Bug fix and enchantment[*] in cygpath.cc

Dave Korn dave.korn@artimi.com
Thu May 17 12:22:00 GMT 2007

On 17 May 2007 13:13, Christopher Faylor wrote:

> On Thu, May 17, 2007 at 01:07:52PM +0100, Dave Korn wrote:
>> On 14 May 2007 11:56, Corinna Vinschen wrote:
>>> On May 12 15:47, Ilya Bobir wrote:
>>>>>> [...]
>>>>>> 	* cygpath.cc (get_long_name): Fallback to get_long_path_name_w32impl.
>>>>>> 	Properly null-terminate 'buf'.
>>>> [...]
>>>> I've submitted this patch on 30.07.2006, but it seems that the bug still
>>>> exists in cygwin-1.5.24-2 that was released on 31.01.2007.
>>>> I can see that HEAD CVS brunch contains the fix.
>>>> Why is that so?  Is it some kind of mistake?
>>> No, the patch has gone into CVS HEAD, but the current releases are
>>> taken from the cr-0x5f1 branch.  CVS HEAD is not yet ready for prime
>>> time.
>> Is there a summary of the branches and what they're for anywhere?
> The branches aren't for public consumption, so no.
> cgf

  Well, the cr-0x5f1 branch clearly is, or was, or perhaps was when it was
HEAD but isn't any more, and there are/have been other cr-xxx branches in the
past, and how are people meant to know what to build?  Anyone who wants to try
and fix a problem in 1.5.24 (for reasons of supporting some current
installation) by building from CVS *needs to know* to be on the branch, and
anyone who wants to work on 1.7 needs to know to be on mainline.

  So is it at least the case that cr-0x5f1 and HEAD are /all/ that anyone
needs to know about?

Can't think of a witty .sigline today....

More information about the Cygwin-developers mailing list