[Ecls-list] COMPILED-FUNCTION-FILE returning logical-pathname namestrings

[Juan Jose Garcia-Ripoll]

On Wed, Mar 10, 2010 at 10:29 AM, Tobias C. Rittweiler <tcr at freebits.de>wrote:

> Is ECLSRCDIR looked at in a toplevel form in config.lsp; Is that
> toplevel form executed only during build, or also during boot of ECL?
>

At boot time


> > (ext:get-annotation 'si::decode-ihs-env 'si::location :all)
>
> (((DEFUN SI::DECODE-IHS-ENV) #P"/src/tcr/from-upstream/ecl/src/lsp/top.lsp"
>  . 31487))
>

Ok, thanks, I now know where to look for the problem.

 > > Also SI::EXPAND-SET-DOCUMENTATION does not have any annotations.
> > >
> >
> > Documentation is most of the time stored in the core (generic functions,
> > etc), except in those cases that are stored as annotations. Again, an
> > example of "lost" documentation would be needed.
>
> So annotations are a supplemental ressource? I.e. that
> EXPAND-SET-DOCUMENTATION does not have a :location annotation does not
> constitute case of lost documentation because it does have a file
> location via COMPILED-FUNCTION-FILE?
>

Sorry, Tobias, but I do not follow that paragraph and I think I also did not
follow your brief sentence. I understood that you were discussing the
documentation mechanism and how it relates to annotations. What I mean is
the following: when a definition has a documentation string, this string is
passed, at macroexpansion time, to EXPAND-SET-DOCUMENTATION, which in turn
returns the form that will store the documentation somewhere. Normally this
form will be (SET-DOCUMENTATION ...) which uses internally the annotation
mechanism.

Now, if you are not discussing documentation at all, but rather the fact
that the function SI::EXPAND-SET-DOCUMENTATION does not have ITSELF any
annotation, that should be pretty clear to you if you inspect the code.
SI::EXPAND-SET-DOCUMENTATION is in a file, helpfile.lsp, which is loaded
very early at boot time. This file is the one that sets up annotations.
Thus, when this function is defined the annotation mechanism is not set up
and thus it can not have any annotation itself.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20100310/94933e6c/attachment.html>