[mapserver-dev] MOTION: Symbol-level keyword removal

[Daniel Morissette]

I declare this motion passed with +1 from ThomasB, SteveW, DanielM, 
FrankW, AssefaY and SteveL.

I have committed the changes to the source in r11472 and r11473.

Docs and migration guide updates are still pending.

http://trac.osgeo.org/mapserver/ticket/3797
http://trac.osgeo.org/mapserver/ticket/3798

Daniel

On 11-04-01 02:55 PM, Daniel Morissette wrote:
> Hi PSC,
>
> *** MOTION ***
>
> Removal of deprecated symbology-related keywords in MapServer 6.0:
>
> 1- Remove any references to CARTOLINE still left in the code and
> documentation. CARTOLINE had stopped working a few releases ago anyway.
>
> 2- Remove the following GAP, PATTERN, LINECAP/JOIN and POSITION keywords
> from symbolObj with the understanding that they were deprecated since
> around version 5.4 and conceptually those values really belong at the
> STYLE level. (see Thomas' explanation below)
>
> ***
>
> And I (Daniel) vote +1.
>
> We've already got formal +1 on this in the previous thread from ThomasB
> and SteveW. A few others have expressed support without voting officially.
>
> So please, PSC members, vote on this motion, and if we get no objections
> by Tuesday April 5 then I will take care of this before next week's beta.
>
> Related tickets:
>
> http://trac.osgeo.org/mapserver/ticket/3797
> http://trac.osgeo.org/mapserver/ticket/3798
>
> Daniel
>
>
> On 11-04-01 12:47 PM, thomas bonfort wrote:
>> Christy,
>> Supporting those keywords on STYLES and SYMBOLS makes us jump through
>> hoops to try and support it consistently throughout the codebase. I
>> understand the need you have, and would hope that someone comes up
>> with funding for real style inheritance in mapserver.
>> Until then, using gap/pattern on a symbol is conceptually incorrect,
>> as a SYMBOL is just the reference to a marker type, and it is the
>> STYLE that references it that is used to determine how it should be
>> styled for each feature. I agree that the previous version of
>> mapserver did not clearly make the distinction between the two
>> concepts, but that derived due to the limitations of the GD renderer
>> with respect to advanced symbology.
>>
>> As the mapfiles have to be updated anyhow in 6.0 due to other
>> backwards-compatibility breaking changes, now would seem the good time
>> to update them for those keywords too, rather than forcing another
>> change in 7.0.
>>
>> regards,
>>
>> Thomas
>>
>> On Fri, Apr 1, 2011 at 16:44, Christy Nieman<cnieman at dmsolutions.ca>
>> wrote:
>>> I just wanted to put in my $0.02 on this.
>>>
>>> I'm a little worried that removing parameters like these from the symbol
>>> file will make large and complex mapfiles more difficult to manage
>>> without
>>> some sort of named style inheritance (which I don't think exists
>>> yet?). I
>>> have many symbols for line styles that I reuse throughout a mapfile and
>>> having to change gap values, line caps and/or patterns for each class
>>> that
>>> uses the style will make management of the mapfiles a lot more involved.
>>>
>>> Christy
>>>
>>> --
>>> View this message in context:
>>> http://osgeo-org.1803224.n2.nabble.com/6-0-backwards-compatibility-tp6226113p6231271.html
>>>
>>> Sent from the Mapserver - Dev mailing list archive at Nabble.com.
>>> _______________________________________________
>>> mapserver-dev mailing list
>>> mapserver-dev at lists.osgeo.org
>>> http://lists.osgeo.org/mailman/listinfo/mapserver-dev
>>>
>> _______________________________________________
>> mapserver-dev mailing list
>> mapserver-dev at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/mapserver-dev
>
>


-- 
Daniel Morissette
http://www.mapgears.com/
Provider of Professional MapServer Support since 2000