Stephan I have created a patch that might solve this problem. It is attached to the ticket:
http://trac.osgeo.org/mapserver/ticket/2803 could you try it and report if it works? Thanks, Umberto On Tue, Nov 4, 2008 at 9:40 AM, Stephan Holl <[EMAIL PROTECTED]> wrote: > Hello Umberto, > > "Umberto Nicoletti" <[EMAIL PROTECTED]>, [20081104 - 09:10:26] > >> On Tue, Nov 4, 2008 at 9:05 AM, Stephan Holl >> <[EMAIL PROTECTED]> wrote: >> > Hello Umberto, >> > >> > "Umberto Nicoletti" <[EMAIL PROTECTED]>, [20081104 - >> > 08:46:24] >> > >> >> I was hit by the same bug: >> >> >> >> when you specify hilite you MUST have a style section for each of >> >> your classes. The fix is therefore to add a style section to ALL >> >> of your classes. >> > >> > You mean each of my classes of layers which are being queryable? >> >> exactly. > > uff, my mapfile is more than 10000 lines long and has _many_ classes... > > But yes, testing with one layer and it worked. > >> >> Failure to comply will cause the new agg code (I think the gd >> >> renderer is immune, but it needs to be checked) to dereference a >> >> null pointer. The problem also exists with cgi mapserver and is >> >> not only related to mapscript. >> > >> > Ah, I am using the gd-rendering though. >> >> then I must correct my statement: also the gd driver is affected. The >> bug was introduced when we (mostly I, for those who like finger >> pointing ;-) ) rewrote the layers, classes, styles arrays to be arrays >> of pointers. >> >> I'll file an issue on bugzilla later today. > > Please drop a note (and a URL) on this thread so we find it. > >> > Ugh, thank your for your hint. I was not aware of that. Did I >> > overlook this issue in the migration-guide? >> >> Nope, I think we (the developers) were not aware of this problem until >> now so it was probably not documented there. > > Thank you for clarifying! > > Best regards > > Stephan > >> >> On Mon, Nov 3, 2008 at 3:20 PM, Stephan Holl >> >> <[EMAIL PROTECTED]> wrote: >> >> > Dear mapserver-users, >> >> > >> >> > I am updating a mapserver 4.1.0.3 app to mapserver 5.2.0 and >> >> > found the issue that the following mapfile-block produces a seg >> >> > fault with php-mapscript. >> >> > >> >> > QUERYMAP >> >> > COLOR 0 255 255 >> >> > STATUS on >> >> > STYLE hilite >> >> > END >> >> > >> >> > With mapserver 4.10.3 everything was fine. Is this a known issue? >> >> > Setting it the querymap to something else than STYLE 'hilite' >> >> > everything works normal. >> >> > >> >> > Did anybody run into this? I found a bug about another >> >> > hilite-stuff[1], but also the 5.2-branch from SVN has this >> >> > behaviour. >> >> > >> >> > Thanks for any comments on this. >> >> > >> >> > Best >> >> > >> >> > Stephan >> >> > >> >> > -- >> >> > Stephan Holl <[EMAIL PROTECTED]> | Tel.: +49 (0)541-33 >> >> > 50 8 32 Intevation GmbH, Neuer Graben 17, 49074 OS | AG >> >> > Osnabrück - HR B 18998 Geschäftsführer: Frank Koormann, >> >> > Bernhard Reiter, Dr. Jan-Oliver Wagner >> >> > >> >> > _______________________________________________ >> >> > mapserver-users mailing list >> >> > [email protected] >> >> > http://lists.osgeo.org/mailman/listinfo/mapserver-users >> >> > >> >> > >> > >> > >> > -- >> > Stephan Holl <[EMAIL PROTECTED]> | Tel.: +49 (0)541-33 50 >> > 8 32 Intevation GmbH, Neuer Graben 17, 49074 OS | AG Osnabrück - >> > HR B 18998 Geschäftsführer: Frank Koormann, Bernhard Reiter, Dr. >> > Jan-Oliver Wagner >> > > > > -- > Stephan Holl <[EMAIL PROTECTED]> | Tel.: +49 (0)541-33 50 8 32 > Intevation GmbH, Neuer Graben 17, 49074 OS | AG Osnabrück - HR B 18998 > Geschäftsführer: Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner > _______________________________________________ mapserver-users mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/mapserver-users
