🚀 go-pugleaf

I've just compiled transfig 2.1.5 on our RS/6000 system running
AIX 3.1 and there seems to be a problem in some string constants. It seems
that some versions of cpp must treat braces specially, so they have been
escaped in the code:

> 	cc -O   -I/usr/include -I..     -DXAW_BC -c mkfile.c
>        89 |                 fprintf(mk,"\techo \'\\strut\\psfig\{figure=%s.
>             ....................................................a..........
> a - 1506-235: (W) Illegal escape sequence { ignored.
>             ps\}\' >%s.tex\n",
>             ...a..............
> a - 1506-235: (W) Illegal escape sequence } ignored.

OK, this is harmless and the braces should be output as intended (although
it would be nice to have ifdefs around this so the compile is clean. I'm
not aware of any systems treating curly braces specially, so i'm not sure
what the default should be ... ?

> 	cc -O   -I/usr/include -I..    -DEPIC -DIBMGL -DLATEX -DPIC -DPICTEX -DPS -DPSTEX -DTEXTYL -DTPIC -DXAW_BC -c iso2tex.c
>        86 |   "{\rlap{\raise.3ex\hbox{--}}D}", /* Eth */
>             .....................a......................
> a - 1506-235: (W) Illegal escape sequence h ignored.

This one is the real problem! Of course '\r' is a carriage return, so
this will become:

	"{<cr>lap{<cr>aise.3exhbox{--}}D}"

... ouch! Fortunately, the '\h' warning brought this to my attention, but
i wonder how many other '\r's or other incorrect escape sequences are
hiding in the TeX output ? I had a quick look and i couldn't see any others.
Surely this would break (at least) iso2tex ... ? Am i missing something,
or is this a new bug, or one that just hasn't been found yet ?

rab
------------------------------------------------------------------------
Richard Brown                     | E-mail: rab@tauon.ph.unimelb.EDU.AU
School of Physics                 | Phone : +61 3 344 5081
University of Melbourne           | Fax   : +61 3 347 4783
Parkville Victoria AUSTRALIA 3052 | Telex : AA35185