Hermann Peifer wrote:
> Roland Rau wrote:
>> Dear all,
>>
>> I am processing rather large files (at least large in my opinion; the
>> combined number of records is about 94million) and I used successfully
>> gawk so far. At home on my GNU/Linux computer, I tried awka and it
>> gave me a speed boost of about 10-15times in comparison to gawk. At
>> work, however, I am confined to Windows and I was unable to work with
>> awka (0.7.5).
>>
>> I tried the following (intentionally simple) script called
awkascript.awk
>>
>> BEGIN {
>> mycounter = 0;
>> }
>> {
>> mycounter += 1;
>> }
>> END {
>> print mycounter;
>> }
>>
>>
>> as follows:
>> gawk -f awkascript.awk afile.txt
>> and everything worked fine.
>>
>> If I do, however,
>> awka -X -f awkascript.awk
>
> I have no idea about awka, but one observation is that you
> don't mention any data file above.
>
> Hermann
awka translates the awkscript into C. If you specify the flag -X, it
automatically generates an executable file which you can use then for
your data file.
for example (specifying the name of the output-file):
awka -X -f awkascript.awk -o awkacompiled.exe
awkacompiled.exe afile.txt
In case this caused some confusion, I hope this helped a bit.
Best,
Roland
|
