Re: Testing automake on Cygwin 1.7

[Peter Rosin]

Stefano Lattarini skrev 2011-11-28 20:43:
> On Monday 28 November 2011, Peter Rosin wrote:
>> Regarding the commit message for the 2/2 patch, the trouble with
>> the double leading slash is not Cygwin specific, but instead a
>> run-of-the-mill POSIX violation.
>>
>> A leading double slash is the key into an implementation defined
>> namespace.
>>
>> From the end of section 4.11 at:
>> http://pubs.opengroup.org/onlinepubs/009695399/basedefs/xbd_chap04.html#tag_04_11
>>
>>      "A pathname that begins with two successive slashes may be
>>      interpreted in an implementation-defined manner, although
>>      more than two leading slashes shall be treated as a single
>>      slash."
>>
> Still, Cygwin is AFAIK the only non-museum implementation that handles
> leading `//' differently from `/'.  That said, we are in bikeshedding
> territory, and the patch is in your name, so ...

It's pretty nifty, so I wouldn't be surprised if other uses crop up
elsewhere, or already have cropped up in some option somewhere, just
waiting to bite the unsuspecting user. Turn on some optional kernel
module in Samba e.g., and have UNC paths in Linux just like in
Cygwin. Or whatever...

>> So, please use this commit message (or something like it)
>>
>>      tests: fix 'distcheck-override-infodir.test' on Cygwin
>>      * tests/distcheck-override-infodir.test (Makefile.am): Do not add
>>      any `/' between $(DESTDIR) and the following paths.  Otherwise,
>>      when $(DESTDIR) is empty, the recipes will try to access files
>>      with a leading double slash, which have an implementation defined
>>      interpretation.  For Cygwin they mean UNC paths.
>>
> ... I've amended the ChangeLog entry accordingly (next time, I'll leave 
> to you the burden of distilling your fixes in proper patches ;-).

Yes, I know I'm lazy. Thanks!

> Attached is what I'll push in 72 hours or by your ACK (whichever comes
> first).

ACK

Cheers,
Peter