3

I have folders with hundreds of pdf and xls(x) files that were mass exported from legal e-discovery systems. The filenames in these exports correspond to bates # such as ABCD_00000001.pdf, ABCD_00000002.pdf, ... , ABCD_00002000.pdf. These mass exports include a blank pdf file for every single xls(x) file - with both having the exact same filename. E.g., ABCD_00000005.xlsx is the xlsx file that was produced in the ediscovery system and ABCD_00000005.pdf is an extraneous blank pdf file that was created in the mass export.

These extraneous .pdf files probably result from a user error on the part of the people running these mass exports, but I don't usually have control over that side of the process. So I wanted to know if any relatively straightforward way to delete these extraneous .pdf without forcing someone to go through them manually.

Improve this question
1
  • 1
    Are they .xls or .xlsx? Or could they be either? Why can't you just delete all pdf files in that directory? Are there some pdf files you would like to save? Commented Aug 10, 2018 at 16:28

2 Answers 2

Reset to default
6

Loop over the pdf files, use parameter expansion to extract the basename:

#!/bin/bash
for pdf in *.pdf ; do
    basename=${pdf%.pdf}
    if [[ -f $basename.xls || -f $basename.xlsx ]] ; then
        rm "$pdf"
    fi
done

Update: I got the logic backwards, should be fixed now.

Improve this answer
6
  • Added slightly more efficient alternative. If this folder is extensive, it could make a difference in processing time. Commented Aug 10, 2018 at 17:00
  • @Xalorous: But your "alternative" removed the files my solution tried to keep. Commented Aug 10, 2018 at 17:01
  • It's possible that I pasted the wrong one. I'll post a separate answer with the more efficient version. Commented Aug 10, 2018 at 17:04
  • I agree with your decision to reject the suggested edit — new answers should be posted as new answers, not edits.  But I believe that Xalorous got it right and you (choroba) got it backwards.  If ABCD_0000005.xlsx and ABCD_0000005.pdf both exist, your code leaves ABCD_0000005.pdf alone.  But if important.pdf exists, and there’s no corresponding spreadsheet, your code deletes important.pdf. Commented Aug 10, 2018 at 17:43
  • Actually, I missed that his logic is inverted. Taking out the ! would fix it though. The real question is what if there's a pdf without matching xls(x). What does OP want to happen then? If they want them gone, then they need to just rm -rf *.pdf. If not, then my answer. Reading the question strictly, we're only to delete pdf files with matching xls(x) files. Commented Aug 10, 2018 at 17:51
4

Loop over the .xls(x) files and remove matching pdf files.

for xls in *.xls* ; do
    /bin/rm -f "${xls%.xls*}"".pdf"
done

If there's no matching pdf it won't hurt anything.

Improve this answer
2
  • You don’t really need the ""; i.e., you could do /bin/rm -f "${xls%.xls*}.pdf".  But this looks like it should work. Commented Aug 10, 2018 at 17:42
  • I tried it both ways, the match failed without the "" in the middle. Or maybe I made a different mistake at the same time. I'm new to bash. Commented Aug 10, 2018 at 17:46

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.