Re: Useless dependency assumption libxml2 -> libxslt in MSVC scripts

Michael Paquier <michael(dot)paquier(at)gmail(dot)com> writes:
> On Thu, Sep 8, 2016 at 9:39 PM, Magnus Hagander <magnus(at)hagander(dot)net> wrote:
>> Pretty sure this goes back to *our* XML support requiring both. As in you
>> couldn't turn on/off xslt independently, so the "xml requires xslt" comment
>> is that *our* xml support required both.
>>
>> This may definitely not be true anymore, and that check has just not been
>> updated.
>>
>> Also this was 10 years ago, so I'm of course not 100% sure, but I think it
>> was something like that...

> Was it for contrib/xml2? For example was it because this module could
> not be compiled with just libxml2?

The core code has never used xslt at all. Some quick digging in the git
history suggests that contrib/xml2 wasn't very clean about this before
2008:

commit eb915caf92a6805740e949c3233ee32bc9676484
Author: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Date: Thu May 8 16:49:37 2008 +0000

Fix contrib/xml2 makefile to not override CFLAGS, and in passing make it
auto-configure properly for libxslt present or not.

or even 2010, depending on how large a value of "work" you want:

commit d6a6f8c6be4b6d6a9e90e92d91a83225bfe8d29d
Author: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Date: Mon Mar 1 18:07:59 2010 +0000

Fix contrib/xml2 so regression test still works when it's built without libxslt.

This involves modifying the module to have a stable ABI, that is, the
xslt_process() function still exists even without libxslt. It throws a
runtime error if called, but doesn't prevent executing the CREATE FUNCTION
call. This is a good thing anyway to simplify cross-version upgrades.

Both of those fixes postdate our native-Windows port, though I'm not sure
of the origin of the specific test you're wondering about.

regards, tom lane