Added yarp.pl - (Yet Another Raptor Parser) and dump_recipe_at_line.pl
Usage for yarp.pl:
perl yarp.pl <compile_log> <output_csv>
Usage for dump_recipe_at_line.pl:
perl dump_recipe_at_line.pl <compile_log> <line|line_list_file>
The idea of dump_recipe_at_line.pl is so you can take a column of line numbers from the CSV created by yarp and see what happened (without having to open the 300MB file)
with a single column:
...
226242
...
you get:
...
226242 <recipe name blah blah blah...
...
With two columns in the file, you can have some more readable output:
...
226242 m:/sf/mw/classicui/group/bld.inf
...
you get:
...
m:/sf/mw/classicui/group/bld.inf(226242): <recipe name blah blah blah...
...
Update 20090519 by MattD
Can now run parselistdirs.pl from anywhere witout lots of intermediate output files:
perl parselistdirs.pl <logs_dir> (<output_dir>)
Note that it will still put most of it's output to the console as before. The second dir is optional.
Added find_collisions.pl
Prefers to be run from the root on the build machine (as it's looking for files to compare)
It's a hacked up version of merge_csv.pl
Usage:
perl find_collisions.pl what_results.log >collision_result.txt
It creates a CSV based on the name of the file passed to it (ie what_results.log_collisions.csv)
Yes, They are all quick and dirty.
Older stuff:
F:\6.2\generated\parselistdirs.pl - creates sets from the listdirs outputs...
F:\6.2\generated\parsewhatlog.pl - creates a CSV-style log from the whatlogs.
F:\6.2\generated\merge_csv.pl - takes the output CSV from parsewhatlog.pl and merges in the output of parselistdirs.pl
in generated>
perl parselistdirs.pl ..\ >list_results.log
perl parsewhatlog.pl ..\ >what_results.log
perl merge_csv.pl what_results.log list_results.log >summary.txt
The Test dir is a directory filled with test text files...