voldenuit
09-24-2005, 01:58 AM
The Unix command strings is quite useful to do that kind of thing:
strings mycrashedwordfile.doc | less
will come up with the text parts of those files in a comfortably paged way.
To speed things up more, you could even do
strings *.doc | less
and if you are sure of the presence of a significant string, you can directly pick the right file, foobar being the string you know is inside:
grep foobar *.doc
The syntax here supposes you cd to the directory where all these doc files live, otherwise you'll have to give the full path.
Good luck.
----
Just saw the edit:
You can also use strings to clean up the text, but BareBones free editor TextWrangler has a very useful "zap gremlins" option.
Still, the way of the get-a-larger-hammer unix.wizard might be something like:
strings borked.doc > interview.txt
If I were you, I'd use a bulletproof text-editor for raw text, then import it for layout. Should silly doc formats be required, I'd convert to that at the last possible moment.
strings mycrashedwordfile.doc | less
will come up with the text parts of those files in a comfortably paged way.
To speed things up more, you could even do
strings *.doc | less
and if you are sure of the presence of a significant string, you can directly pick the right file, foobar being the string you know is inside:
grep foobar *.doc
The syntax here supposes you cd to the directory where all these doc files live, otherwise you'll have to give the full path.
Good luck.
----
Just saw the edit:
You can also use strings to clean up the text, but BareBones free editor TextWrangler has a very useful "zap gremlins" option.
Still, the way of the get-a-larger-hammer unix.wizard might be something like:
strings borked.doc > interview.txt
If I were you, I'd use a bulletproof text-editor for raw text, then import it for layout. Should silly doc formats be required, I'd convert to that at the last possible moment.
