[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: (sxml match)
From: |
Ludovic Courtès |
Subject: |
Re: (sxml match) |
Date: |
Wed, 26 May 2010 16:12:36 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/23.2 (gnu/linux) |
Hi Andy!
Andy Wingo <address@hidden> writes:
> On Tue 25 May 2010 23:42, address@hidden (Ludovic Courtès) writes:
>> I integrated the doc in the manual. Unfortunately it can’t be made to
>> appear close to the automatically generated doc of (sxml simple), so it
>> actually appears before. I’m open to suggestions to improve that.
>
> You could put the documentation in the module commentary of (sxml
> match), as texinfo. Or you could extend module-commentary somehow to
> look for commentaries in other sources. Probably easiest just to paste
> the texinfo you already wrote into the commentary, no?
Actually I don’t want that much Texinfo in a Scheme file, and I don’t
want the automatically-generated “Usage” subsection.
Besides, I initially started by just putting address@hidden sxml-match.texi’
in the module’s commentary but that didn’t work (I forgot the details.)
>> The test suite reveals an interesting psyntax “problem”: renaming the
>> ‘xyzpq’ variable in sxml-match-tests.ss to ‘x’ leads to:
>>
>> ERROR: In procedure macroexpand:
>> ERROR: identifier out of context in subform `x' of `(quote (d (a 1 2 3) (a 4
>> 5) (a 6 7 8) (a 9 10)))'
>>
>> Ideas?
>>
>> (Would be good if psyntax error reporting contained source location
>> info.)
>
> It would certainly be possible to maintain a stack of source locations
> in the expander, and modify syntax-violation to look in that stack.
>
> There is also the possibility of simply inspecting the syntax objects
> given to syntax-violation for source info, but it might not be there --
> best to have a stack to at least give approximate info.
Thanks for the hints, I’ll try to look into it.
I note that you subtly omitted to comment on the issue above. :-)
Thanks,
Ludo’.