On Thu, 2010-07-08 at 15:23 +0200, Stefan Sperling wrote:
> On Thu, Jul 08, 2010 at 02:08:15PM +0200, Stefan Sperling wrote:
> > On Thu, Jul 08, 2010 at 12:39:22PM +0100, Julian Foad wrote:
> > > I've only skimmed through it so far, and probably can't do a full review
> > > before the end of the mo
On Thu, Jul 08, 2010 at 02:08:15PM +0200, Stefan Sperling wrote:
> On Thu, Jul 08, 2010 at 12:39:22PM +0100, Julian Foad wrote:
> > I've only skimmed through it so far, and probably can't do a full review
> > before the end of the month. However from what I've seen it looks good
> > enough to go a
On Thu, Jul 08, 2010 at 12:39:22PM +0100, Julian Foad wrote:
> (Oops, I forgot to trim 60 KB of quoted text in my previous reply :-( )
>
>
> Would you mind writing a test. The thing that most needs testing, I
> should think, is interleaving read(), readline(), mark() and seek().
> Here's an idea
(Oops, I forgot to trim 60 KB of quoted text in my previous reply :-( )
Would you mind writing a test. The thing that most needs testing, I
should think, is interleaving read(), readline(), mark() and seek().
Here's an idea for a test:
Stream content = "Line one.\nLine two.\n"
# Some point
On Thu, Jul 08, 2010 at 09:52:51AM +0100, Julian Foad wrote:
> Hi Stefan. I've just started looking at this. Overall impression is it
> looks like you've done it very well - thank you!
>
> Could you possibly get the readline_detect_eol() API changes out of this
> patch, and make them in a separa
This change was proposed by Julian some time ago, because having
callback types that change the behaviour of svn_stream_readline(),
namely svn_io_line_filter_cb_t and svn_io_line_transformer_cb_t,
was the wrong approach.
See http://subversion.tigris.org/issues/show_bug.cgi?id=3555
The patch below
6 matches
Mail list logo
