Test suite fails on x86_64 OS X build
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| subvertpy |
Fix Released
|
High
|
Jelmer Vernooij | ||
| subvertpy (Debian) |
Fix Released
|
Unknown
|
|||
| subvertpy (Ubuntu) |
Fix Released
|
High
|
Unassigned | ||
Bug Description
When building subvertpy 0.8.0 on Mac OS X x86_64 with any of 2.5, 2.6 or 2.7 pythons, I get the following test failure:
=======
FAIL: test_transmit_
-------
Traceback (most recent call last):
File "subvertpy/
AssertionError: 1 != 4294967297
-------
Ran 154 tests in 29.462s
FAILED (failures=1)
The same test passes on a i386 build. It looks like a variable length mismatch somewhere; something is expecting a 32 bit value but gets a 64 bit one instead. That's a guess, since 4294967297 = 0x100000001 which is a 64 bit value and looks suspiciously like it's supposed to be 0x00000001, which is 32 bit.
Related branches
| Changed in subvertpy: | |
| status: | New → Triaged |
| status: | Triaged → Incomplete |
| Daniel Johnson (daniel-daniel-johnson) wrote | #2 |
| Jelmer Vernooij (jelmer) wrote | #3 |
| Daniel Johnson (daniel-daniel-johnson) wrote | #4 |
| Launchpad Janitor (janitor) wrote | #5 |
| Changed in subvertpy: | |
| status: | Incomplete → Expired |
| Changed in subvertpy: | |
| status: | Expired → Triaged |
| importance: | Undecided → Medium |
| Changed in subvertpy (Ubuntu): | |
| status: | New → Triaged |
| importance: | Undecided → Medium |
| Changed in subvertpy: | |
| importance: | Medium → High |
| Changed in subvertpy (Ubuntu): | |
| importance: | Medium → High |
| Changed in subvertpy: | |
| status: | Triaged → In Progress |
| assignee: | nobody → Jelmer Vernooij (jelmer) |
| Changed in subvertpy: | |
| status: | In Progress → Fix Committed |
| Daniel Johnson (daniel-daniel-johnson) wrote | #6 |
| Launchpad Janitor (janitor) wrote | #7 |
| Changed in subvertpy (Ubuntu): | |
| status: | Triaged → Fix Released |
| Changed in subvertpy: | |
| status: | Fix Committed → Fix Released |
| status: | Fix Released → Fix Committed |
| milestone: | none → 0.8.3 |
| Changed in subvertpy: | |
| status: | Fix Committed → Fix Released |
| Changed in subvertpy (Debian): | |
| status: | Unknown → Fix Released |
svn_revnum_t should be a long according to /usr/include/
Can you check the size of svn_revnum_t and see if it matches that of long?