Re: svn commit: r1382204 - /subversion/trunk/subversion/libsvn_subr/utf_validate.c

2012-09-07 Thread Branko Čibej
On 08.09.2012 03:59, Peter Samuelson wrote: >>for (; len > 0; ++data, --len) >> - if (*data < 0) >> -return data; >> +if (*data < 0 || *data >= 0x80) >> + return data; > A reasonable compiler will collapse it anyway, but this is shorter and > more direc

Re: svn commit: r1382204 - /subversion/trunk/subversion/libsvn_subr/utf_validate.c

2012-09-07 Thread Peter Samuelson
>for (; len > 0; ++data, --len) > - if (*data < 0) > -return data; > +if (*data < 0 || *data >= 0x80) > + return data; A reasonable compiler will collapse it anyway, but this is shorter and more direct: if (*data & 0x80) Peter

Re: svn commit: r1381808 - in /subversion/trunk/subversion/libsvn_fs_fs: fs_fs.c fs_fs.h

2012-09-07 Thread Stefan Fuhrmann
On Fri, Sep 7, 2012 at 9:36 PM, Daniel Shahaf wrote: > I think this fix is wrong. svn_fs_fs__read_noderev() should return the > parsed noderev as is --- if it's a revision file and the noderev > contains an is-fresh-txn-root field, then it needs to reflect that. > The problem is that the whole

Re: Consolidating private/svn_subr_private.h

2012-09-07 Thread Stefan Fuhrmann
On Tue, Sep 4, 2012 at 10:03 PM, Stefan Fuhrmann < stefan.fuhrm...@wandisco.com> wrote: > Hi there, > > While looking for the appropriate place to declare a few > svn-private constants, I realized that libsvn_subr is the > only lib that comes with multiple private headers. > > So, I prepared a pat

Re: svn commit: r1381766 - /subversion/trunk/subversion/libsvn_subr/utf_validate.c

2012-09-07 Thread Stefan Fuhrmann
On Fri, Sep 7, 2012 at 7:19 PM, Peter Samuelson wrote: > > [stef...@apache.org] > > + for (; len > 0; ++data, --len) > > + if (*data < 0) > > +return data; > > Two major problems: > > 1) It doesn't account for character sets like Shift-JIS, where state >change is sig

Re: svn commit: r1381766 - /subversion/trunk/subversion/libsvn_subr/utf_validate.c

2012-09-07 Thread Stefan Fuhrmann
On Fri, Sep 7, 2012 at 6:18 PM, Branko Čibej wrote: > On 06.09.2012 23:30, stef...@apache.org wrote: > > Author: stefan2 > > Date: Thu Sep 6 21:30:40 2012 > > New Revision: 1381766 > > > > URL: http://svn.apache.org/viewvc?rev=1381766&view=rev > > Log: > > Many strings we need to convert to UTF-

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Paul Burba
On Fri, Sep 7, 2012 at 6:46 PM, Stefan Fuhrmann wrote: > On Fri, Sep 7, 2012 at 9:18 PM, Julian Foad > wrote: >> >> Paul Burba wrote: >> >> > On Thu, Sep 6, 2012 at 7:32 PM, wrote: >> >> URL: http://svn.apache.org/viewvc?rev=1381800&view=rev >> >> Log: >> >> Re-implement svn_io_read_length_l

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Stefan Fuhrmann
One minute after my fix ;) On Sat, Sep 8, 2012 at 12:35 AM, Paul Burba wrote: > On Fri, Sep 7, 2012 at 3:29 PM, Julian Foad > wrote: > > (Just forwarding this thread to Stefan.) > > > > > > - Julian > > > > > > I (Julian Foad) wrote: > > > >> Paul Burba wrote: > >> > >>> On Thu, Sep 6, 2012 at

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Stefan Fuhrmann
On Fri, Sep 7, 2012 at 9:18 PM, Julian Foad wrote: > Paul Burba wrote: > > > On Thu, Sep 6, 2012 at 7:32 PM, wrote: > >> URL: http://svn.apache.org/viewvc?rev=1381800&view=rev > >> Log: > >> Re-implement svn_io_read_length_line as this is one of the > >> most-called functions in SVN. Instead

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Paul Burba
On Fri, Sep 7, 2012 at 3:29 PM, Julian Foad wrote: > (Just forwarding this thread to Stefan.) > > > - Julian > > > I (Julian Foad) wrote: > >> Paul Burba wrote: >> >>> On Thu, Sep 6, 2012 at 7:32 PM, wrote: URL: http://svn.apache.org/viewvc?rev=1381800&view=rev Log: Re-impl

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Philip Martin
Paul Burba writes: > Anybody else seeing this? > > I haven't figured out why yet, but r1381800 is causing failures on my > Windows box, all similar to this: > > C:\SVN\src-trunk>win-tests.py -d -c --test=basic#1 --log-to-stdout > --log-level=DEBUG > Testing Debug configuration on local repository

Re: svn commit: r1381808 - in /subversion/trunk/subversion/libsvn_fs_fs: fs_fs.c fs_fs.h

2012-09-07 Thread Daniel Shahaf
I think this fix is wrong. svn_fs_fs__read_noderev() should return the parsed noderev as is --- if it's a revision file and the noderev contains an is-fresh-txn-root field, then it needs to reflect that. stef...@apache.org wrote on Thu, Sep 06, 2012 at 23:52:17 -: > Author: stefan2 > Date: Th

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Julian Foad
(Just forwarding this thread to Stefan.) - Julian   I (Julian Foad) wrote: > Paul Burba wrote: > >> On Thu, Sep 6, 2012 at 7:32 PM,  wrote: >>>   URL: http://svn.apache.org/viewvc?rev=1381800&view=rev >>>   Log: >>>   Re-implement svn_io_read_length_line as this is one of the >>>   most-call

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Julian Foad
Paul Burba wrote: > On Thu, Sep 6, 2012 at 7:32 PM,  wrote: >> URL: http://svn.apache.org/viewvc?rev=1381800&view=rev >> Log: >> Re-implement svn_io_read_length_line as this is one of the >> most-called functions in SVN. Instead of reading data a byte >> at a time, read 128 byte chunks and s

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Paul Burba
On Thu, Sep 6, 2012 at 7:32 PM, wrote: > Author: stefan2 > Date: Thu Sep 6 23:32:11 2012 > New Revision: 1381800 > > URL: http://svn.apache.org/viewvc?rev=1381800&view=rev > Log: > Re-implement svn_io_read_length_line as this is one of the > most-called functions in SVN. Instead of reading data

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread C. Michael Pilato
On 09/07/2012 11:30 AM, C. Michael Pilato wrote: > On 09/07/2012 09:59 AM, Bert Huijben wrote: >> Can you add a note to the release notes to tell that this changes the >> default behavior? > > Attaching a patch for the release notes, lest I fail to remember to patch > them. I committed my relea

Re: svn commit: r1381766 - /subversion/trunk/subversion/libsvn_subr/utf_validate.c

2012-09-07 Thread Peter Samuelson
[stef...@apache.org] > + for (; len > 0; ++data, --len) > + if (*data < 0) > +return data; Two major problems: 1) It doesn't account for character sets like Shift-JIS, where state change is signalled by ASCII 14 and 15 ("shift out", "shift in") 2) It assumes a 'char

Re: svn commit: r1381766 - /subversion/trunk/subversion/libsvn_subr/utf_validate.c

2012-09-07 Thread Branko Čibej
On 06.09.2012 23:30, stef...@apache.org wrote: > Author: stefan2 > Date: Thu Sep 6 21:30:40 2012 > New Revision: 1381766 > > URL: http://svn.apache.org/viewvc?rev=1381766&view=rev > Log: > Many strings we need to convert to UTF-8 (paths, even log messages) > contain large ASCII chars only sections

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread C. Michael Pilato
On 09/07/2012 09:59 AM, Bert Huijben wrote: > Can you add a note to the release notes to tell that this changes the default > behavior? Attaching a patch for the release notes, lest I fail to remember to patch them. -- C. Michael Pilato CollabNet <> www.collab.net <> Enterprise Cloud D

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread Mark Phippard
On Fri, Sep 7, 2012 at 10:51 AM, C. Michael Pilato wrote: > > I may be wrong, but I kinda get the sense that the GUI clients take their > pick of which Subversion runtime config options make sense for them to > honor. At the moment, the only code which honors the state of the new > configuration

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread Stefan Küng
On 07.09.2012 16:51, C. Michael Pilato wrote: On 09/07/2012 10:14 AM, Stefan Küng wrote: On 07.09.2012 16:07, Mark Phippard wrote: On Fri, Sep 7, 2012 at 10:04 AM, C. Michael Pilato mailto:cmpil...@collab.net>> wrote: On 09/07/2012 09:57 AM, Mark Phippard wrote: > Shouldn't this def

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread C. Michael Pilato
On 09/07/2012 10:14 AM, Stefan Küng wrote: > On 07.09.2012 16:07, Mark Phippard wrote: >> On Fri, Sep 7, 2012 at 10:04 AM, C. Michael Pilato > > wrote: >> >> On 09/07/2012 09:57 AM, Mark Phippard wrote: >> > Shouldn't this default to 'yes' to match the current b

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread Stefan Küng
On 07.09.2012 16:07, Mark Phippard wrote: On Fri, Sep 7, 2012 at 10:04 AM, C. Michael Pilato mailto:cmpil...@collab.net>> wrote: On 09/07/2012 09:57 AM, Mark Phippard wrote: > Shouldn't this default to 'yes' to match the current behavior? I went with Joe Orton's explicit suggestion

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread C. Michael Pilato
On 09/07/2012 09:59 AM, Bert Huijben wrote: > Nice patch. > > Can you add a note to the release notes to tell that this changes the default > behavior? Pending the result of Mark's challenge to this change of behavior, yes I'll do that. > Are you also looking into steveking's request to allow r

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread Mark Phippard
On Fri, Sep 7, 2012 at 10:04 AM, C. Michael Pilato wrote: > On 09/07/2012 09:57 AM, Mark Phippard wrote: > > Shouldn't this default to 'yes' to match the current behavior? > > I went with Joe Orton's explicit suggestion (and Karl's implicit approval > thereof) that we break compatibility and chang

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread C. Michael Pilato
On 09/07/2012 09:57 AM, Mark Phippard wrote: > Shouldn't this default to 'yes' to match the current behavior? I went with Joe Orton's explicit suggestion (and Karl's implicit approval thereof) that we break compatibility and change the default behavior. See http://svn.haxx.se/dev/archive-2008-07

RE: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread Bert Huijben
> -Original Message- > From: cmpil...@apache.org [mailto:cmpil...@apache.org] > Sent: vrijdag 7 september 2012 15:53 > To: comm...@subversion.apache.org > Subject: svn commit: r1382028 - in /subversion/trunk/subversion: > include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.

Re: svn commit: r1382028 - in /subversion/trunk/subversion: include/svn_config.h libsvn_subr/cmdline.c libsvn_subr/config_file.c

2012-09-07 Thread Mark Phippard
On Fri, Sep 7, 2012 at 9:53 AM, wrote: > Author: cmpilato > Date: Fri Sep 7 13:53:05 2012 > New Revision: 1382028 > > URL: http://svn.apache.org/viewvc?rev=1382028&view=rev > Log: > Finish issue #2410 ("Allow client to avoid SSL certificate prompts"). > This adds a runtime configuration knob for

RE: Externals: Multiple definitions for the same path

2012-09-07 Thread Bert Huijben
> -Original Message- > From: MARTIN PHILIP [mailto:codematt...@ntlworld.com] On Behalf Of > Philip Martin > Sent: vrijdag 7 september 2012 15:27 > To: Walter Klust > Cc: dev@subversion.apache.org > Subject: Re: Externals: Multiple definitions for the same path > > I suppose we could defi

Re: Externals: Multiple definitions for the same path

2012-09-07 Thread Philip Martin
I suppose we could define the order for a single svn:externals but users can also use multiple svn:externals: svn ps svn:externals ^/X1 A/B/C' . svn ps svn:externals ^/X2 B/C' ./A There has been at least one report that something like (I don't remember the exact details): svn ps svn:externals ^/

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Stefan Fuhrmann
On Fri, Sep 7, 2012 at 10:11 AM, Ivan Zhakov wrote: > On Fri, Sep 7, 2012 at 3:32 AM, wrote: > > Author: stefan2 > > Date: Thu Sep 6 23:32:11 2012 > > New Revision: 1381800 > > > > URL: http://svn.apache.org/viewvc?rev=1381800&view=rev > > Log: > > Re-implement svn_io_read_length_line as this

Re: Externals: Multiple definitions for the same path

2012-09-07 Thread Walter Klust
Another minor issue: The sequence in which external definitions are applied on checkout/update is not explicitly specified in the documentation. The current behaviour should be documented and an according testcase provided. Thoughts ? On 06.09.2012 17:18, C. Michael Pilato wrote: Walter, I ag

Workaround for updating directories with external files (was: Updating root external files results in an error message)

2012-09-07 Thread Maciej Jaros
Hi. Since I have many boundles (use them as editions) with externals only that include various files I needed a workaround for the error. In brief the error occurs when you have an "external" that points to the same directory/repo (or at least this is my case). The workaround script runs upd

Re: [PATCH] Re: [RFC] svnwcsub post-update hook

2012-09-07 Thread Daniel Shahaf
Daniel Shahaf wrote on Thu, Sep 06, 2012 at 09:48:51 +0100: > Daniel Shahaf wrote on Thu, Sep 06, 2012 at 09:07:58 +0100: > > RFC: teach svnwcsub to optionally run an arbitrary command (a hook) > > after updating the working copy. The interface might be: > > % ${/path/to/hook/from/config} "post-up

Re: svn commit: r1381800 - /subversion/trunk/subversion/libsvn_subr/io.c

2012-09-07 Thread Ivan Zhakov
On Fri, Sep 7, 2012 at 3:32 AM, wrote: > Author: stefan2 > Date: Thu Sep 6 23:32:11 2012 > New Revision: 1381800 > > URL: http://svn.apache.org/viewvc?rev=1381800&view=rev > Log: > Re-implement svn_io_read_length_line as this is one of the > most-called functions in SVN. Instead of reading data