[pylucene-dev] Buggy LuceneBenchmark.py

[Jarek Zgoda]

LuceneBenchmark.py has bugs:

$ python LuceneBenchmark.py
LuceneBenchmark.py:140: RuntimeWarning: tempnam is a potential security 
risk to your program
   self.workDir = os.tempnam(None, ".lucene_benchmark")
  - setup in /tmp/.lucevRkCIh
     retrieving 
http://kdd.ics.uci.edu/databases/20newsgroups/mini_newsgroups.tar.gz to 
mini_newsgroups.tar.gz
     reading 1860687 bytes..........
* Unpacking reference collection: 
mini_newsgroups.tar.gz++++++++++++++++++++++++++++++++++++++++
  - 2000 source files in 20 directories.
* Creating jumbo files...
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/alt.atheism
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/comp.graphics
  - creating jumbo files in 
/tmp/.lucevRkCIh/jumbo/jumbo/comp.os.ms-windows.misc
  - creating jumbo files in 
/tmp/.lucevRkCIh/jumbo/jumbo/comp.sys.ibm.pc.hardware
  - creating jumbo files in 
/tmp/.lucevRkCIh/jumbo/jumbo/comp.sys.mac.hardware
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/comp.windows.x
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/misc.forsale
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/rec.autos
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/rec.motorcycles
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/rec.sport.baseball
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/rec.sport.hockey
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/sci.crypt
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/sci.electronics
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/sci.med
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/sci.space
  - creating jumbo files in 
/tmp/.lucevRkCIh/jumbo/jumbo/soc.religion.christian
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/talk.politics.guns
  - creating jumbo files in 
/tmp/.lucevRkCIh/jumbo/jumbo/talk.politics.mideast
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/talk.politics.misc
  - creating jumbo files in /tmp/.lucevRkCIh/jumbo/jumbo/talk.religion.misc
  - 3100 jumbo files in 20 directories.
Traceback (most recent call last):
   File "/root/pylucene/PyLucene-2.0.0-5/test/LuceneBenchmark.py", line 
993, in ?
     LuceneBenchmark.main(sys.argv[1:])
   File "/root/pylucene/PyLucene-2.0.0-5/test/LuceneBenchmark.py", line 
462, in main
     qs = cls.createQueries(LuceneBenchmark.queries, a)
   File "/root/pylucene/PyLucene-2.0.0-5/test/LuceneBenchmark.py", line 
505, in createQueries
     q = qp.parseQuery(_qs)
AttributeError: 'PyLucene.QueryParser' object has no attribute 'parseQuery'

Changing this to qp.parse(_qs) makes the program to pass few lines 
further then fails:

Traceback (most recent call last):
   File 
"/home/jarek/install/PyLucene-src-2.1.0-1/test/LuceneBenchmark.py", line 
993, in ?
     LuceneBenchmark.main(sys.argv[1:])
   File 
"/home/jarek/install/PyLucene-src-2.1.0-1/test/LuceneBenchmark.py", line 
477, in main
     bench.runBenchmark(param)
   File 
"/home/jarek/install/PyLucene-src-2.1.0-1/test/LuceneBenchmark.py", line 
374, in runBenchmark    iw.mergeFactor = params.mergeFactor
AttributeError: 'PyLucene.IndexWriter' object has no attribute 'mergeFactor'

Is the LuceneBenchmark.py program of any importance? If yes, I'll try to 
fix it today, but if not, I'll propose removing it ffrom distribution.

-- 
Jarek Zgoda

"We read Knuth so you don't have to."