I have a question about xdoc and performance. I've been working with
statcvs to generate my commit log files, and it generates approximately
1000 commit_log*.xml. When xdoc starts its translation it gets slower
and slower and after approximately 400 files I get out of memory. If I
use MAVEN_OPTS=-Xmx512m I get to about 700 files before getting out of
memory. Each of the xml files themselves do not exceed 100kb in size.
It seems that xdoc is leaking memory, or the xml/xsl parsers are
leaking. Instead of iterating over the entire directory would the memory
performance be better by looping in the jelly code of each of the
files?? Anyone else experienced this????
- J
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
