[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [dev-serveez] current status
From: |
stefan |
Subject: |
Re: [dev-serveez] current status |
Date: |
Fri, 8 Jun 2001 17:39:31 +0200 (CEST) |
On Thu, 7 Jun 2001, Raimund 'Raimi' Jacob wrote:
> On Thu, 7 Jun 2001, Martin Grabmueller wrote:
>
> > > * Why is the (current-load-port) undefined in exception and how can
> > > this be fixed ? Since it does not work you will not have any chance
> > > finding the source of the exception in the file.
> >
> > Well, this won't be ``fixed'', because its broken by
> > design. (current-load-port) returns the current load port while a file
> > is loaded. When an error occurs and the corresponding handler is
> > invoked, the file loading has been aborted and there is no `current
> > guile --debug -s some-buggy-scheme-file.scm
> > and you will see a backtrace. Internally, enough information is
> > available to get at file name and line number information for the
> > failing expression. This is already done in CVS Guile, where this
> > information is printed in backtraces.
>
> mmmhh... can we traverse internal data structures to extract this
> information ? perhaps it would be nice to provide this information in some
> nice way ?
Made another (half working) approach to it: Saved the current-load-port in
some function and used it in the exception, worked fine. The problem is
the location of the saving: Where can I put it if none of my exported
procedures (gh_new_procedure) gets called *before* the exception occures ?
address@hidden