UNIX for Dummies Questions & Answers

HI All,
I am new to perl ,need your inputs for my better understanding.

I have 5 UNIX files ,inside which only required fields will be loaded except with junk data mentioned in double quotes .For Ex

File has below information:
Process=High
Name=Albert
Age=27
Occupation=Army
"Medlife Cont'd Capital partnership 12.36 Cont'd Sillicon valley Cont'd Global management overseas....................................................................
...........................Cont'd traditional quora etc....."

Fields starting with " " , wont be loaded as its invalid .Also has more than 199 fields which pops up inbetween correct records.When i use AWK ,its throwing error due to thresold limit.I cannot upgrade HP-UX to use gwak due to constraints.

Can PERL be used to search for pattern (Say "Medlife") and take count of each file line record ,if it exceeds 199 fields.It should delete that record and new file should be generated.

Please help me on this regard .





Constraints :
I cannot upgrade HP-UX to use gwak.Please advice how should i

Why don't you remove the records with " (double quotes)?

Code:

egrep -v '"' File | <run your program here>

@Shell Life,
Yes i tried it .But Problem is in one file ,it has more than 95lacs line records.
Even one line/two also contains with quotes" " .Which are usually ignored because of <199 fields. Though time taking to get required records excepts quotes is increasing processing speed .

I want perl code
1) to check for that pattern
2) If count is >=199 for that line record it should be deleted.

Only 3-5 cases have >199 lines in entire 95lacs record file.So if this condition is checked outcome would be pretty fast. Please guide if any other method make justice . Thanks

It will help if you display a sample of the actual data.

File
Process=Low
Name=Andrew Place=NY
Age=25 Sex=M
Occupation=Business
"Medlife Cont'd 12.23 AM transword"

Process=High Place=NE
Name=Micheal Sex=M
Age=63
Occupation=Banker
"Medlife Cont'd 16.6 AM Silicon valley Cont'd {Majority
Setback} Cont'd 13.54 AM port ........ " HAS MANY LINES

If pattern matching is "Medlife and line record
Count >=198 It should delete.

you could store $. at start of the field and continue to add the lines to the field until the end of the field or current $. - $start_of_field > 199.
-

Given the below is in file.in :-

File
Process=Low
Name=Andrew Place=NY
Age=25 Sex=M
Occupation=Business
"Medlife Cont'd 12.23 AM transword"

Process=High Place=NE
Name=Micheal Sex=M
Age=63
Occupation=Banker
"Medlife Cont'd 16.6 AM Silicon valley Cont'd {Majority
Setback} Cont'd 13.54 AM port ........ " HAS MANY LINES

If pattern matching is "Medlife and line record
Count >=198 It should delete.

The following basic perl oneliner will negate any line that matches Medlife that is above or equal to line 15 ie in the above case "If pattern matching is "Medlife and line record" will be negated. The output will be directed to file.out.

Code:

# perl -lane 'print unless (/Medlife/) && ($. >= 15)'  <file.in > file.out

Change the 15 for 198.
Not sure if this is what you are after but hope it helps.

Cheers,
Steve