[Laszlo-dev] explicit replication test cases?

[André Bargull]

I vote for declaring xpath as a String, because I know many developers 
have problems with double-quoting things.
Just take <basevaluecomponent> with its "value"-field which is also 
declared as expression. If you perform a search on the forum for 
problems with "value", you'll get a dozen of threads, because people 
don't know they must double-quote strings for this field.
And I think it's confusing for users to remember to use a double-quoted 
xpath for <replicator>, but a single-quoted xpath for <datapointer>. The 
same goes for the various "itemdatapath", "contentdatapath" fields on 
components. So, almost everywhere you need to use a single-quoted xpath. 
That'd be a huge break to be forced to write double-quotes...

Additionally, if you ever need to have a computed xpath, you can use 
i.e. an "once-constraint".

André


P T Withington wrote:
> Because I made it be an expression, so you could have a computed  
> xpath.  If that is not useful, we can change it to a String.
>
> On 2007-09-03, at 18:25 EDT, Sarah Allen wrote:
>   
>> thanks!  hmmm.... why is xpath declared with double quotes:
>>
>>    <replicator name="foo"
>>                dataset="mydata"
>>                xpath="'/product/'">
>>
>> ?
>>
>> P T Withington wrote:
>>     
>>> I put them in tests/explicit-replicators
>>>
>>> On 2007-09-03, at 17:33 EDT, Sarah Allen wrote:
>>>       
>>>> Are there existing test for explicit replication?  I don't see  
>>>> anything in /wafflecone/test/data that seems like it would cover  
>>>> that based on filenames...
>>>>
>>>> Thanks,
>>>> Sarah
>>>>         

-- 
Official OpenLaszlo-Committer ^_^
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.openlaszlo.org/pipermail/laszlo-dev/attachments/20070904/0c8769e8/attachment.html