Parsing PostScript files for data

[Edward]

Ok, I know you all do six impossible things before breakfast, but I
don't do breakfast, so I can't do anything much at all. Which is why
I'm posting this, videlicet:

ASP.NET
VB.NET

*My* client receives a report from *his* client in Postscript format
(NOT PDF, probably LaTeX or something, anyway with a .ps extension).
The Third Party (the PS generator) flat out refuses to release the
report in anything other than .PS format. My client wants me to parse
this report (it looks like a badly designed Word table, BTW) and
extract the data, to be reformated into a new report.

As far as I can see the PS file is just a list of co-ordinates in no
very useful order. For example, there is a field "Consignor" with (I
guess) X and Y positions, but the data that appears in the Consignor
"box" is about 200 lines away. So it's not like I could parse the
file as a big string, find the word "Consignor" and be able to rely on
the next string being the associated datum.

Badly expressed, I know, but mine own. Anyone out there solved this
problem?

MTIA

Edward
--
The reading group's reading group:
http://www.bookgroup.org.uk

[Cor]

Hi Edward,

This group is a program language group, I would try it in a more document
newsgroup like Word.

Although there are a lot of people who visits this newsgroups who know a lot
of document, I give you a better chance in such a newsgroup.

I hope you have success

Cor

[Herfried K. Wagner [MVP]]

* te********@hotmail.com (Edward) scripsit:

*My* client receives a report from *his* client in Postscript format
(NOT PDF, probably LaTeX or something, anyway with a .ps extension).
The Third Party (the PS generator) flat out refuses to release the
report in anything other than .PS format. My client wants me to parse
this report (it looks like a badly designed Word table, BTW) and
extract the data, to be reformated into a new report.

As far as I can see the PS file is just a list of co-ordinates in no
very useful order. For example, there is a field "Consignor" with (I
guess) X and Y positions, but the data that appears in the Consignor
"box" is about 200 lines away. So it's not like I could parse the
file as a big string, find the word "Consignor" and be able to rely on
the next string being the associated datum.

Did you have a look at the specification for the file formats?

<http://www.wotsit.org>

--
Herfried K. Wagner [MVP]
<http://www.mvps.org/dotnet>