Shell Programming and Scripting

Code:

nawk '{c+=gsub("ST[*]850[*]", "&")}END{print c}' myFile

rdcwayx
it works fine, but now is more complex, so I have a Directory with 70,000 files with differents dates, so the goal is for example get how many times the string is repeated in all files from a date ( for example the September month)
Thanks for your help.

take a look at my previous suggestion - substitute the 'myFile' with the wild-carded file names for month of September. I we don't know how your files are named, it's hard to provide a more detailed hint.
The below will 'grab' all the files and provide the total. Start with that.

Code:

nawk '{c+=gsub("ST[*]850[*]", "&")}END{print c}' /path/2/dir/with/files/*

Here is the update for your new request

Code:

awk -F "ST\\\*850\\\*" '{sum+=NF-1}END{print sum}' *sept*

ok guys, Im trying this scritp:

clear
echo "Please enter the start date in the format MMM DD, example: Jun 25"
read strtdt
echo "please enter the end date in the format MMM DD, example: Ago 26"
read enddt
touch -t ${strtdt}0000 /gentran/SI51/install/EDIS_Inbound/datefrom
touch -t ${enddt}2359 /gentran/SI51/install/EDIS_Inbound/dateto
find /gentran/SI51/install/EDIS_Inbound \( -newer /gentran/SI51/install/EDIS_Inb
ound/datefrom ! -newer /gentran/SI51/install/EDIS_Inbound/dateto \) -print | xar
gs awk '{c+=gsub("ST[*]850[*]", "&")}END{print c}'

but when the dates are small, for instead 20111024 to 20111025 it appears work fine, but when dates are bigger for instead 20111001 to 20111025 then I get several lines like: 1031
691
463
98
132
148
16
Do you know why?

xargs causes awk to run several times. Change "xargs awk" to "xargs cat |awk". This way xargs invokes cat as many times as needed then all the concatenated data is fed into one awk.

All.
Thanks so much for your help, I have gotten the result I wanted, binlib, your help was very useful also.

Refugio.