[Python-bugs-list] [ python-Bugs-562585 ] build problems on DEC Unix 4.0f

[noreply@sourceforge.net]

Bugs item #562585, was opened at 2002-05-30 22:27
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=105470&aid=562585&group_id=5470

Category: Build
Group: Python 2.2.1
Status: Open
Resolution: None
Priority: 5
Submitted By: Garey Mills (gareytm)
Assigned to: Nobody/Anonymous (nobody)
Summary: build problems on DEC Unix 4.0f

Initial Comment:
Building with cc instead of gcc (as recommended) and with configure switch '--with-dec-threads' (also 
recommended. 'make test yields the following errors and messages:

test test_bsddb crashed -- bsddb.error: (22, 'Invalid argument')

test test_format produced unexpected output:
**********************************************************************
*** lines 2-3 of actual output doesn't appear in expected output after line 1:
+ '%#o' % 0 == '00' != '0'
+ u'%#o' % 0 == u'00' != '0'
**********************************************************************


2 tests failed:
    test_bsddb test_format
27 tests skipped:
    test_al test_audioop test_cd test_cl test_curses test_dl test_gdbm
    test_gl test_gzip test_imageop test_imgfile test_linuxaudiodev
    test_locale test_minidom test_nis test_ntpath test_pyexpat
    test_rgbimg test_sax test_socket_ssl test_socketserver
    test_sunaudiodev test_unicode_file test_winreg test_winsound
    test_zipfile test_zlib
Ask someone to teach regrtest.py about which tests are
expected to get skipped on osf1V4.


Are test failures important? Who do I ask "to teach regrtest.py about which tests are expected to get 
skipped on osf1V4"?

----------------------------------------------------------------------

>Comment By: Martin v. Löwis (loewis)
Date: 2002-06-01 23:05

Message:
Logged In: YES 
user_id=21627

Ok, I don't feel qualified to analyse this further from
remote. It seems that something on your system does not like
this file name; not sure whether this is the operating
system, bsddb, or something else. One would need to use
debugging techniques to get more information about the
nature of this problem. If you don't plan to use bsddb, you
can probably safely ignore this problem.

----------------------------------------------------------------------

Comment By: Garey Mills (gareytm)
Date: 2002-06-01 00:47

Message:
Logged In: YES 
user_id=555793

Here is the output of the test with the file name:

# ./python Lib/test/regrtest.py -v test_bsddb
test_bsddb

Testing:  BTree

fname:  /tmp/@18218.0
test test_bsddb crashed -- bsddb.error: (22, 'Invalid argument')
Traceback (most recent call last):
  File "Lib/test/regrtest.py", line 305, in runtest
    the_module = __import__(test, globals(), locals(), [])
  File "Lib/test/test_bsddb.py", line 79, in ?
    test(type[0], type[1])
  File "Lib/test/test_bsddb.py", line 21, in test
    f = openmethod(fname, 'c')
error: (22, 'Invalid argument')
1 test failed:
    test_bsddb

Garey

----------------------------------------------------------------------

Comment By: Martin v. Löwis (loewis)
Date: 2002-06-01 00:12

Message:
Logged In: YES 
user_id=21627

Can you please edit the fragment in test_bsddb.py that reads

    f = openmethod(fname, 'c')

to print fname before that line? Apparently, mktemp returns
a bad file name.

----------------------------------------------------------------------

Comment By: Garey Mills (gareytm)
Date: 2002-05-31 00:40

Message:
Logged In: YES 
user_id=555793

Here are the command and output

# ./python Lib/test/regrtest.py -v test_bsddb
test_bsddb

Testing:  BTree
test test_bsddb crashed -- bsddb.error: (22, 'Invalid argument')
Traceback (most recent call last):
  File "Lib/test/regrtest.py", line 305, in runtest
    the_module = __import__(test, globals(), locals(), [])
  File "./Lib/test/test_bsddb.py", line 76, in ?
    test(type[0], type[1])
  File "./Lib/test/test_bsddb.py", line 18, in test
    f = openmethod(fname, 'c')
error: (22, 'Invalid argument')
1 test failed:
    test_bsddb

----------------------------------------------------------------------

Comment By: Martin v. Löwis (loewis)
Date: 2002-05-31 00:18

Message:
Logged In: YES 
user_id=21627

the test_format bug is not important; it indicates a bug in
the system's C library.

For recording expected skipped tests, see Lib/regrtest.py.
Search for win32, and submit a patch that records the
expected skips. Alternatively, just don't worry about this.

For test_:bsddb, please run "python Lib/regrtest.py -v
test_bsddb" separately, and report the output.

----------------------------------------------------------------------

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=105470&aid=562585&group_id=5470