[Mapcss] Bug in the mapcss introduction slieds of Maskim?

[Thomas Davie]

The problem with implicit ordering in the style is things like this:

way[waterway]
{
  colour: blue;
  width: 5pt;
}

way[highway]
{
  colour: red;
  width: 11pt;
}

way[waterway="river"]
{
  text: name;
}

Which gets rendered first.  Secondly, the other issue is that it requires you to tie the renderer and styler more closely together – suddenly the renderer needs to care about when the rule(s) for a particular object were matched in the stylesheet, rather than just the computed style.

I'd much rather see the MapCSS workflow look like

StyledObject = (Object, ComputedStyle)

[Object] --styler-> [StyledObject] --renderer-> PrettyPicture

Bob
if (*ra4 != 0xffc78948) { return false; }

On 27 Jan 2012, at 12:43, Peter Wendorff wrote:

> Hi.
> If you are right (and yes, that's possible, I can't prove the opposite), I think this should be stated clearly in the documentation.
> 
> On the other hand I'm not sure if this variant is really the best idea.
> It leads to verbose, hard do maintain mapcss stylesheets if the developer wants to set up a fixed rendering order, and usually at maps that's the case for most rules, I think.
> 
> I would change it in the wiki, but I would like to get more comments on that from the list, so I don't change it now.
> 
> regards (and thanks for the answer)
> 
> Peter
> 
> Am 27.01.2012 13:34, schrieb Thomas Davie:
>> 
>> At least in my interpretation of MapCSS, the behaviour when two objects are defined to have the same z-index is undefined.  The order of rendering here should (in my book) be effectively random.
>> 
>> Bob
>>  if (*ra4 != 0xffc78948) { return false; }
>> 
>> On 27 Jan 2012, at 12:24, Peter Wendorff wrote:
>> 
>>> Hi.
>>> I have to give a short introduction to someone who should implement a mapcss renderer and looked into the docs availlable for that.
>>> For that I especially looked into the sotm-eu presentation done by Maskim [1], and I fear there is a small bug on slide 7 (and the following pages).
>>> 
>>> The stylesheet presented does not contain any z-index or layer definitions. To quote the slide, it's defined as:
>>> 
>>> line[highway] {
>>>    color: orange;
>>>    width: 11;
>>> }
>>> 
>>> area[building] {
>>>    fill-color: gray;
>>> }
>>> 
>>> line[waterway] {
>>>    color: blue;
>>>    width:3;
>>> }
>>> 
>>> The image presented nearby shows a similar image, but the highway is rendered on top of the river.
>>> Shouldn't that be the other way around (according to the mapcss code, not to "good maps")?
>>> 
>>> regards
>>> Peter
>>> 
>>> [1] http://sotm-eu.org/slides/10_MaksimGurtovenko_MapCSS.pdf
>>> 
>>> _______________________________________________
>>> Mapcss mailing list
>>> Mapcss at openstreetmap.org
>>> http://lists.openstreetmap.org/listinfo/mapcss
>> 
>> 
>> 
>> _______________________________________________
>> Mapcss mailing list
>> Mapcss at openstreetmap.org
>> http://lists.openstreetmap.org/listinfo/mapcss
> 
> _______________________________________________
> Mapcss mailing list
> Mapcss at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/mapcss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/mapcss/attachments/20120127/68533720/attachment-0001.html>