Hello Phelype,

On Fri, Jul 28, 2023 at 12:04:28PM -0300, Phelype Oleinik wrote:
> Hi Thierry,
> 
> On Fri, Jul 28, 2023 at 11:59?AM Joseph Wright
> <[log in to unmask]> wrote:
> >
> > Older LaTeX used \openin + \ifeof to test for file existence; we now use
> > \(pdf)filesize, which is expandable and reports "0" for a non-existent
> > file. Both methods should find "foo.tex" from "foo" in the same way.
> 
> A wild guess:  maybe your implementation of |\pdffilesize| doesn't try
> to guess the extension as |\input| does, thus the difference you are
> seeing.  Because as others have already said, |\input{docstrip}| works
> in both older and newer (web2c-based) LaTeX.
> 
> Sorry if this isn't useful:  can't run any code right now.

No, I think you are right: my implementation does not do acrobatics: the
filesize of file "this_file" is asked. This is "this_file" that is
opened (or not) without any acrobatics.

How can one rely on the size if the code can advertise the size that is
not the size of the file requested but of another one for which it has
added, on its own, an extension?

Best,
-- 
        Thierry Laronde <tlaronde +AT+ polynum +dot+ com>
                     http://www.kergis.com/
                    http://kertex.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C