CODEWITHSUNDEEP
bashshellpdf
ᴜsᴇʀ
ᴜsᴇʀ

Reputation: 1158

Divide a pdf into some pdf of n pages everyone

I had try to make a script that divide a single pdf into n parts of NUM pages everyone, but I'm not able to do do the second part of this script (a cycle):

#!/bin/sh

#Ask pdf filename
echo -n "Filename: "
read FILENAME

#Ask number of pages of every pdf
echo -n "Pages: "
read NUM

#Save pages number of the original pdf
PAG=$(pdftk $FILENAME.pdf dump_data|grep NumberOfPages| awk '{print $2}')

#Divide pdf into a pdf file for every page
pdftk $FILENAME.pdf burst output $FILENAME-%03d.pdf

Now I must do a cycle that merge NUM pages into a single pdf, for example, for NUM=3:

pdftk $FILENAME-001.pdf $FILENAME-002.pdf $FILENAME-003.pdf cat output PART-1.pdf
pdftk $FILENAME-004.pdf $FILENAME-005.pdf $FILENAME-006.pdf cat output PART-2.pdf
...
pdftk $FILENAME-(PAG-2).pdf $FILENAME-(PAG-1).pdf $FILENAME-PAG.pdf cat output PART-X.pdf

Then, for every obtained PDF (PART-1.pdf, PART-2.pdf, ..., PART-X.pdf) I must do an operation like this:

pdftops PART-X.pdf PART-X.ps

Someone can help me?

Upvotes: 1

Views: 234

Answers (2)

emzap79
emzap79

Reputation: 46

I think a while loop is the way to go here:

#!/bin/sh

...


#Save pages number of the original pdf
PAG=$(pdftk $FILENAME.pdf dump_data|grep NumberOfPages| awk '{print $2}')


# create parts 'from'-'to' given by NUM
PAG_FROM=1; PAG_TO=$NUM
while [ "$PAG_TO" -le "$PAG" ]; do
    FILE_OUT="PART-$(printf %03d $PAG_FROM)-$(printf %03d $PAG_TO)"
    pdftk $FILENAME.pdf cat $PAG_FROM-$PAG_TO output $FILE_OUT.pdf
    pdftops $FILE_OUT.pdf $FILE_OUT.ps
    PAG_FROM=$((PAG_FROM+$NUM))
    PAG_TO=$((PAG_TO+$NUM))
done

# execute pdftk on trailing pages
if [ $PAG_FROM -le $PAG ]; then
    FILE_OUT="PART-$(printf %03d $PAG_FROM)-$(printf %03d $PAG)"
    pdftk $FILENAME.pdf cat $PAG_FROM-$PAG output $FILE_OUT.pdf
    pdftops $FILE_OUT.pdf $FILE_OUT.ps
fi

I'd recommend to change your FN-variable in the beginning to FILENAME=$(basename $1), that you can easily trigger the script like so:

scriptname.sh <name of file>.pdf

Tested on a 8 pages file myself, should be working just fine!

Upvotes: 2

Jacobo de Vera
Jacobo de Vera

Reputation: 1933

There are several ways to do that. You could use your knowledge of the number of pages and then do the math to construct the gluing commands. I would choose a solution that uses the files that are sure to exist, and rely on the numbering to give you the right order:

NUM=3
pages_done=0
files=''
part=1
for file in ./$FILENAME-*
do
    files="$files $file"
    (( pages_done++ ))

    if [[ $pages_done -eq $NUM ]]
    then
        echo glue $files into part-$part
        pages_done=0
        files=''
        (( part++ ))
    fi
done


# Take care of any dangling files that did not make a group of size NUM
if [[ -n $files ]]
then
    echo glue $files into part-$part
fi

Upvotes: 0

Related Questions