[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: display path of directory of a file and bug?
From: |
Catonano |
Subject: |
Re: display path of directory of a file and bug? |
Date: |
Tue, 8 Sep 2020 12:32:05 +0200 |
Il giorno dom 6 set 2020 alle ore 17:13 <2QdxY4RzWzUUiLuE@potatochowder.com>
ha scritto:
> On 2020-09-06 at 16:03:17 +0200,
> Zelphir Kaltstahl <zelphirkaltstahl@posteo.de> wrote:
>
> > If I move a file using `mv`, a recomplation seems to be not
> > required. If I copy the file instead, Guile recompiled it. Does this
> > have something to do with inodes or creation date, which is
> > differently handled when using `cp` instead of `mv`?
>
> That was my first thought. Renaming a file (using mv) doesn't change
> the timestamp, and it's likely that guile tries to use a cached version
> but some of the pieces don't fit right.
Probably the cached version contains, hardwired, to say so, the _old_ name
of the file so when the compiled version runs, the file can't be found ?
Also if this is the case, the error message is... suboptimal ?
Maybe, being able to process the exception that gets raised, you could
extract more information about the error
The interesting line is this one
1736:10 7 (with-exception-handler _ _ #:unwind? _ #:unwind-for-type _)
If you can manage to somehow process such exception, it'd be nice if you
would share what you find ☺