FM> I'm not sure that a profiler would show you the problem. If it's just
FM> timing your *code* execution, it may not indicate where *DOS* is taking
FM> a long time, such as in file I/O.

I can't remember which one I used to do the profiling, but since
I last did profiling, I discovered that Watcom's profiler is
inadequate, but that IBM's CSET is very good albeit difficult to
set up, but it generates hooks at every function call, so I get
great results, which I have provided below.  As you can see, the
bottleneck is in reading a message, which is because of the silly
way it is doing it with continual-reseeking (simplifies the code
though), so the bottleneck is not in writing the output file.
Since the next bit of optimization requires rewriting code, I'm
not keen on the idea.  Of course, if anyone wants to jump in and
volunteer, don't let me stand in your way...


E:\DEVEL\PQWK\PKT2QWKP.EXE run 11/02/95 at 07:40:10  PM
  
Executables generating events: 1 of 3
Functions generating events: 23
Threads generating events: 1
Maximum nesting depth of each thread: 
 thread:  1 depth    6
Total events: 36044
User events: 0
Number of trace buffer flushes: 0
  
Total trace time excluding overhead: 14139106 tics or 11.850 seconds
Trace overhead: 702858 tics or 0.589 seconds
  
All times in the details window are given in milliseconds.
  
Number Function Name
------ -------------
1      get_msg
2      main
3      xwriteqwkmsg
4      write_line
5      qwkwritehdr
6      bmgStrstr
7      bmgSearch
8      qwkwriteidx
9      fdt_split
10     bmgCompile
11     get_numerical_token
12     asctoi
13     assign
14     lstrip
15     rstrip
16     dt_split
17     IEEEToMSBIN
18     qwkwriteblk
19     parse_addr
20     to_delim
21     qwkfreeareas
22     stripcr
23     skip_white
  
Func    %   %    Total     Run      Active     Min       Max       Avg
Num    Run Act   Calls     Time     Time       Time      Time      Time
------ --- --- -------- --------- --------- --------- --------- ---------
1       59  64      241     6,986     7,619     0.713      41.7      29.0
2       15 100        1     1,796    11,832     1,796     1,796     1,796
3       10  20      240     1,187     2,410      2.22      26.1      4.94
4        6   6    9,312       719       719     0.003      5.15     0.077
5        3   3      480       312       312     0.556      4.04     0.650
6        3   5    1,296       301       625     0.029      5.37     0.232
7        2   2    1,296       289       289     0.014      5.46     0.223
8        1   1      240      79.3      81.1     0.029      4.32     0.330
9        1   1      240      66.4       108     0.219      3.33     0.277
10       0   0    1,104      34.7      34.7     0.022      1.46     0.031
11       0   0    1,200      23.7      41.4     0.005      5.19     0.020
12       0   0    1,200      17.6      17.6     0.011     0.232     0.015
13       0   0      192      6.57      6.57     0.031     0.041     0.034
14       0   0      242      4.79      4.79     0.003     0.060     0.020
15       0   0      242      3.22      3.22     0.004     0.913     0.013
16       0   1      240      2.41       110     0.006     0.057     0.010
17       0   0      240      1.79      1.79     0.004     0.013     0.008
18       0   0        1     0.629     0.629     0.629     0.629     0.629
19       0   0        1     0.192     0.324     0.192     0.192     0.192
20       0   0        5     0.119     0.119     0.012     0.045     0.023
21       0   0        2     0.076     0.076     0.002     0.075     0.039
22       0   0        6     0.048     0.048     0.004     0.022     0.008
23       0   0        1     0.006     0.006     0.006     0.006     0.006
@EOT:

---