Alexander Hansen wrote: > On 10/19/07, Alexander K. Hansen <[EMAIL PROTECTED]> wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> David Reiser wrote: >>> I'm trying to build libofx3 against curl4 (obsolete runaround occurs >>> trying to use curl3). But fink complains about test failures building >>> libxml2 because it can't find symbols expected in /usr/lib/ >>> libxml2.2.dylib. The first appearance of /usr/lib is: >>> >>> ## Error cases stream regression tests >>> cdata.xml result >>> 1,5d0 < ./test/errors/cdata.xml:2: parser error : Input is not proper >>> UTF-8, indicate encoding ! < Bytes: 0x5B 0x43 0xE1 0x72 < <A><![CDATA >>> [C?r]]></A> < ^ < ./test/errors/cdata.xml : failed to parse >>> ## Running the API regression tests this may take a little while >>> dyld: lazy symbol binding failed: Symbol not found: _htmlNewParserCtxt >>> Referenced from: /sw/src/fink.build/libxml2-2.6.27-1001/ >>> libxml2-2.6.27/./testapi >>> Expected in: /usr/lib/libxml2.2.dylib >>> >>> dyld: Symbol not found: _htmlNewParserCtxt >>> Referenced from: /sw/src/fink.build/libxml2-2.6.27-1001/ >>> libxml2-2.6.27/./testapi >>> Expected in: /usr/lib/libxml2.2.dylib >>> >>> after a bunch of other instances, fink barfs with: >>> >>> ## Threaded regression tests >>> ## generating dba100000.xml >>> ## Timing tests to try to detect performance >>> ## as well a memory usage breakage when streaming >>> ## 1/ using the file interface >>> ## 2/ using the memory interface >>> ## 3/ repeated DOM parsing >>> ## 4/ repeated DOM validation >>> Parsing took 831 ms >>> Parsing took 839 ms >>> 100 iterations took 1142 ms >>> 100 iterations took 2574 ms >>> ## Module tests >>> dyld: lazy symbol binding failed: Symbol not found: _xmlModuleOpen >>> Referenced from: /sw/src/fink.build/libxml2-2.6.27-1001/ >>> libxml2-2.6.27/./testModule >>> Expected in: /usr/lib/libxml2.2.dylib >>> >>> dyld: Symbol not found: _xmlModuleOpen >>> Referenced from: /sw/src/fink.build/libxml2-2.6.27-1001/ >>> libxml2-2.6.27/./testModule >>> Expected in: /usr/lib/libxml2.2.dylib >>> >>> make[2]: *** [ModuleTests] Error 133 >>> make[1]: *** [check-am] Error 2 >>> make: *** [check-recursive] Error 1 >>> ### execution of make failed, exit code 2 >>> ### execution of /var/tmp/tmp.40.qfHqoB failed, exit code 2 >>> Removing runtime build-lock... >>> Removing build-lock package... >>> /sw/bin/dpkg-lockwait -r fink-buildlock-libxml2-2.6.27-1001 >>> (Reading database ... 4208 files and directories currently installed.) >>> Removing fink-buildlock-libxml2-2.6.27-1001 ... >>> Failed: phase test: error (2) >>> >>> >>> I'm building this on an almost pristine /sw fink install. >>> >>> Anyone know why the test modules are looking in /usr/lib? >>> >>> Dave >>> -- >>> David Reiser >>> [EMAIL PROTECTED] >>> >>> >> Confirmed on my own setup. >> >> I might guess that the issue is that the package predates fink running >> the test suites as part of maintainer mode builds, and so they weren't >> tested, but I'm not sure about that, >> >> > D.Macks had an idea to check; but I'm trying a bootstrap on 10.3 to > check Fink upgrade scenarios for D.R.M. If you've got .debs for them, > try installing all 3 libxml packages and then do a rebuild to see if > the test suites are happy. > Based on similar behavior (and cause?) in gnucash2, I think make check interacts undesirably with DYLD_LIBRARY_PATH. D.Macks added a make check line in libxml2 in august, my deb in another of my trees was built in February just after the last upstream revup and I didn't have a problem then. But that could have been the rebuild-is-not-a-problem result.
-- David Reiser [EMAIL PROTECTED] ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Fink-devel mailing list Fink-devel@lists.sourceforge.net http://news.gmane.org/gmane.os.apple.fink.devel