[Laszlo-dev] Does the doc tool pick up @initarg? [Was: I think I see a general bug with the 4.x doc]

[Donald Anderson]

I don't know that there's a way that works at present.
What you are doing *appears* to be right - It is honored by js2doc,
and I see <initarg> entries being produced in intermediate files
when @initarg is used.  However, the xslt seems
to fall down on the job even though there are quite a few references  
to initarg
in the transforms.

Please file a JIRA.

On May 8, 2008, at 4:03 PM, P T Withington wrote:

> I tried adding
>
>  * @initarg 'x'|'y' axis: (If lazy replication is requested) The
>  * direction in which the replicated elements will be layed out.
>  * Default is 'y'.
>
> to the doc comment above LzDatapath, but it does not seem to be  
> picked up by the doc tool.  What is the correct way to document an  
> initialize-only argument to a class?
>
> On 2008-05-07, at 15:46 EDT, P T Withington wrote:
>
>> The doc tool used to pick up documentation from http://svn.openlaszlo.org/openlaszlo/branches/3.4/WEB-INF/lps/schema/lzx.rnc 
>>  .  I think that Jim tried to move the relevant documentation into  
>> the LFC source, but perhaps someone needs to systematically go over  
>> it.
>>
>> For example:  In 3.4, the datapath tag has an axis attribute that  
>> can be x or y.  There is no documentation on this attribute, but at  
>> least it is listed.  This attribute is missing altogether from 4.x.
>>
>> It should be the case that every property of a tag that is  
>> specified in the schema is listed in the documentation.  If the  
>> property is not reflected in an instance var of the same name  
>> (e.g., it is handled specially by the constructor, as is the case  
>> for the axis property of datapath) it should be listed as an  
>> @initarg above the class declaration.
>>
>> Maybe everybody else already knew this, but I did not.
>


--

Don Anderson
Java/C/C++, Berkeley DB, systems consultant

voice: 617-547-7881
email: dda at ddanderson.com
www: http://www.ddanderson.com