On Wed, May 13, 2009 at 05:19:58PM -0500, Shawn Walker wrote:
> http://cr.opensolaris.org/~swalker/pkg-6342/
pkgrecv.1.txt
- line 51, 52: I think "evaluates" and "list" are probably too low-level
for this document. Maybe just "recursively retrieve"?
- line 54, 55: "... from which to receive ..."
pull.py:
- line 167: why CachedManifest? Is this to speed up
get_hashes_and_sizes()?
- line 306, 313: do you really need both?
- line 322: the way you're using this, you could use a set.
- line 390: you could put these tests inside the try statement, save a
few lines.
- line 393: isn't the second argument to copy() just "dest"?
- line 397: I think it'd be useful to have a comment here saying that we
don't bother cleaning up a partially copied file because it'll be
overwritten on the next attempt. Though I think it'd be better to
clean up partially copied files because someone might not decide to
retry, and then their repo wouldn't just be incomplete, it'd be broken.
- line 420, 424: spaces around equals.
- line 452: does this really wrap no better? Also, couldn't you just put
this in abort()?
- line 489, 490: getting a repo's catalog seems like a candidate for some
part of the API, so that you can call it regardless of the URI scheme.
- line 492, 493: why not abort(err=e)?
- line 521: use the setdefault() trick like above
- line 526: spaces around equals
- line 605: shouldn't it be possible to republish to a file:// repo
without having to decompress and recompress?
- line 622: put the first close-paren on the previous line
- line 701, 702: rewrap
Danek
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
