Re: Request release of new LSBdev packages for Etch
On Thu, 8 Mar 2007, Steve Langasek wrote:
lsb-build-base3_3.1.1-3
-----------------------
When viewing the debdiff, there are some non-trivial looking changes in
some perl scripts. These scripts are not part of the binary packages, and
are not even used at build time (they are used by the upstream to generate
many of the file in this package), so I hope they do not become sticking
points when evaluating this package.
Which bits, precisely, are ignorable as no-ops?
The diffs to these script can be ignored as they are not used during the
build, or included in the package.
lsb-build-base3-3.1.1/build_env/stub_libs/check_lsb_symbols
lsb-build-base3-3.1.1/build_env/stub_libs/mkstublibs
lsb-build-base3-3.1.1/build_env/headers/getheaderlist
lsb-build-base3-3.1.1/build_env/headers/mkdatadef
lsb-build-base3-3.1.1/build_env/headers/mkfuncprototype
lsb-build-base3-3.1.1/build_env/headers/mkheader
That should leave just the headers, and a few .c files under the
stub_libs directory.
There are quite a few whitespace changes in the headers, which are obviously
no-ops but nevertheless make reviewing the diff awkward.
Yes, that is unfortunate noise, but comes from upstream.
I could maybe see unblocking this just on the grounds that this is a
standards implementation, and shipping what matches upstream is important
even if it introduces practical regressions for building against...
Most of the changes are additions to what was there, only few are actually
corrections that are needed to make things build. The chance of a
regression should be small, and in fact, the chances that something now
builds that didn't use to build is improved.
Stuart
Stuart R. Anderson anderson@netsweng.com
Network & Software Engineering http://www.netsweng.com/
1024D/37A79149: 0791 D3B8 9A4C 2CDC A31F
BD03 0A62 E534 37A7 9149
Reply to: