2016-03-18 10:55 GMT+01:00 Philip Taylor <p.tay...@rhul.ac.uk>:
>
>
> Arthur Reutenauer wrote:
>
> > Of course they can /know/: by inspecting the log file. It contains
> > the exact transcript of the TeX run, and thus reflects all of TeX's
> > knowledge about what happened when compiling the file; as far as
> > overfull \hbox'es, etc. are concerned.
>
> Augmented by anything that the program elects to write to the log file.
> A program can elect to write an "overfull \hbox" message to the log
> file even if no overfull \hbox has actually occurred.
>
Even now it is possible to swich to \scrollmode or \nonstopmode and issue
\errmsg although no error appeared. As another test I inserted the
following:
\setbox254=\hbox to .1pt{A}
It reports an overfull hbox although the box is never used. Thus if the
code does not behave in a decent way, it can always be cheated.
>
> > Do you *understand* what I /mean/ ?
>
> Not only do I understand what you mean, I also understand that you fail
> to appreciate the difference between "know if an overfull \hbox has
> occurred " (which is *TeX's state) and "believe that an overfull \hbox
> has occurred" (which is the state of any adjunct program attempting to
> determine the state of *TeX's knowledge simply by inspecting the log file).
>
> Philip Taylor
>
>
Zdeněk Wagner
http://ttsm.icpf.cas.cz/team/wagner.shtml
http://icebearsoft.euweb.cz
>
> --------------------------------------------------
> Subscriptions, Archive, and List information, etc.:
> http://tug.org/mailman/listinfo/xetex
>
--------------------------------------------------
Subscriptions, Archive, and List information, etc.:
http://tug.org/mailman/listinfo/xetex
