Robert Collins wrote:
On Fri, 2009-11-20 at 21:02 +0100, Francesco Chemolli wrote:
------------------------------------------------------------
revno: 10149
committer: Francesco Chemolli <[email protected]>
branch nick: trunk
timestamp: Fri 2009-11-20 21:02:00 +0100
message:
skip performing C libTrie unit tests
Please include motivation in commit messages. The diff shows that you
skipped the C tests, but not why.
And because the motivation is missing, I'm left asking 'why?' Untested
code is broken code, so this really can't be the right answer.
-Rob
For the record. From IRC (non-relevant discussion elided):
...
(04:03:06) kinkie: but it fails on trie
(04:06:31) kinkie: I'll probably have to copy to libTrie/configure.in
some checks from the main configure.in
...
(04:14:55) kinkie: that may be the reason.. maybe libtrie is not built
unless esi is enabled. And libtrie doesn't use libtool, thus no lib--aths
(04:15:08) amosjeffries: yes its not.
...
(04:18:56) kinkie: Now I'm trying to understand how to add library paths
(-L..) in a portable manner
(04:19:04) kinkie: solaris scatters libraries all over the place
(04:19:15) rousskov: info automake
(04:27:24) ***kinkie tries to libtoolize the easy way
(05:04:45) kinkie: *sigh*. Libtoolizing wasn't enough
(05:09:35) ***kinkie fails to understand where extra lib-paths are detected
(05:14:53) amosjeffries: $(LIBS) or $(LDFLAGS) ?
(05:16:24) kinkie: maybe.
(05:16:33) kinkie: but LDFLAGS seems to be explicitly overridden.
(05:21:54) kinkie: it's not, but it definitely doesn't pass through.
(05:22:08) amosjeffries: CXXLINK = $(CXXLD) $(AM_CXXFLAGS) $(CXXFLAGS)
$(AM_LDFLAGS) $(LDFLAGS)
(05:22:48) amosjeffries: where is it being added?
...
(05:24:34) kinkie: It's explicitly added only in libTrie.
(05:24:46) kinkie: worse.
(05:24:51) kinkie: Only in libTrie/test
(05:25:12) kinkie: let's try just removing it
(05:25:26) rousskov: which probably means it should not be
(05:25:30) amosjeffries: which lib?
(05:27:17) kinkie: libstdc++
(05:27:27) kinkie: and libm
(05:27:38) kinkie: libm I'm willing to believe in. stdc++ ? less likely.
(05:28:26) amosjeffries: I added -lm since it would not link without it.
might be better as a configure.in addition. stdc++ was their earlier.
...
(05:40:31) kinkie: Undefined first referenced
(05:40:31) kinkie: symbol in file
(05:40:31) kinkie: __1cG__CrunKpure_error6F_v_
../src/libTrie.a(Trie.o)
(05:40:31) kinkie: __1cG__CrunMex_rethrow_q6F_v_
../src/libTrie.a(Trie.o)
(05:40:31) kinkie: __1c2n6FI_pv_
../src/libTrie.a(Trie.o)
(05:40:32) kinkie: __1c2k6Fpv_v_
../src/libTrie.a(Trie.o)
(05:42:53) amosjeffries: I'm thinking we don't actually need the
C-interface in Squid-3 do we?
(05:43:08) kinkie: I'd guess not.
(05:45:41) amosjeffries: wait for lifeless agreement before comitting
anythign hat cuts it out. but the tests should be able to go further
with trie-c commented away.
Amos
--
Please be using
Current Stable Squid 2.7.STABLE7 or 3.0.STABLE20
Current Beta Squid 3.1.0.14
