RFC 26 - MapServer Terminology Cleanup
dmorissette at MAPGEARS.COM
Tue May 8 11:14:39 EDT 2007
Yes, I think the default has to be set to FALSE. And I agree that this
can be an annoying change for some people with lots of queryable layers,
but I can't think of any other alternative with a lesser impact on
users. SteveW's suggestion of a Perl script to upgrade a mapfile could
help for those with access to Perl (may be a bit too much trouble for
Thinking about it some more, a simple search/replace in any text editor
to search for "TEMPLATE" and replace with "QUERYABLE TRUE TEMPLATE"
would do... except for the WEB and LEGEND TEMPLATEs that would end up
Steve Lime wrote:
> That's a pretty big change for folks to have to go through every layer and set "QUERYABLE TRUE".
> Default should probably be conservatively set to FALSE.
>>>> On 5/7/2007 at 10:06 AM, in message <463F4093.2080206 at mapgears.com>, Daniel
> Morissette <dmorissette at MAPGEARS.COM> wrote:
>> Steve Lime wrote:
>>> Is there a way we could do this in the code without adding a new parameter?
>>> thinking probably not since you can use layers with WxS service parameters
>>> with the CGI or MapScript... So, I would propose recognizing the template or
>> a queryable
>> What if someone sets TEMPLATE "something.html" and also sets QUERYABLE
>> FALSE. With what you propose the QUERYABLE setting would be ignored and
>> the layer would still be queryable, right? That would be creating
>> another source of confusion. I think if we add a QUERYABLE TRUE/FALSE
>> setting then TEMPLATE should be treated as a simple parameter and
>> queryability should be determined only by the QUERYABLE setting.
>> That's just my opinion, let's see what others think.
>>> Since we have a function like msIsLayerQueryable that should be easy
>> I think there's more to do than just updating msIsLayerQueryable().
>> First there is the class by class queryability that is not handled by
>> msIsLayerQueryable(). I also did a quick check and there is lots of code
>> that checks the layer and class template values directly and would need
>> to be updated. There's also some code (mapwfs.c) that sets dummy
>> template values that we could finally get rid of. I could take care of
>> the implementation if you want.
More information about the mapserver-dev