kiron@db9.CS.Concordia.CA (Kiron Bondale) (05/02/89)
I have a PostScript file, that I would like to convert to DVI (or actually if possible even to ASCII (fonts/pix are not important in this file))... I have seen DVI to PostScript converters on uunet and pawl.rpi...but is there anything that goes backwards??? thanks in advance Kiron _________________________________________________________________ Kiron D. Bondale Concordia University, Mtl., Quebec, Canada "Never mourn Black Omne" ___Habitat:kiron@sunkisd.CS.Concordia.CA -----------------------------------------------------------------
cliff@ficc.uu.net (cliff click) (05/02/89)
In article <808@sunkisd.CS.Concordia.CA>, kiron@db9.CS.Concordia.CA (Kiron Bondale) writes: > >I have a PostScript file, that I would like to convert to DVI (or actually >if possible even to ASCII (fonts/pix are not important in this file))... >I have seen DVI to PostScript converters on uunet and pawl.rpi...but is >there anything that goes backwards??? >Kiron Postscript is a general purpose programming language (ok, that's stretching it a bit :-). In general, it's possible to write impossible twisted code that will eventually produce character output. The only sure-fire way is to print the page, and read it with a scanner. If your Postscript was generated by an application, and the application's Postscript is not the twisted nightmare I've seen so many applications produce (I helped write a Postscript clone, so I've seen the worst), then it might be possible to edit the prep script (Postscript-ese for the stuff at the front of the file), to make executing the program just spit out the text instead. Of course some applications build their own fonts, and swap the ASCII meanings of the various letters around, so this is no guarentee - unless you print the page in the first place. TANSTFL. -- Cliff Click, Software Contractor at Large Business: uunet.uu.net!ficc!cliff, cliff@ficc.uu.net, +1 713 274 5368 (w). Disclaimer: lost in the vortices of nilspace... +1 713 568 3460 (h).
paul@moncam.co.uk (Paul Hudson) (05/03/89)
In article <4054@ficc.uu.net>, cliff@ficc.uu.net (cliff click) writes: > In article <808@sunkisd.CS.Concordia.CA>, kiron@db9.CS.Concordia.CA (Kiron Bondale) writes: > > > >I have a PostScript file, that I would like to convert to DVI (or actually > Postscript is a general purpose programming language (ok, that's stretching > it a bit :-). In general, it's possible to write impossible twisted code > that will eventually produce character output. The only sure-fire way is > to print the page, and read it with a scanner. This is a bit pessimistic. Sure, one would have to write a PS interpreter, but that's not too bad (I speak having done just that), or snaffle a free one (like GhostScript), but if the PS file produces output that makes sense for DVI (ie characters, mostly) the conversion wouldn't be impossible. Even arbitary PS->DVI is possible (one could alwasy use the '.' as a "pixel" and do all the scan-coversion - one TeX plotting package does this). On the other hand, I don't know of a program that does it. -- Paul Hudson MAIL: Monotype ADG, Science Park, Cambridge, CB4 4FQ, UK. PHONE: +44 (223) 420018 EMAIL: paul@moncam.co.uk, ;" FAX: +44 (223) 420911 ...!ukc!acorn!moncam!paul `"";";" "/dev/null full: please empty the bit bucket"
cliff@ficc.uu.net (cliff click) (05/05/89)
[ Long fruitless response because I hate Postscript, press 'n' quick! ] In article <170@marvin.moncam.co.uk>, paul@moncam.co.uk (Paul Hudson) writes: > In article <4054@ficc.uu.net>, cliff@ficc.uu.net (cliff click) writes: > > In article <808@sunkisd.CS.Concordia.CA>, kiron@db9.CS.Concordia.CA (Kiron Bondale) writes: > > >I have a PostScript file, that I would like to convert to DVI (or actually > > The only sure-fire way is to print the page, and read it with a scanner. > This is a bit pessimistic. Sure, one would have to write a PS > interpreter, but that's not too bad (I speak having done just that), I, too, have written a PS interpreter for a commercially available laser printer. What I found was that 90% of the PS producing applications wrote code that breaks 90% of the PS clones (5 or 6 PS clones we tested against our suite of application-produced PS code). The *interpreter* is fairly easy. Getting the font machinery, error handling, and special primitives (clippath, setscreen, settransfer, reversepath, etc) right is a nightmare. Note: the Redbook lies about the required behavior of several primitives, and is mum about much of the magic that goes on behind the scenes in the font machinery. How would you handle programs that remap the fonts? (It looks like: "100 100 moveto (a) show" but the page is rotated 90 degrees, and "a" in the current font produces "Company Foobar Logo" using a combo of Times-Roman and Times-Bold, and is actually printing a logo up the left margin. Further the BuildChar routines are recursively stacked 2 deep.) How about programs that roll their own fonts, and not using an ASCII ordering? At least one I know looks to see if he has downloaded that character in the font in his BuildChar routine, and if not he sends stuff back to the host which causes the host to send down a series of moveto/lineto commands to draw the missing character. In this case the page output is some complex interaction between the PS application and the host driver. *Bleah*. We also have a suite of programs that break (lockup and/or cold boot) the Apple Laserwriter. I wonder what would happen if we fed them to DisplayPostscript on NeXT or Sun? %-) -- Cliff Click, Software Contractor at Large Business: uunet.uu.net!ficc!cliff, cliff@ficc.uu.net, +1 713 274 5368 (w). Disclaimer: lost in the vortices of nilspace... +1 713 568 3460 (h).
greid@adobe.com (Glenn Reid) (05/05/89)
In article <4079@ficc.uu.net> cliff@ficc.uu.net (cliff click) writes: >We also have a suite of programs that break (lockup and/or cold boot) the >Apple Laserwriter. I wonder what would happen if we fed them to >DisplayPostscript on NeXT or Sun? %-) Go ahead and run them. I predict that they'll work fine. You didn't mention what model LaserWriter you ran these on, but if it is really the "LaserWriter", or even the Plus, and not one of the II models, then it isn't surprising. That original printer was discontinued in 1986, when the Plus came out, and the Plus is also now a discontinued product, as of 1987, I think. Also, the old LaserWriter used to reboot sometimes when it ran out of memory at an inopportune time. This was a bug, but it means it was very easy to write a program on purpose to cause that printer to reboot. All you had to do was to over-allocate memory. Glenn Reid Adobe Systems