Normand,<br><br>I doubt if it was a libxml issue. It seems like mapserv->sendheaders should be set to false in mapserv.c in order not to print out the headers along with the output.<br>This switch is contolled by the mapserv -nh commandline option, and I suspect this haven't been set by the wxs test scripts yet.<br>
<br>Best regards,<br><br>Tamas<br><br><br><br><div class="gmail_quote">2009/3/6 Normand Savard <span dir="ltr"><<a href="mailto:nsavard@mapgears.com">nsavard@mapgears.com</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div><div class="h5">Tamas Szekeres wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi All,<br>
<br>
I'm wondering how the msautotest should be executed on windows. Currently I<br>
get much of the failures because mapserv provides the result with the<br>
content-type included at the beginning of each file (including the png-s)<br>
which is compared to the expected files without having that extra<br>
information.<br>
Is there a way to suppress this header inclusion without touching the code?<br>
(I know mapserv has the '-nh' option but it seems mstestlib.py should be<br>
changed to include this switch)<br>
<br>
<br>
<br>
</blockquote></div></div>
Tamas,<br>
<br>
I think it may be related to libxml2 version being lower then 2.6.31.<br>
<br>
See <a href="http://trac.osgeo.org/mapserver/ticket/2776" target="_blank">http://trac.osgeo.org/mapserver/ticket/2776</a><br>
<br>
I opened 2777 to try to find a fix in the msautotest tests. I have something in mind and will test it.<br>
<br>
<br>
Norm<br>
</blockquote></div><br>