Dear friends,
I'm a novice Unix user and I'm trying to learn the ropes. I have a big task I have to accomplish and I'm convinced Unix can get the job done, I just haven't figured out how. I recently posted on the topic of cutting text between unique text patterns and somebody helped me a great deal. It worked great.
There are other tasks, however, that I want to accomplish.
I'm doing a content analysis of newspaper articles that I've exported in .txt format from a ProQuest database. The .rtf files look like this when I cat them in Unix.

The information contained in the 'fields' such as author, date, document type, and text word count would be immensely valuable to me. I would like to be able to extract the information (preferably only the text after the field titles (author, date, text word count) although this is not necessary. I imagine find and replace functions in Excel could be used to delete that easily. I would like to be able to get this information to some kind of excel database, probably via a .csv file.
Ultimately, I will have hundreds of these news stories to extract the information from.

Does anybody have any suggestions?

Simon

just a snippet of it:
output:

That's really helpful. For some reason, however, the gsub command doesn't delete the backslashes in the files. When I play with it, the command doesn't return an error; it seems to complete, but the files still contain the slashes.

Also, the field variables don't write exactly right. The /Author/{print "Author: "$2 }string only prints the first, not both names of the author. I had to modify it to read "$2, $3" to get first and last names. Any clue why that might be?

Thanks again though, this is what I was looking for!
Simon

gsub() modifies what you passed to it "in place". eg gsub("","",$1);print $1


it should be no problem if the field separator is specified as ":" . show your code if you want.

Hi, On the first point, this is the code I was playing with from the command line, based on your suggestion.
awk 'BEGIN { gsub(/\\/,"")}' 03152000.rtf

Did you put the gsub command in there just to get rid of the backslashes? Or is it related to the field extraction process?

Regarding the second point:
Here is the text, copied straight from the .txt file I'm trying to extract information from.

Author(s): Ashley Geddes, Provincial Affairs Writer

Here is the code (entered in the command line) I got from you and it's output.

awk 'Begin{FS=":"} ; /Author/{print $2 }' 03152000.rtf > Author.txt

Ashley

Here is the modified way I wrote the code and its result

awk 'Begin{FS=":"} ; /Author/{print $2,$3 }' 03152000.rtf > Author.txt

Ashley Geddes,

the same problem occurred with the field "Document Types" I had to change around the fields as you wrote them to get the result.

Obviously I'm not that concerned, because it seems to be working. However I am very curious because I'd like to know how the thing works. Thanks for your help!

use BEGIN not Begin. and yes ,the gsub is there to remove slash