[metapost] file naming question.
burkhard at cs.ucsd.edu
Thu Jan 21 03:18:59 CET 2010
Thanks for the note.
yes, your suggestion is operational. e.g.
mpost --jobname=data data.out creates data.0 data.1 data.2 ...
Dan Luecking wrote:
> At 08:46 AM 1/20/2010, you wrote:
> [Oops, forgot to hit reply-all. Here's what I wrote to Taco:]
>> Walt Burkhard wrote:
>>> I have used MetaPost for many years. I notice a change in how .mp
>>> files are created.
>>> old version: mpost data.out creates data.0 data.1 data.2
>>> new version: mpost data.out creates data.out.0 data.out.1
>>> data.out.2 etc ...
>>> Is there any outputtemplate to continue the "old" naming convention?
>> Unless you count
>> outputtemplate := "data.%c";
>> or trickery with substrings of jobname as a valid solution, no.
>> I am in two minds about the new behaviour: on the one hand it is
>> backward incompatible, on the other hand you *did* name the file
>> data.out, and I am not a fan of all the automatic black magic
>> that always goes on with file extensions in tex/mf/mp.
> The behavior in TeX is rather simple and has been stable
> for some time in the systems I use (TeX Live), namely
> \jobname is the filename without its last extension (if
> there is any).
> I don't see why all programs with a jobname capability can't
> do the same: follow tex's rules.
> There is precedent: On the search side, tex searches for the
> file.name.tex and then for file.name (at least in TL2009).
> We had a discussion here about the search rules for mpost
> and came to a consensus that it be essentially the same as
> tex's: file.name.mp and then file.name.
> As a work-around, surely
> mpost --jobname=data data.out
> still works?
> Daniel H. Luecking
> Department of Mathematical Sciences
> 1 University of Arkansas
> Fayetteville, Arkansas 72701
More information about the metapost