As Dave notes, Re/m and Re/data from George James
(http://georgejames.com) are the most capable, and what I recommend when
anyone asks.
Note that if you are using GT.M, there is a very handy code
histogramming feature. Any code executed between VIEW "TRACE":1:glvn
and VIEW "TRACE":0 is histogrammed. For example:
bhaskark ~/demo 2:54pm 1044: cat trace.m
trace view "TRACE":1:"^trace($j)"
f i=1:1:2 d
.d ^hello
view "TRACE":0
zwr ^trace
q
bhaskark ~/demo 2:54pm 1045: cat hello.m
hello w "Hello, universe",!
q
bhaskark ~/demo 2:54pm 1046: mumps -run ^trace
Hello, universe
Hello, universe
^trace(6683,"hello","hello")="2:0:0:0"
^trace(6683,"hello","hello",0)="2:0:0:0"
^trace(6683,"hello","hello",1)="2:0:0:0"
^trace(6683,"trace","trace",1)="1:0:0:0"
^trace(6683,"trace","trace",1,"FOR_LOOP",1)=2
^trace(6683,"trace","trace",2)="2:0:0:0"
This is very handy, e.g., for finding dead code, debugging, performance
tuning, etc.
-- Bhaskar
-------------------------------------------------------
This SF.Net email is sponsored by Yahoo.
Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
Search APIs Find out how you can build Yahoo! directly into your own
Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
_______________________________________________
Hardhats-members mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/hardhats-members
