Optimize using mxTextTools parser in case it's available

[GoogleCodeExporter]

Some proof-of-concept parser is already written:
http://code.google.com/p/xdebugtoolkit/source/browse/trunk/xdebugtoolkit/cgparse
rmx.py

Need to:
 * Make it memory-efficient, because currently generated tree is too sparse
because of function names, file names and some garbage;
 * Make it generate structures compatible with the current cgparser.py, and
perform it still quicker;
 * Make a wrapper that will use this parser when it's available, or the
current cgparser.py otherwise.

Original issue reported on code.google.com by Alexey.K...@gmail.com on 14 May 2009 at 6:43

[GoogleCodeExporter]

The link to the file in the issue should be considered
http://code.google.com/p/xdebugtoolkit/source/browse/trunk/xdebugtoolkit/cgparse
rmx.py?r=44

Original comment by Alexey.K...@gmail.com on 16 May 2009 at 3:45

[GoogleCodeExporter]

> * Make it memory-efficient, because currently generated tree is too
> sparse because of function names, file names and some garbage;

Done.

> * Make it generate structures compatible with the current cgparser.py,
> and perform it still quicker;

Almost done. At least it generates the same amount of RawEntry and RawCall 
objects
and performs parsing 2x times faster (3.2 sec instead of 6.4 sec on a 16mb cg 
file).
It still needs to calculate some cumulative statistics.

Pure file parsing of the same file takes 0.4 sec, so it can be usable for other 
tools
that don't need complete file parsing, e.g. for initial version of a coming 
cgfilter
tool that will act like:

grep "^summary: [0-9]\{7,\}" *

Original comment by Alexey.K...@gmail.com on 16 May 2009 at 3:51

[GoogleCodeExporter]

Original comment by Alexey.K...@gmail.com on 4 Jun 2009 at 3:04

• Added labels: Maintainability, Performance