Tax Time PDF collector

pdf-split-and-collect.fish

set tmp_dir (mktemp -d) # create a temporary directory

for file in (ls Account*.pdf) # match all files to be included
  pdfseparate $file $tmp_dir/$file--%03d.pdf # split pdfs page wise into tmp dir
end

for file in (ls $tmp_dir/*.pdf) # iterate through all split pages 
  if pdftotext $file - | grep 'EUROPEAN' # check if they contain our search string
    set matches $matches $file # append to our match list
  end
end

pdfunite $matches all.pdf # take all the matched pages and merge them into all.pdf

test -n "$tmp_dir"; and rm -r $tmp_dir # make sure tmp dir exists and remove it