On 10.12.2013 04:40, Evgeny Kotkov wrote:
> Hi,
>
> I've recently discovered that the new code that creates the svn_fs_id_t
> objects
> does something quite odd. There is a discrepancy in how functions from
> subversion/libsvn_fs_fs/id.c initialize the GENERIC_ID field of the
> 'fs_fs__id_t' stru
On 10.12.2013 04:34, Evgeny Kotkov wrote:
> Hi,
>
> I've noticed that the svn_fs_fs__get_changes function currently invokes
> undefined behavior when there is no CHANGES_CACHE present.
Committed in r1549766, thanks!
-- Brane
--
Branko Čibej | Director of Subversion
WANdisco // Non-Stop Data
e.
Hi,
I've recently discovered that the new code that creates the svn_fs_id_t objects
does something quite odd. There is a discrepancy in how functions from
subversion/libsvn_fs_fs/id.c initialize the GENERIC_ID field of the
'fs_fs__id_t' structs. For example, if you compare svn_fs_fs__id_deserial
Hi,
I've noticed that the svn_fs_fs__get_changes function currently invokes
undefined behavior when there is no CHANGES_CACHE present. According to the
fs_fs_data_t contract, this is a perfectly valid situation (see
subversion\libsvn_fs_fs\fs.h around trunk@1549686):
[[[
/* Private (non-shared
On 12/9/13 1:56 PM, Ivan Zhakov wrote:
> I was worried a but that we have to care that mergeinfo --log will
> never have include_merged_revisions option.
I'm not sure there really is a problem with supporting include_merged_revisions
for mergeinfo --log. For instance even with our (Subversion pro
On 5 December 2013 00:15, Stefan Fuhrmann wrote:
> Hi all,
>
> I've spent the last two days getting our test suite to pass
> with our backward compat option. That mostly worked for
> 1.5 through 1.9 (with one exception).
>
> The problem is however, that we not only create repositories
> in the res
On 6 December 2013 02:09, Ben Reser wrote:
> On 12/5/13 12:30 PM, Ivan Zhakov wrote:
>> On 5 December 2013 06:20, wrote:
>>> Author: breser
>>> Date: Thu Dec 5 02:20:41 2013
>>> New Revision: 1547995
>>>
>>> URL: http://svn.apache.org/r1547995
>>> Log:
>>> Make mergeinfo and log commands share
Hi,
I'm pleased to announce the release of serf 1.3.3.
This is a small patch release containing a fix to solve a problem
connecting to multi-homed servers (e.g. ipv4/ipv6) and some
improvements in the use of error codes during ssl certificate
validation and handling of timed out connections.
Yo
Martin Furter writes:
> I could also argue that it is inconsistent as long as you do not write
> *c as *(c+0).
>
> But this is irrelevant. As you found out the log message is wrong. So
> here is a honest one:
>
> [[[
> What led to at least one bug will probably lead to more bugs so remove
> the u
Wrong button :/ Forwarding to the list too...
Original Message
Subject: Re: [PATCH] fix for programmer error in path split text logic
Date: Mon, 09 Dec 2013 21:30:54 +0530
From: Martin Furter
To: Philip Martin
On 12/09/13 17:13, Philip Martin wrote:
Martin Furter writes:
Martin Furter writes:
> On 12/09/13 15:27, Philip Martin wrote:
>> Martin Furter writes:
>>
>>> Wouldn't last_dot[1] be more readable than (*(last_dot + 1) ?
>>
>> Probably. I approve a patch if you want to commit.
>
> A quick grep shows 21 of those constructs in the following files:
>
> ./subve
On 12/09/13 15:27, Philip Martin wrote:
Martin Furter writes:
Wouldn't last_dot[1] be more readable than (*(last_dot + 1) ?
Probably. I approve a patch if you want to commit.
A quick grep shows 21 of those constructs in the following files:
./subversion/libsvn_subr/path.c
./subversion/lib
On 09.12.2013 11:01, Branko Čibej wrote:
> Hi Stefan,
>
> could you please shed some light on your "minor opimization" commit:
>
>>
>> r1532186 | stefan2 | 2013-10-15 06:59:00 +0200 (Tue, 15 Oct 2013) | 5
>> lines Minor optimi
Hi Stefan,
could you please shed some light on your "minor opimization" commit:
>
> r1532186 | stefan2 | 2013-10-15 06:59:00 +0200 (Tue, 15 Oct 2013) | 5
> lines Minor optimization in svn_membuf code. *
> subversion/libsvn_s
Martin Furter writes:
> Wouldn't last_dot[1] be more readable than (*(last_dot + 1) ?
Probably. I approve a patch if you want to commit.
--
Philip Martin | Subversion Committer
WANdisco // *Non-Stop Data*
15 matches
Mail list logo
