Changes between Version 6 and Version 7 of CupsFaq

Timestamp:

Nov 28, 2010, 5:23:35 AM (13 years ago)

Author:

Alex Taylor

Comment:

Clarified known problems & limitations sections, added link to GS printing

CupsFaq

@@ -340 +340 @@
 <h3>1.6.  What limitations does eCups have?</h3>
 
-<p>CUPS involves a long and exceedingly complex chain of interlocking stages, many
+<p>There are several.  Some of the more prominent ones:
+<ul>
+<li>CUPS involves a long and exceedingly complex chain of interlocking stages, many
 of which consist of separate programs.  If something goes wrong somewhere along
-the line, it can be very difficult to pinpoint the failing component.
-
-<p>At the moment, installing and configuring eCups is a lengthy and sometimes
-difficult process.
-
-<p>Due to the lack of properly-integrated OS/2 tools, configuring and managing
+the line, it can be tricky to pinpoint the failing component.<p>
+
+<li>At the moment, installing and configuring CUPS is a lengthy and sometimes
+difficult process.<p>
+
+<li>Due to the lack of properly-integrated OS/2 tools, configuring and managing
 CUPS is also awkward.  Simply creating a new printer, for example, is tedious and
-often confusing.
+often confusing.<p>
+</ul>
 
 
@@ -357 +360 @@
 <ul>
 <li>Importing certain (mostly Foomatic) PPD files into ECUPS.DRV can cause the
-OS/2 spooler to crash when printing.  Until a fix is found, you should be able
-to use the Generic PostScript Printer driver.<p>
+OS/2 spooler to crash when printing.  This is probably due to both CUPS and
+Foomatic using some very convoluted PPD syntax which the OS/2 PostScript driver
+cannot handle.  Until a fix can be developed, you should be able to use the Generic 
+PostScript Printer driver (and configuring all printer-specific settings through
+the CUPS administration GUI).<p>
 
 <li>Printing to printers which use Foomatic plus a built-in GhostScript driver
 doesn't work properly just yet: the print job may abort about 99% of the way 
-through (resulting in the last few words/characters being cut off).
-
-<p>This problem is currently under investigation.  In the meantime, it does not
-seem to occur when printing files which are already in either PS or PDF format;
-therefore, one workaround is to print to a PostScript file first (by selecting
-"output to file" on the output port page of the OS/2 printer object's
-properties), and then print the file manually using <em>cupslpr</em> or
-<a href="http://pages.cs.wisc.edu/~ghost/gsview/">gsview</a>.  (This has only
-been tested with single-page documents.)
-
-<p>(Similarly, in programs like OpenOffice, which have the option of exporting
-directly to PDF, you can generate a PDF file which can be printed from
-<em>cupslpr</em> or Lucide.)
+through (resulting in the last few words/characters being cut off).  This appears
+to be caused by some kind of incompatibility in the PostScript data generated
+by the OS/2 drivers, and the validation of same performed by Foomatic.
+
+<p>This problem is currently under investigation.  In the meantime, a couple of
+workarounds are available.  First, it does not seem to occur when printing files 
+which are already in PDF format, so if your application (like OpenOffice or some
+recent Mozilla builds) supports exporting to PDF, you can do that and then print
+the PDF manually using <em>cupslpr</em> or Lucide.
+
+<p>Alternatively, you can bypass CUPS (and Foomatic) entirely, and print using
+GhostScript as an inline print processor.  This has the advantage of allowing
+basically seamless printing, at a cost of being somewhat awkward to configure 
+properly.  There are somewhat generalized instructions
+<a href="http://svn.netlabs.org/ecups/wiki/GhostScriptRasterPrinting">here</a>.
+
 </ul>