MIGRATION_GUIDE.TXT

[Yewondwossen Assefa]

Stephen Woodbridge wrote:
> OK, I could be wrong, but this might also be a regression that was 
> introduced at some point.
> 
> Does anyone else remember what the behavior used to be with this?
> 
> Assefa, if you have some older versions of mapserver already loaded can 
> you try this on them also. If the behavior is consistent, then document 
> it. And I'll try to update my memory :)
> 
  Oldest I have locally is Mapserver 4.8.x : the code in the function 
seems to be the same as the one in the current svn. I will update the 
docs (http://mapserver.gis.umn.edu/docs/reference/symbology) to reflect 
this.

> Thanks,
>   -Steve W.
> 
> Yewondwossen Assefa wrote:
>>
>> Stephen Woodbridge wrote:
>>> Yewondwossen Assefa wrote:
>>>> Stephen Woodbridge wrote:
>>>>> I think we need to clarify this point.
>>>>>
>>>>>> ----------------------------------------------------
>>>>>> Styling/Symbolgy changes
>>>>>> ----------------------------------------------------
>>>>>> Prior to MapsServer 5.0, if a pixmap symbol was used in style on a 
>>>>>> Line
>>>>>> Layer, the symbol was used as a brush to draw the line. In 
>>>>>> MapServer 5.0,
>>>>>> it is possible to draw the pixmap symbol along the line (Note that 
>>>>>> this
>>>>>> was available using a true type symbol). To achieve  this, the 
>>>>>> user needs
>>>>>> to use the parameter GAP with it's pixmap symbol definition in the 
>>>>>> symbol
>>>>>> file. The GAP represents the distance between the symbols. If the 
>>>>>> GAP is not
>>>>>> given, the pixmap symbol will be used as a brush.
>>>>>
>>>>> With true type symbols you can specify negative values for GAP to 
>>>>> cause the symbol to align with the line segment. Is this also 
>>>>> supported for pixmap symbols? This should be made explicit in here.
>>>>>
>>>>
>>>>  The pixmap symbols (and the true type symbols) always align with 
>>>> the line segment. The negative gap should have the effect of adding 
>>>> 180deg to the calculated angle.
>>>> I will update the guide to reflect this.
>>>
>>> Assefa,
>>>
>>> The current behavior is that with a positive gap, the symbol will 
>>> always point north, when the gap is negative it will align the symbol 
>>> north with the direction of the line segment. I do not think we want 
>>> to eliminate either of these options. For oneway arrows along 
>>> streets, you need to know if the direction of travel is A-B or B-A 
>>> and use a CLASS to select a flipped symbol for the other direction. 
>>> You can not change the GAP to flip the symbol on a per shape basis in 
>>> the mapfile.
>>>
>>
>>  Steve,
>>
>>  I am not sure that when using a true type fonts,  a positive gap will 
>> always make the symbol point north.  Here is a test I did with a true 
>> type font :
>>
>>  SYMBOL
>> NAME "I"
>> TYPE
>> TRUETYPE
>> FONT "Arial"
>> CHARACTER "I"
>> ANTIALIAS TRUE
>> GAP 15
>> END
>>
>>  and the STYLE looks
>>     STYLE
>>         SYMBOL "I"
>>          SIZE 15
>>         COLOR 120 120 120
>>      END
>>
>>  Attached is the a what it gives in Mapserver
>>
>>  Looking at the result, I interpret this as being the symbol is 
>> aligned with the line and does not always point north. Is that correct ?
>>
>>> -Steve
>>>
>>>>
>>>>> -Steve W
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>>
>> ------------------------------------------------------------------------
>>
> 


-- 
----------------------------------------------------------------
Assefa Yewondwossen
Software Analyst

Email: assefa at dmsolutions.ca
http://www.dmsolutions.ca/

Phone: (613) 565-5056 (ext 14)
Fax:   (613) 565-0925
----------------------------------------------------------------