On Thu, Jan 14, 2010 at 23:21, Eddie Drapkin <[email protected]> wrote:
> On Thu, Jan 14, 2010 at 4:43 PM, Hannes Magnusson
> <[email protected]> wrote:
>> I.. don't know.
>> I look at the class reference page when I want quickly see the access
>> modifiers of the methods.
>> And, its documented on the method pages..
>>
>> I don't really see the need to parse the access modifiers and then
>> manipulating all <methodname>, <xref>, <refname>, and wherever it may
>> be mentioned.
>>
>> Having a strict markup convention for this is going to be biatch to
>> enforce, and eventually leading to confusion.
>>
>>
>> I'm sure we can implement this in PhD though if you all feel strongly about 
>> it.
>>
>> -Hannes
>>
>
> Having not looked a lot at the source of PhD, how difficult would this
> be to implement?  I think it should just be a check if there's a
> <modifier>static</modifier> child of each <methodsynopsis>, shouldn't
> it?

Not exactly.
This has to be done in the indexerer so <xref
linkend="datetime.createfromformat" /> will generate the correct text.
Also, it would have to check *all* <function> and <methodname>, look
it up in the index, and fetch the "preferred" format and magically
replace the content.
Same with <refname>, and all TOCs linking to it, will have to be
checked against the index to print the correct format.

I can't think of any other edge cases at the moment, but this as to be
consistent throughout every mention of every method, otherwise it just
creates confusion for the end-user.

Also, we still do have some <link
linkend="classname.methodname">Classname::methodName</link> around,
which would have to be found and fixed.

-Hannes

Reply via email to