On Wednesday 02 July 2008 12:55:51 Andrea Aime wrote:
> Mauricio Pazos ha scritto:
> > On Wednesday 02 July 2008 11:05:09 you wrote:
> >> Jody Garnett ha scritto:
> >>> Andrea Aime wrote:
> >>>> Anyone against this?
> >>>
> >>> We need the check with CQL parser people; make sure "." is an
> >>> acceptable thing.
> >>
> >> Hi Victor,
> >> how is CQL going to handle this one? We're looking for a way to specify
> >> "this", that is, something that returns the feature as opposed to one
> >> of its attributes. Will the CQL parser break trying to parse "." or
> >> will it consider it a valid "attribute name"?
> >
> > Sure, "." is acceptable, I have extracted the following sample from CQL
> > tiest
> >
> > gmd:aa:bb.gmd:cc.gmd:dd < 100
> >
> > (please take a look in CQL test)
>
> Wow... what's that? Not sure I understand what's the role of the
> columns in that expression.
Yes the sintax alows that expression
<identifier start [ { <colon> | <identifier part> }... ]
but, I cannot imagine if it has an application (this is the reason why I write
an "abstract" test sample (gmd:aa:bb.gmd:cc.gmd:dd).
> Anyways, I've tried parsing ". < 3" and it seems . is not considered
> a valid identifier. What we do want is to use "." as a shortcut for
> "this", that is, a way to specify the parameter provided to the filter,
> as opposed to one of its attributes.
ok
>
> Like, I need to feed "abcd" into a filter defined as "len(.) < 3"
> and have the filter returns "false". If it's not possible to do this
> with CQL or TXT too bad, but it's not the end of the world, since
> feature validity filters can be built using the filter factory
> directly and "." is a valid xpath.
>
> >> I'm also wondering, would CQL work with other xpath like expressions
> >> as well? Think something like name1/name2 or name1[att1 = val1]?
> >> If not CQL, what about TXT?
> >
> > Well, you know that CQL was defined by the OGC, so I would prefer to
> > maintain this as "canonical" language, but TXT is our escape or back
> > door. TXT is being developed thinking in our needs.
> >
> > My questions are:
> >
> > Why is needed more than one way to specify attributes or property names?
> > If we accept that posibility TXT could be a big language (hard to learn).
> > I prefer an othogonal solution (a few elements and many composition
> > posibilities)
> >
> > Maybe the CQL definition (<identifier><dot><identifier> ...) is not the
> > common use, so we could change this by that you propose (name1/name2 or
> > name1[att1 = val1]) in TXT.
>
> I don't think we have to follow the xpath syntax exactly, but as complex
> features come in, it may become interesting (or necessary) to support
> the same semantics. So a.b.c is fine. xpath has lots more, like the
> filtering example I specified above, but I'm not sure wheter we'll
> support it or not.
>
> My question was just intended to see how powerful TXT was compared
> to xpath, but only as a curiosity.
Intersting work, if you have online (or available) documentation I would like
take a look.
>
> Cheers
> Andrea
>
>
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> _______________________________________________
> Geotools-devel mailing list
> Geotools-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/geotools-devel
--
Mauricio Pazos
www.axios.es
Tel.: +34 94 682 42 86
-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Geotools-devel mailing list
Geotools-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel
