Shell Script: Reading Pdf files

funkymunky · 06-17-2004, 12:03 AM

HI!

Iv stumbled upon a treasure trove of ebooks, 5gb in all, on a local hard drive. But since they are not arranged topic-wise, i want to write a shell script that will arrange them topically.
For this i need some utility that would read a pdf file, so that i can grep for key words in the output and send the file to the appropriate folder.
Does anyone here know of such a utility? Much like a "cat" for pdf?

Thanx in anticipation

Mayank

320mb · 06-17-2004, 12:22 AM

http://man.linuxquestions.org/index....ction=0&type=2

Code:

man awk/gawk

funkymunky · 06-17-2004, 12:36 AM

thanx..thats one way..any other options?

jlliagre · 06-17-2004, 01:08 AM

pdftotext should help (http://www.foolabs.com/xpdf/about.html)

keefaz · 06-17-2004, 02:57 PM

grep can grep pdf file on another way
grep -i stringToFind docToSearch.pdf

jlliagre · 06-17-2004, 07:04 PM

Keefaz,

Pdf files are binary files with strings encoded, grep is unlikely to find anything but gibberish in them ...

funkymunky · 06-17-2004, 10:25 PM

yes, thats why i was looking for a util. pdftotext is fine for my use, it takes the starting and ending pages as arguments. i can make a temporary text file, and then grep for strings in it

thanx jlliagre

keefaz · 06-18-2004, 02:28 PM

jlliagre
I know that pdf files are binary file lol, but grep can test if a string is in a pdf file. Try yourself

grep -i "a string" yourfile.pdf

see the output, grep can do a test
[edit]
if nothing output echo $?, the test is false string not found

jlliagre · 06-18-2004, 03:34 PM

Keefaz,

I understand you happen to have successfully tested grep on a file, luckily not encoded nor compressed, but that doesn't make your suggestion correct.
Even in your case, strings embedded in pdfs can be cut in arbitrarily positions, so you are likely to miss many hits.
Moreover, eBooks, for obvious reasons, are usually compressed, so grep is not there an option.

Thanks for trying.