Help with Text Parsing Script

NiallC · 06-20-2006, 02:11 PM

Hello,

I have a text file that's in the following format:
------------------------------------------------
Company Name Date Type Num Due Date Amount Open Balance Balance
"Abernathy Drive, 4473"
James and Barbara -240.00 -240.00 0.00

"Total for Abernathy Drive, 4473"

"Abernathy Drive, 4481"
Jane 01/01/2003 Invoice 3297 01/01/2003\

------------------------------------------------

I only want the text between the " marks.

So my result could be:

"Abernathy Drive, 4473"

"Total for Abernathy Drive, 4473"

"Abernathy Drive, 4481"

or that without the " marks.

Any help with a script would be great. You guys have always been great in the past, and I thank you in advance for any assistance.

Thanks again,

NiallC

kaz2100 · 06-20-2006, 02:40 PM

pipe your input into

Code:

grep \"*\"

NiallC · 06-21-2006, 01:07 PM

I tried something similar to that, but it doesn't work, because the data is not split into separate lines, so I still get all the data returnded. Sorry, I should have been clearer

Thanks

NiallC

homey · 06-21-2006, 01:36 PM

I don't understand what isn't working with that command.
It gives me this output....

Code:

grep \"*\" file.txt

"Abernathy Drive, 4473"
"Total for Abernathy Drive, 4473"
"Abernathy Drive, 4481"

Here are a couple of similar commands:

Code:

grep -o '"[^"]*"' file.txt

grep \"*\" < file.txt | sed 's/^"\(.*\)"$/\1/'

NiallC · 06-24-2006, 12:38 PM

Ok,like I mentioned the text is not split on separate lines like I originally posted, it looks more like this:

Company Name Date Type Num Due Date Amount Open Balance Balance
"Abernathy Drive, 4473" James and Barbara -240.00 -240.00 0.00 "Total for Abernathy Drive, 4473" "Abernathy Drive, 4481"

Grep searches and returns the line. which includes all the other data too, unfortunately.

xhi · 06-24-2006, 01:02 PM

Code:

while(<>){
    while(m/\"(.+?)\"/ig) {print "$1\n";}
}

specify the datafile at the command line

edit> i guess i should mention that this is perl..

sundialsvcs · 06-24-2006, 03:28 PM

Check out the awk command.

Basically, an awk program consists of one or more of these:
<pattern> { <what to do when this pattern is seen> }.

That sounds quite simple, and it is, but it is also remarkably powerful and there are many good examples provided.

The two main "interesting" features of this file are that the first line is treated separately and that the data includes quoted strings. Both of these can be handled quite directly in awk.

Awk is a very useful tool for jobs that are more complex than what grep can do. The "regular expression" capabilities of both are the same. Awk includes a fully-featured programming language based loosely upon "C."