrl/3; only verbosity can be controlled
this way.
