# pdftotext is a command line program that converts PDF files into text files
# first get the pdftotext installer package from http://www.bluem.net/downloads/pdftotext_en/
# pdftotext code written by Glyph & Cog, LLC: http://www.glyphandcog.com/Xpdf.html

which pdftotext     # /usr/local/bin/pdftotext

pdftotext -h
pdftotext example.pdf
pdftotext example.pdf example.txt
pdftotext example.pdf -
pdftotext -layout -eol unix -enc UTF-8 /path/to/file.pdf /path/to/file.txt

snippet 345
snippet 1268
snippet 1281
snippet 182
snippet 1282
snippet 1115
snippet 5
snippet 23444

# $ cat $HOME/.bashrc

function snippet() {

   declare NL OPWD download_dir outputfile postnum title url

   if [[ "${1//[[:digit:]]/}" != "" ]]; then echo "Argument error. No positive integer: ${1}"; return 1; fi

   postnum="${1}"
   url="http://codesnippets.joyent.com/posts/show/${postnum}"
   download_dir="${HOME}/Desktop/Snippets"
   download_dir="${download_dir%/}"
   #printf "\n\e[0K\e[1;30m%s\e[0m:  %s\n\n" "download directory" "${download_dir}"
   /bin/mkdir -p "${download_dir}"
   OPWD="$PWD"
   cd "${download_dir}"
   /usr/bin/curl -L -O -s --max-time 25 "${url}" || exit 1    # download snippet
   file="${download_dir}/${url##*/}"

   title="$(/usr/bin/sed -E -n -e '/<[tT][iI][tT][lL][eE]>/{s/^.*<[tT][iI][tT][lL][eE]>(.*)<\/[tT][iI][tT][lL][eE]>.*$/\1/p;q;}' "${file}" | \
             sed -E -e 's/\[[^][:space:]]*\]//g')"    # delete [xxx] tag elements of title

   title="${title//CodeSnippets:/}"
   title="${title//\//:}"
   title="${title// /_}"
   title="${title//[[:cntrl:]]/}"
   title="${title%"${title##*[!_]}"}"   # remove trailing "_" characters

   if [[ $title == '_CodeDrive_Snippets_courtesy_of_Peter_Coopers_handy_little_app' ]] || [[ -z "$title" ]]; then
      printf "\e[0K\e[31m%s\e[0m:  %s\n" "couldn't access" "${url}"
      /bin/rm "${file}"
      return 1
   fi

   outputfile="${HOME}/Desktop/Snippets/${postnum}_${title}.txt"
   #outputfile="${outputfile//__/_}"  # uniq underscores

   printf "\n\e[0K\e[1;30m%s\e[0m:  %s\n\n" "saved as" "${outputfile}"

   /usr/bin/textutil -output "${outputfile}" -convert txt "${file}"
   /bin/rm "${file}"

   # escape backslashes
   # man bash 2>/dev/null | less -p 'Each command in a pipeline'
   #outputfile="$(printf "%q" "${outputfile}")"  # cf. help printf
   outputfile="${outputfile//\\/\\\\}"

   NL=$'\\\n'

cat <<EOF | /bin/ed -s "${outputfile}"
H
,g/Snippets is a public source code repository/1,/Snippets is a public source code repository/d
,g/You need to create an account or log in to post comments to this site//You need to create an account or log in to post comments to this site/,\$d
,g|(See related posts)$|s|.See related posts.|${NL}${NL}|
,g|^to.* by.* on .*[[:digit:]]$|s|^to\(.*\) by\(.*\) on \(.*[[:digit:]]\)$|${NL}${NL}Author:\2${NL}Date: \3${NL}URL: ${url}${NL}Tags:\1${NL}|
,g|^Comments on this post$|s|\(Comments on this post\)|${NL}\1:|
,g| posts on .* at |s|\(.* posts on .* at .*\)|${NL}\1:|
w
EOF

   cd "$OPWD"
   return 0

}