Re: Importing EPS

*To*: mathgroup at smc.vnet.net*Subject*: [mg29114] Re: Importing EPS*From*: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>*Date*: Wed, 30 May 2001 05:50:29 -0400 (EDT)*Organization*: Universitaet Leipzig*References*: <9eksi2$7qi@smc.vnet.net>*Sender*: owner-wri-mathgroup at wolfram.com

Hi, the PostScript import is *very* simple and will not work with a complicated image. The best solution seems to write a Mathematica backend for pstoedit. The power of ghostscript may be better than Mathematica's own PostScript interpreter. But you should keep in mind, that Mathematica's simple graphics (no native Bezier curves, no color gradients, no fill patterns) can't support all the stuff that can be done with a good drawing program. Regards Jens Michael wrote: > > Hello, > > I am trying to important some figures from applications such as VISIO into > Mathematica. I have exported the figures to encapsulated Postscript (EPS), > but when I go to import from Mathematica, I get the following errors: > > In[1]:= > Import["C:\\Michael\\test.eps"] > > >From In[1]:= > LinkObject::"linkv": "Argument \!\(System`Dump`localRasterValue\) in \ > \!\(LinkReadyQ[System`Dump`localRasterValue]\) is not a valid LinkObject." > > >From In[1]:= > LinkObject::"linkv": "Argument \!\(System`Dump`localRasterValue\) in \ > \!\(LinkError[System`Dump`localRasterValue]\) is not a valid LinkObject." > > >From In[1]:= > Syntax::sntup: > Unexpected end of file (probably unclosed parenthesis). > (line 12 of "C:\WINDOWS\TEMP\000001a99639") > > I tried to use TracePrint to see what was happening and noticed that some > experimental functions were being referenced, perhaps to import binary > values. However the trace was rather long (I would attach it but the file > is 6.8MB long) and I didn't really get anywhere with this. > > I have seen earlier articles suggest that the Graphics[] statement being > constructed can be salvaged from the temp file referenced above, however > there isn't enough of the figure I am trying to import for this to be > useful. From what I can see, the problem might be converting text??? > > I have version 4.0 of Mathematica, and am wondering if this has been fixed > in 4.1, or if Wolfram is aware of the problem. I have also had intermittent > problems importing other types of files, but usually in those cases retrying > it works. There is no mention of Import being experimental so I would > expect it to work!? Also, I have installed J/Link, not that it should make > any difference but who knows? > > Any ideas? > > Thanks, > > Michael Schmidt > michael at science.edu