On Wed, Aug 27, 2014 at 6:58 PM, Ivan Zhakov <ivan_at_visualsvn.com> wrote:
> On 27 August 2014 19:41, Stefan Fuhrmann <stefan.fuhrmann_at_wandisco.com>
> wrote:
> > On Wed, Aug 27, 2014 at 9:13 AM, Ivan Zhakov <ivan_at_visualsvn.com> wrote:
> >>
> >> On 26 August 2014 21:18, Branko Čibej <brane_at_wandisco.com> wrote:
> >> > On 26.08.2014 18:46, Ivan Zhakov wrote:
> >> >
> >> > On 18 August 2014 19:50, <stefan2_at_apache.org> wrote:
> >> >
> >> > Author: stefan2
> >> > Date: Mon Aug 18 15:50:44 2014
> >> > New Revision: 1618641
> >> >
> >> > URL: http://svn.apache.org/r1618641
> >> > Log:
> >> > Within FSFS, replace the use of svn__strtol with a variant that
> >> > provides overflow detection. FSFS needs a locale-independent
> >> > and preferrably quick number parser.
> >> >
> >> > [...]
> >> >
> >> > Stefan,
> >> >
> >> > It maybe worth to make tests for this function (making it private to
> >> > libsvn_subr for this). What do you think?
> >> >
> >> >
> >> > Currently, this function is only used in a couple places within FSFS.
> >> > FWIW
> >> > this does not prevent writing tests for it; we do already have C tests
> >> > for
> >> > libsvn_fs_fs.
> >> >
> >> > I did raise the same question in Sheffield, but was convinced that
> >> > moving
> >> > that code around for formality's sake serves no useful purpose.
> >> >
> >> I'm fine to keep libsvn_fs_fs if we could write test for this
> >> function, but I think that test will be very useful for this function.
> >
> >
> > Test added in r1620913.
> >
> Is it true that you uncovered the bug in svn_fs_fs__id_parse() fixed
> in r1620912 using this new test?
>
Not in that sense. While verifying that the new test covers
all overflow detection paths, I saw and opportunity to make
the parser more strict that it has ever been in previous
releases. That has kind of been the point of making all
ID elements proper numbers instead of converting them
from and to arbitrary strings.
-- Stefan^2.
Received on 2014-08-28 00:49:15 CEST