You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Karl Heinz Marbaise <kh...@gmx.de> on 2009/03/16 17:38:02 UTC
CHANGEs File for Subversion 1.6 - Questions - #3349
Hi,
I found the following:
In theory, BDB backend could allow `changes' table out-of-sync with
actual changes (#3349)
A comment in the #3349 issue:
-------------
NOTE: The API usage demonstrated in the regression test is not valid
per the instructions on the APIs given in svn_fs.h, which say that you
must send windows including the NULL terminating window through the
returned handlers before making other changes to the transaction. FSFS
seems to enforce this, but BDB does not. In BDB, this code path will
result in node revisions that still have dangling
representations-in-progress, but that was true before my changes, too.
-------------
Based on the diff
svn diff -r 34896:34897 http://svn.collab.net/repos/svn
i don't see a real change in an API ?
Or did i misunderstand a thing ?
Kind regards
Karl Heinz Marbaise
--
SoftwareEntwicklung Beratung Schulung Tel.: +49 (0) 2405 / 415 893
Dipl.Ing.(FH) Karl Heinz Marbaise ICQ#: 135949029
Hauptstrasse 177 USt.IdNr: DE191347579
52146 Würselen http://www.soebes.de
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1334489
Re: CHANGEs File for Subversion 1.6 - Questions - #3349
Posted by "C. Michael Pilato" <cm...@collab.net>.
Karl Heinz Marbaise wrote:
> Hi,
>> Nope, no API change. Just a bugfix to a codepath folks aren't supposed to
>> be executing anyway.
> So it will be part of the developer-visible changes as a simple fixed
> entry...
>
> * fixed: In theory, BDB backend could allow `changes' table
> out-of-sync with actual changes (#3349)
>
> Might there be a better text for this ?
Maybe:
fixed: BDB `changes' table inconsistency when APIs are misused (#3349)
?
--
C. Michael Pilato <cm...@collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1334590
Re: CHANGEs File for Subversion 1.6 - Questions - #3349
Posted by Karl Heinz Marbaise <kh...@gmx.de>.
Hi,
> Nope, no API change. Just a bugfix to a codepath folks aren't supposed to
> be executing anyway.
So it will be part of the developer-visible changes as a simple fixed
entry...
* fixed: In theory, BDB backend could allow `changes' table
out-of-sync with actual changes (#3349)
Might there be a better text for this ?
Kind regards
Karl Heinz Marbaise
--
SoftwareEntwicklung Beratung Schulung Tel.: +49 (0) 2405 / 415 893
Dipl.Ing.(FH) Karl Heinz Marbaise ICQ#: 135949029
Hauptstrasse 177 USt.IdNr: DE191347579
52146 Würselen http://www.soebes.de
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1334537
Re: CHANGEs File for Subversion 1.6 - Questions - #3349
Posted by "C. Michael Pilato" <cm...@collab.net>.
Nope, no API change. Just a bugfix to a codepath folks aren't supposed to
be executing anyway.
Karl Heinz Marbaise wrote:
> Hi,
>
> I found the following:
>
> In theory, BDB backend could allow `changes' table out-of-sync with
> actual changes (#3349)
>
> A comment in the #3349 issue:
> -------------
> NOTE: The API usage demonstrated in the regression test is not valid
> per the instructions on the APIs given in svn_fs.h, which say that you
> must send windows including the NULL terminating window through the
> returned handlers before making other changes to the transaction. FSFS
> seems to enforce this, but BDB does not. In BDB, this code path will
> result in node revisions that still have dangling
> representations-in-progress, but that was true before my changes, too.
> -------------
>
> Based on the diff
> svn diff -r 34896:34897 http://svn.collab.net/repos/svn
> i don't see a real change in an API ?
>
> Or did i misunderstand a thing ?
>
> Kind regards
> Karl Heinz Marbaise
--
C. Michael Pilato <cm...@collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1334503