fix fuse impl_c a.k.a. blackmatch #811

New Issue

zooko · 2009-10-02T16:24:57Z

zooko commented

2009-10-02 16:24:57 +00:00

Thomas Delaet contributed a patch to make fuse impl_c a.k.a. blackmatch work:

http://allmydata.org/pipermail/tahoe-dev/2009-September/002923.html

I've downloaded his patch from that email and attached it to this ticket. Please review and apply!

Thomas Delaet contributed a patch to make fuse impl_c a.k.a. blackmatch work: <http://allmydata.org/pipermail/tahoe-dev/2009-September/002923.html> I've downloaded his patch from that email and attached it to this ticket. Please review and apply!

zooko added the

labels 2009-10-02 16:24:57 +00:00

zooko added this to the undecided milestone 2009-10-02 16:24:57 +00:00

zooko commented

2009-10-02 16:25:13 +00:00

Attachment fix_blackmatch.patch.txt (2047 bytes) added

**Attachment** fix_blackmatch.patch.txt (2047 bytes) added

fix_blackmatch.patch.txt

2.0 KiB

francois commented

2009-12-29 23:58:47 +00:00

I've just tested this patch on latest trunk and it really does make the blackmatch implementation work again, so it should be included in the next Tahoe release.

The initial breakage was due to this patch:

Sat Jul 18 00:15:49 CEST 2009  Brian Warner <warner@lothar.com>
  * rename NewDirectoryNode to DirectoryNode, NewDirectoryURI to DirectoryURI

The test in the parsing of the '--root-uri' option is commented by the initial patch and I don't really see why because the test worked during my tests. What do you think about keeping it uncommented ? I'll attach a new patch with this modification.

        # test the uri for structural validity:
        try:
            DirectoryURI.init_from_string(root_uri)
        except:
            raise usage.error('root-uri must be a valid directory uri (not %r)' % (root_uri,))

I've just tested this patch on latest trunk and it really does make the blackmatch implementation work again, so it should be included in the next Tahoe release. The initial breakage was due to this patch: ``` Sat Jul 18 00:15:49 CEST 2009 Brian Warner <warner@lothar.com> * rename NewDirectoryNode to DirectoryNode, NewDirectoryURI to DirectoryURI ``` The test in the parsing of the '--root-uri' option is commented by the initial patch and I don't really see why because the test worked during my tests. What do you think about keeping it uncommented ? I'll attach a new patch with this modification. ``` # test the uri for structural validity: try: DirectoryURI.init_from_string(root_uri) except: raise usage.error('root-uri must be a valid directory uri (not %r)' % (root_uri,)) ```

tahoe-lafs modified the milestone from undecided to 1.6.0

2009-12-29 23:58:47 +00:00

francois commented

2009-12-29 23:59:02 +00:00

Attachment fix_blackmatch-2.patch.txt (1792 bytes) added

**Attachment** fix_blackmatch-2.patch.txt (1792 bytes) added

fix_blackmatch-2.patch.txt

1.8 KiB

warner commented

2009-12-30 04:08:10 +00:00

keeping the DirectoryURI.init_from_string(root_uri) test is probably ok. I'd like to reduce the use of things like DirectoryURI in the future, but it's probably ok for the near-term.

Sorry about breaking that.. I grepped through src/ for users of those symbols, but I didn't think to grep through misc/ too.. I'd assumed that everything there was speaking webapi rather than importing tahoe code directly.

keeping the `DirectoryURI.init_from_string(root_uri)` test is probably ok. I'd like to reduce the use of things like `DirectoryURI` in the future, but it's probably ok for the near-term. Sorry about breaking that.. I grepped through src/ for users of those symbols, but I didn't think to grep through misc/ too.. I'd assumed that everything there was speaking webapi rather than importing tahoe code directly.

zooko commented

2010-01-09 20:34:35 +00:00

fixed by changeset:eb9358ff6de5cd23

zooko added the

fixed

label 2010-01-09 20:34:35 +00:00

zooko closed this issue

2010-01-09 20:34:35 +00:00

Sign in to join this conversation.