Mapserver conectivity to native informix spatial module

[joluinfante@…]

Mapserver don't have native conection to informix spatial module.
I'm working on it.

[dmorissette]

Reassigned bug to Jorge Luis.

Note that we had the feature freeze for the realease of MapServer 4.6 last week,
so this new feature would have to wait until the next release after 4.6 (that
will be 4.8 or 5.0). It should not be committed to CVS until we create the 4.6
branch in CVS in 4-6 weeks.

[hobu]

Was there any ever patch supplied with this work? There will be at least two other native-style drivers from the ISC guys coming soon as well -- SpatialDB and Katmai (MSSQL 2008).

[joluinfante]

Hi. I did prepare the patch to version 4.4.1, and, I'm using it for one year, in the site of "Municipalidad de Rosario" (​http://www.rosario.gov.ar/infomapas/). I'll prepare the patch to the last version. To upload only need to attach here?

TIA

jorge infante direccion de catastro municipalidad de rosario rosario - santa fe - argentina

[hobu]

Jorge,

Attaching the patch here would be great. I would note that this patch will not make it into the upcoming 5.0 release, but it would be an excellent addition for 5.2.

Howard

[joluinfante]

Ok, its ok for me. But, where I can download the sources for 5.2? What version I need to download to work with my patch?

TIA

jorge infante

Replying to hobu:

Jorge,

Attaching the patch here would be great. I would note that this patch will not make it into the upcoming 5.0 release, but it would be an excellent addition for 5.2.

Howard

[hobu]

​http://mapserver.gis.umn.edu/development/cvs contains information about how to get the current development version of the code.

[joluinfante]

The patch

[joluinfante]

Ok. This is a "primary version" of the driver to informix for mapserver, compiled with last version (I did download it from cvs repository). I'm attaching: . informix_spatial.patch . mapinformixspatial.c I have in "production" mapserv+informix, in a 4.4.1. I need to test the new version, but, I'm attaching, to do faster the integration process. I see 2 "bugs" in the cvs version: . Is not working, in routine loadWeb(), the replacing of parameters map_web_*, when you put in the map "set in index.html". Up when I did find the correct solution, my solution is not use "set in index.html" in the map, and then change in the mapfile.c:

in mapfile.c:

loadWeb()

default:

if(strlen(msyytext) > 0) {

msSetError(MS_IDENTERR, "Parsing error near (%s):(line %d)", "loadWeb()"

, msyytext, msyylineno);

return(-1); <-- I'm replacing this code, ignoring return(0); the error.

} else {

return(0); /* end of a string, not an error */

}

. The other "bug", is I can't to compile mapogr.cpp. My "solution" is comment the lines with "ACQUIRE_OLD_OGR_LOCK" and "RELEASE_OLD_OGR_LOCK".

I'll continue working in do better the code in mapinformixspatial.c, but, it is working with a good performance.

Next steps to include my patch in the next release version?

[joluinfante]

Well... I did put the patch. I suppouse I need to work more on it, but, this is working, and I'd like have a feedback about next steps, and work more integrated to the release preparation.

[dmorissette]

The current SVN trunk is v5.0 that we are in the process of releasing so we cannot add the new driver to it at the moment.

After the release of MapServer 5.0 around mid-September, we will branch the SVN tree and the SVN trunk will become the 5.2 development version. At that point I'd invite you to contact the developers on the mapserver-dev list to discuss the integration of the driver.

FYI, a RFC will probably be required describing/documenting the new driver, its features and options, etc and then following a vote of the MapServer PSC to approve the RFC it will be ready to commit to SVN. See ​http://mapserver.gis.umn.edu/development/rfc/ms-rfc-29/ for a sample RFC that could be used as a guide for this.

The process may sound complicated but it's really fairly straightforward. If you have questions then please don't hesitate to ask on the mapserver-dev list.

[pramsey]

Low priority until author submits patch against trunk.

[sdlime]

Moving to FUTURE pending another patch or at least confirmation that what we have works.

Steve

[sdlime]

Oh, changing to "enhancement". Can't break what we don't have. ;-) I hope the author has not vanished...

Steve