Help with grep on script 1

[dadms]

I am trying to count the number records that contain specific words that are not group together by grepping on 2 keywords in my script but having no success.

John Doe Married
Jan Doe Married
Don Doe Single

(i know the syntax is not right)
example....
#how many Doe's are Married
for file in 'blahblahblah'
do
grep -n -c "Doe"&&"Married" $file
done

 

[motoslide]

There may be other ways, but try this:

grep Doe $file|grep -c Married

 

[SamBones]

This...

awk '{print $3,$2}' $FILE | sort | uniq -c

...will give you...

   2 Married Doe
   1 Single Doe

Hope this helps.

 

[dadms]

ok, just one more thing, what do I do if the fields are comma delimited. I do I tell it that the comma separates the fields?


John,Doe,Married
Jan,Doe,Married
Don,Doe,Single

 

[motoslide]

If you stick with "grep", it couldn't care less. It is simply a pattern matching utility. It doesn't care about what order the matching paterns are in either. You may want to explore the "awk" solution SamBones provided. You'll have much more control.

 

[p5wizard]

with just one grep:

grep -c "Doe.*Married" $file

. = any character
.* = any number of occurences of any character

There are other ways of specifying REs (Regular Expressions) so see man page for grep and ed - ed is where the REs in grep originated. In fact the name grep comes from: g RE p (global i.e. whole file; RE; p -print)


HTH,

p5wizard

 

[dadms]

I have figured out that -F, separates the fields however I get told that there are too many fields for the awk command. There are over 108.

HMMM...back to the grep..

 

[dadms]

thanks for all the help. I used the following:

grep -c "Doe.*Married" $file


It didn't initially work because I had a space between the . and *

 

[dadms]

thanks for all the help. I used the following:

grep -c "Doe.*Married" $file


It didn't initially work because I had a space between the . and *