Re: [patch@35050] replace pathify_dirspec in VMS with new version

Craig A. Berry wrote:
> 
> On Dec 7, 2008, at 8:46 AM, John E. Malmberg wrote:
> 
>> An additional note, this patch may cause some tests in vmsfspec.t to 
>> fail.  That is because the old behavior of pathfiy_dirspec() was 
>> inconsistent with vmsify().
>>
> 
> Specifically:
> 
> not ok 92 - pathify('/__down_/__the_/__garden_/__path_.'): 
> '/__down_/__the_/__garden_/__path__/'
> # Failed at [.lib]vmsfspec.t line 24
> #      got '/__down_/__the_/__garden_/__path__/'
> # expected ''
> ...
> not ok 94 - pathify('/__down_/__the_/__garden_.__path_'): 
> '/__down_/__the_/__garden____path_/'
> # Failed at [.lib]vmsfspec.t line 24
> #      got '/__down_/__the_/__garden____path_/'
> # expected ''
> ...
> not ok 106 - pathify('__path_.notdir'): '__path__notdir/'
> # Failed at [.lib]vmsfspec.t line 24
> #      got '__path__notdir/'
> # expected ''
> 
>> I am working on a new version of vmssfspec.t that handles both the 
>> traditional mode of perl and with the decc$efs_charset / 
>> decc$filename_unix_report modes activated.
>> 
> How does this plan address the failures above?

The "got" results will be come the expected results.

This is needed to make the handling of dots in directories consistent 
with what vmsify does when it encounters a dot in a directory path.

See the results of vmsify($unixpath . '/') where $unixpath is not '' for
the same inputs.


It is also needed for better handling of symbolic links which act like 
directories and but are not named .dir.

-John
[email protected]
Personal Opinion Only