Cleanup parallel_emerge a bit.

Cleanup parallel_emerge a bit.

- Add support for --jobs flag.
- Add support for --depclean with no arguments.
- Cleanup comments.
- Print out error details the first time a package fails.

TEST=Ran full build_packages --fast
BUG=none

[Nick Sanders, 2010-07-07 19:40:26]

http://codereview.chromium.org/2853031/diff/1/2
File parallel_emerge (right):

http://codereview.chromium.org/2853031/diff/1/2#newcode51
parallel_emerge:51: print " ./parallel_emerge --board=BOARD --jobs=JOBS [emerge
args] package"
what does jobs default to in build_image.. "-1" I thought? Should we detect
negative or zero and reset to MAX_INT or 256 or something?

http://codereview.chromium.org/2853031/diff/1/2#newcode97
parallel_emerge:97: jobs_arg = int(arg.replace("--jobs=", ""))
should we check for intability here? maybe throw in a "try: int(...) except:
error msg" so bad cmdline args have a friendlier failure mode..

http://codereview.chromium.org/2853031/diff/1/2#newcode98
parallel_emerge:98: elif arg[0] == "-" or arg == "y" or arg == "n":
maybe arg.startswith("-") for consistency instead of arg[0] == "-"

[davidjames, 2010-07-08 21:14:05]

I've addressed the review comments. PTAL.

I also updated parallel_emerge to handle the case where we see the following
output from emerge:

"WARNING: Requested packages will not be merged because they are listed in
package.provided:

  sys-devel/automake pulled in by 'world'
  virtual/os-headers pulled in by 'world'
  virtual/libc pulled in by 'world'
  sys-devel/autoconf pulled in by 'world'
  sys-devel/gcc pulled in by 'world'
  sys-devel/m4 pulled in by 'world'
  sys-devel/gnuconfig pulled in by 'world'
  sys-apps/texinfo pulled in by 'world'
  sys-devel/binutils pulled in by 'world'
  sys-devel/bison pulled in by 'world'

This problem can be solved in one of the following ways:

  A) Use emaint to clean offending packages from world (if not installed).
  B) Uninstall offending packages (cleans them from world).
  C) Remove offending entries from package.provided.

The best course of action depends on the reason that an offending
package.provided entry exists."

Previously, parallel_emerge spotted the phrase "depends on" in the above output
and exploded. So I updated our depends on fail case to be a bit more
conservative. This case doesn't catch the case where a line says "world depends
on", but that case is already caught by our other regexes so I didn't see it
necessary to catch that.

http://codereview.chromium.org/2853031/diff/1/2
File parallel_emerge (right):

http://codereview.chromium.org/2853031/diff/1/2#newcode51
parallel_emerge:51: print " ./parallel_emerge --board=BOARD --jobs=JOBS [emerge
args] package"
On 2010/07/07 19:40:26, Nick Sanders wrote:
> what does jobs default to in build_image.. "-1" I thought? Should we detect
> negative or zero and reset to MAX_INT or 256 or something? 

By default, build_image doesn't send the --jobs argument at all. But I added
detection for negative and zero cases anyway in case the user passes this in.

http://codereview.chromium.org/2853031/diff/1/2#newcode97
parallel_emerge:97: jobs_arg = int(arg.replace("--jobs=", ""))
On 2010/07/07 19:40:26, Nick Sanders wrote:
> should we check for intability here? maybe throw in a "try: int(...) except:
> error msg" so bad cmdline args have a friendlier failure mode..

Done.

http://codereview.chromium.org/2853031/diff/1/2#newcode98
parallel_emerge:98: elif arg[0] == "-" or arg == "y" or arg == "n":
On 2010/07/07 19:40:26, Nick Sanders wrote:
> maybe arg.startswith("-") for consistency instead of arg[0] == "-"

Done.

[Nick Sanders, 2010-07-09 00:06:12]

lgtm!