convert a directory of images into a single PDF

Question

I have a directory of images:

path/to/directory/
   image01.jpg
   image02.jpg
   ...

and would like to convert it into a single PDF file:

path/to/directory.pdf

This is what I managed to code so far:

#!/bin/bash

echo Directory $1
out=$(echo $1 | sed 's|/$|.pdf|')
echo Output $out

mkdir tmp

for i in $(ls $1)
do
    # MAC hates sed with "I" (ignore case) - thanks SO for the perl solution!
    # I want to match "jpg, JPG, Jpg, ..."
    echo $1$i $(echo "tmp/$i" | perl -C -e 'use utf8;' -pe 's/jpg$/pdf/i')
    convert $1$i $(echo "tmp/$i" | perl -C -e 'use utf8;' -pe 's/jpg$/pdf/i')
done

pdftk tmp/*.pdf cat output $out
rm -rf tmp

So the idea was to convert each image into a pdf file with imagemagick, and use pdftk to merge it into a single file. Thanks to the naming of the files I don't have to bother about the ordering.

Since I'm a newbie to this I'm sure there are many refinements one can do:

• only iterate over image-files in the directory (in case there is some Readme.txt,...)
• including the extensions png, jpeg, ...
• using the trailing "/" is not elegant I admint
• etc.

Currently my main problem is, however, that there are cases where my directories and image files contain spaces in their names. The for-loop then iterates over sub-strings of the filename and I imagine that the line with convert will also fail. I have tried out some things but haven't succeeded so far and hope someone will be able to help me here. If anyone has ideas to address the issues I listed above as well I would be very glad to hear them too.

Answer 1

convert can do this in one go:

convert *.[jJ][pP][gG] output.pdf

Or to answer several of your other questions and replace your script:

#!/bin/bash
shopt -s nullglob nocaseglob
convert "$1"/*.{png,jpg,jpeg} "${1%/}.pdf"

will iterate over all the given extensions in the first argument, regardless of capitalization, and write to yourdir.pdf. It will not break on spaces.