On 02/19/2010 06:36 PM, Randy McMurchy wrote:
> Hi all,
>
> Not sure if this is a -dev issue or what but I recently built an LFS-6.5
> machine. Everything smooth as silk, other than 3 tests in the gcc suite
> and 4 in the mudflap suite. Everything else perfect.
>
> I'm encountering errors in recent packages I'm building. For example in
> Python I get:
>
> =================================================================
> 327 tests OK.
> 2 tests failed:
> test_distutils test_urllib2_localnet
> 36 tests skipped:
> test_aepack test_al test_applesingle test_bsddb185 test_bsddb3
> test_cd test_cl test_codecmaps_cn test_codecmaps_hk
> test_codecmaps_jp test_codecmaps_kr test_codecmaps_tw test_curses
> test_gl test_imgfile test_kqueue test_linuxaudiodev test_macos
> test_macostools test_normalization test_ossaudiodev test_pep277
> test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
> test_startfile test_sunaudiodev test_tcl test_timeout
> test_unicode_file test_urllib2net test_urllibnet test_winreg
> test_winsound test_zipfile64
> 1 skip unexpected on linux2:
> test_tcl
> make: *** [test] Error 1
> =================================================================
>
########## BEGINING TESTSUITE ##########
running build
running build_ext
INFO: Can't locate Tcl/Tk libs and/or headers
^[[?1034h
Failed to find the necessary bits to build these modules:
_bsddb _sqlite3 _tkinter
bsddb185 sunaudiodev
To find the necessary bits, look in setup.py in detect_modules() for the
module's name.
I don't have an explanation for tcl or bdb.
...
326 tests OK.
2 tests failed:
test_distutils test_httpservers
37 tests skipped:
test_aepack test_al test_applesingle test_bsddb test_bsddb185
test_bsddb3 test_cd test_cl test_codecmaps_cn test_codecmaps_hk
test_codecmaps_jp test_codecmaps_kr test_codecmaps_tw test_curses
test_gl test_imgfile test_kqueue test_linuxaudiodev test_macos
test_macostools test_normalization test_ossaudiodev test_pep277
test_py3kwarn test_scriptpackages test_smtpnet test_socketserver
test_sqlite test_startfile test_sunaudiodev test_tcl test_timeout
test_urllib2net test_urllibnet test_winreg test_winsound
test_zipfile64
2 skips unexpected on linux2:
test_tcl test_bsddb
make: *** [test] Error 1
Looking at the specific errors....
{{{
test_distutils
/usr/bin/ld: cannot find -lpython2.6
collect2: ld returned 1 exit status
test test_distutils failed -- Traceback (most recent call last):
File "/sources/Python-2.6.4/Lib/distutils/tests/test_build_ext.py",
line 258, in test_get_outputs
cmd.run()
File "/sources/Python-2.6.4/Lib/distutils/command/build_ext.py", line
340, in run
self.build_extensions()
File "/sources/Python-2.6.4/Lib/distutils/command/build_ext.py", line
449, in build_extensions
self.build_extension(ext)
File "/sources/Python-2.6.4/Lib/distutils/command/build_ext.py", line
531, in build_extension
target_lang=language)
File "/sources/Python-2.6.4/Lib/distutils/ccompiler.py", line 845, in
link_shared_object
extra_preargs, extra_postargs, build_temp, target_lang)
File "/sources/Python-2.6.4/Lib/distutils/unixccompiler.py", line 258,
in link
raise LinkError, msg
LinkError: command 'gcc' failed with exit status 1
}}}
Couldn't find "-lpython2.6" -- Bad linker path, test is broken.
{{{
test_httpservers
Traceback (most recent call last):
File "/sources/Python-2.6.4/Lib/CGIHTTPServer.py", line 255, in run_cgi
os.execve(scriptfile, args, os.environ)
OSError: [Errno 13] Permission denied
Traceback (most recent call last):
File "/sources/Python-2.6.4/Lib/CGIHTTPServer.py", line 255, in run_cgi
os.execve(scriptfile, args, os.environ)
OSError: [Errno 13] Permission denied
Traceback (most recent call last):
File "/sources/Python-2.6.4/Lib/CGIHTTPServer.py", line 255, in run_cgi
os.execve(scriptfile, args, os.environ)
OSError: [Errno 13] Permission denied
test test_httpservers failed -- errors occurred; run in verbose mode for
details
}}}
Running as unprivileged user? Didn't bother rebuilding Python to test
as root.
> Then on libxml2 I get:
>
> =================================================================
> an 2262 tests, 15 errors, 0 leaks
> Total 2262 tests, 15 errors, 0 leaks
> See runxmlconf.log for detailed output
> 15 errors were expected
> make[1]: Entering directory `/home/rml/build/libxml2-2.7.4/python'
> Making all in .
> Making all in tests
> ## running Python regression tests
> -- serialize.py
> error serializing HTML document 4
> make[1]: Leaving directory `/home/rml/build/libxml2-2.7.4/python'
> ==================================================================
Test cases: 3 tests
Ran 2262 tests, 15 errors, 0 leaks
Total 2262 tests, 15 errors, 0 leaks
See runxmlconf.log for detailed output
15 errors were expected
make[1]: Entering directory `/sources/libxml2-2.7.6/python'
Making all in .
Making all in tests
## running Python regression tests
-- serialize.py
error serializing HTML document 4
make[1]: Leaving directory `/sources/libxml2-2.7.6/python'
The testsuite is broken. Going backwards, I found that 3 of the
serialize tests are broken since 2.7.4. Fist one seen is because the
meta data isn't case changed for the ISO (change the expected result
from iso-8859-1 to ISO-8859-1 to 'fix' the test). This happens when the
first (ISO-8859-1) is passed. If you omit test 2, the current expected
result is correct for test 4. I wrote out a nice testcase to show that
the difference was due to changes in libxml2, but deleted it as I
figured I could explain better with text. I wish I would have kept it
as it still displayed the difference between the two versions. Somebody
smarter than I should investigate further.
Next error is HTML root 1, best guess is just get rid of the test
completely. Since 2.7.4, even with meta None, Charset is
unconditionally included. Interestingly enough, if you fix the expected
result, the result changes. I think it's only in the comparison,
however put in a 'print str' before the exit, and see the result on
screen. Modify the expected result (it should match identically the 2nd
HTML root test's expected result), and leave your print statement in
there and you get garbled text for charset. I don't get it, but I
*think* that the test is broken because of the last charset change of
htmlSetMetaEncoding(). (diff -Naurp the two version's trees). Again, I
don't know, if somebody wants to investigate further, great.
The 3rd error is the same as the first in HTML root 4.
At anyrate, the tests haven't kept up with the changes...or the devs
actually broke the Python module and didn't care about testing. IDK.
2.7.3 passes the tests. I'm too tired to continue for tonight and go
asking the devs questions about stuff I don't fully understand.
>
> Is anyone else seeing any of this? At this point I really don't want
> to continue, I've never had issues with either of these packages.
>
-- DJ
--
This message has been scanned for viruses and
dangerous content, and is believed to be clean.
--
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
