Ticket #811 (closed defect: duplicate)
Enscript Fails on Win XP
| Reported by: | centgraf@… | Owned by: | jonas |
|---|---|---|---|
| Priority: | normal | Milestone: | |
| Component: | version control/browser | Version: | 0.9b2 |
| Severity: | major | Keywords: | highlighting |
| Cc: | d.mills@… |
Description
The Enscript syntax-highlighter appears to be failing on my WinXP install of Trac 0.7.1. This is the error log when attempting to view a Java file in the Browser:
6:44:25 PM Trac[file] DEBUG: Displaying file: Test.java mime-type: text/x-java
6:44:25 PM Trac[enscript] DEBUG: type: text/x-java enscript-suffix: java
6:44:25 PM Trac[mimeview] WARNING: Display failed: [Errno 2] No such file or directory: 'c:\\program files\\apache group\\apache2\\cgi-bin\\tmpw1ownv'
6:44:25 PM Trac[default] DEBUG: Mimetype: text/x-java filename: Test.java
6:44:25 PM Trac[default] DEBUG: Using default mimeviewer
Each attempt to view a file that should be handled with enscript also produces a stranded file in my cgi-bin directory called "tmpXXXXXX", containing the original (unhighlighted) source text. There is no corresponding tmp file containing the HTML version of the file. Since I can use enscript manually via Command Prompt using parameters stolen from enscript.py, I assume that there is a problem with the "NaivePopen?" call used to run enscript via Python. Note also that highlighting via SilverCity works like a charm. (I've tested with XML files.)
Versions in use:
Trac 0.7.1
GNU Enscript 1.6.3
SilverCity 0.9.5
Python 2.3.4
svn 1.1.0
Apache 2.0.50
Windows XP SP1
P.S. I've marked this as "major", since the bug leaks files into the cgi-bin dir, which may be a security hazard.


