Re: [Boost-docs] Doxygen features

Subject: Re: [Boost-docs] Doxygen features
From: Daniel James (dnljms_at_[hidden])
Date: 2010-05-31 18:30:00

On 31 May 2010 08:53, Bo Jensen <jensen.bo_at_[hidden]> wrote:
> I am using quickbook+boostbook+doxygen all newest development
> versions. It looks nice, but I can not make some minor things work :

Neither of these are currently supported. I'm surprised that arrays
have never come up before.

> Doxygen commands like @see seem to be ignored, doxygen generates :

I think this is possible. It's a little tricky as it'll have to check
the destination to see what type of link to use, and it might go wrong
if there are overloads. But I think it's mostly possible. Can you file
a ticket?

> I like to use array notation for arrays i.e "int somearray[]", but
> this gets ignored and displayed like "int somearray" in the parameters
> section, eventhough doxygen seem to correctly has parsed my source :

The problem here is that boostbook's markup doesn't support type
information after the variable. The best I could quickly manage is
that attached patch, which outputs the incorrect, but hopefully clear,
'int[] somearray'.

To do better, I think we'll need to change boostbook. One possibility
is to automatically work out where the name goes in the xsl. Of the
top of my head, this will only be needed for arrays and function
pointers, so it might be possible to just put the variable name before
the first '[' or ')'. Does anyone know if there are any cases where
that could go wrong? I suppose we'd need a special case for anonymous
enums, classes and structs, although they don't seem to work at the
moment anyway.

Another possibility is to extend boostbook's syntax. Maybe something like:

<type>
    <prefix>int</prefix>
    <postfix>[]</postfix>
</type>

Any other ideas?

Daniel