mralexgray · September 11, 2011 12:49 · Jul 10, 2011 · Jul 10, 2011
diff --git a/bash-stack.bash b/bash-stack.bash
@@ -0,0 1,40 @@
 #!/usr/bin/env bash
 
 # Bash Stack
 
 set -o errtrace
 set -o errexit
 set -o nounset
 
 declare -a stack
 
 stack_push () {
  [ $# -eq 1 ] || _stack_err "stack_push takes one argument" || return 1
  stack[${#stack[@]}]="$1"
 }
 stack_pop () {
  index=$(_stack_index) || _stack_err "index out of range" || return 1
  unset stack[$index]
 }
 stack_get () {
  [ $# -ge 1 ] && item="$1" || item="0"
  index=$(_stack_index $item) || _stack_err "index out of range" || return 1
  echo ${stack[$index]}
 }
 stack_list () {
  echo ${stack[@]}
 }
 stack_size () {
  echo ${#stack[@]}
 }
 
 # internal:
 _stack_index () {
  [ $# -ge 1 ] && index="$1" || index="0" || true
  [ "$index" -ge 0 ] && [ "$index" -lt ${#stack[@]} ] || return 1
  expr ${#stack[@]} - "$index" - 1 || true
 }
 _stack_err () {
  echo "Stack error: $@" 1>&2
  return 1
 }
diff --git a/bash-xml.bash b/bash-xml.bash
@@ -0,0 1,39 @@
 #!/usr/bin/env bash
 
 # Bash XML parser (SAX)
 #
 # EXAMPLE: echo '<foo><bar>hello world</bar></foo>' | ./bash-xml.bash 'echo START:' 'echo END:' 'echo TEXT:'
 #
 
 set -o errtrace
 set -o errexit
 set -o nounset
 
 bash_xml_sax_parse () {
  handle_element_start="$1"
  handle_element_end="$2"
  handle_characters="$3"
  # assumes each line contains one element
  cat "/dev/stdin" | bash_xml_split | while read line; do
  case "$line" in
  "<?"*) ;;
  "</"*) [ -z "$handle_element_end" ] || $handle_element_end "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
  "<"*"/>") [ -z "$handle_element_start" ] || $handle_element_start "$line" "$(expr "$line" : '</*\([^ />]*\)')"
  [ -z "$handle_element_end" ] || $handle_element_end "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
  "<"*) [ -z "$handle_element_start" ] || $handle_element_start "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
  *) [ -z "$handle_characters" ] || $handle_characters "$line" ;;
  esac
  done
 }
 
 # splits an XML document into a stream of lines containing one element each and removes blanks
 # TODO: make this more robust
 bash_xml_split () {
  sed -e 's/</\
 </g' -e 's/>/>\
 /g' | sed -e '/^ *$/d'
 }
 
 if [ "$0" == "$BASH_SOURCE" ]; then
  bash_xml_split | bash_xml_sax_parse "$@"
 fi
diff --git a/gmail.bash b/gmail.bash
@@ -0,0 1,49 @@
 #!/usr/bin/env bash
 
 # Gmail inbox feed parser
 #
 # EXAMPLE: ./gmail.bash GMAIL_USER
 #
 # GMAIL_USER can include username and optionally a password
 
 set -o errtrace
 set -o errexit
 set -o nounset
 
 parent=$(dirname -- "$BASH_SOURCE")
 source "$parent/bash-xml.bash"
 source "$parent/bash-stack.bash"
 
 gmail_parse_feed () {
  bash_xml_sax_parse gmail_parse_feed_element_start gmail_parse_feed_element_end gmail_parse_feed_characters
 }
 gmail_parse_feed_element_start () {
  [ "$2" != "entry" ] || (gmail_entry_name="" && gmail_entry_title="" && gmail_entry_date="")
  stack_push "$2"
 }
 gmail_parse_feed_element_end () {
  [ "$2" != "entry" ] || printf " s | %-25s | %s\n" "$gmail_entry_date" "$gmail_entry_name" "$gmail_entry_title"
  stack_pop
 }
 gmail_parse_feed_characters () {
  case "$(stack_list)" in
  "feed entry title") gmail_entry_title="$1" ;;
  "feed entry issued") gmail_entry_date="$1" ;;
  "feed entry author name") gmail_entry_name="$1" ;;
  esac
 }
 
 gmail_fetch_feed () {
  curl -u "$1" --silent 'https://mail.google.com/mail/feed/atom'
 }
 
 gmail_inbox () {
  if [ $# -gt 0 ]; then
  GMAIL_USER="$1"
  fi
  gmail_fetch_feed "$GMAIL_USER" | gmail_parse_feed
 }
 
 if [ "$0" == "$BASH_SOURCE" ]; then
  gmail_inbox "$@"
 fi
diff --git a/xml-prettify.bash b/xml-prettify.bash
@@ -0,0 1,43 @@
 #!/usr/bin/env bash
 
 # XML prettifier
 #
 # EXAMPLE: echo '<foo><bar>hello world</bar></foo>' | ./xml-prettify.bash
 #
 
 set -o errtrace
 set -o errexit
 set -o nounset
 
 parent=$(dirname -- "$BASH_SOURCE")
 source "$parent/bash-xml.bash"
 source "$parent/bash-stack.bash"
 
 xml_prettify_indent="2"
 xml_prettify_print () {
  printf "%*s%s\n" $(expr $(stack_size) \* $xml_prettify_indent) "" "$1"
 }
 
 xml_prettify () {
  bash_xml_sax_parse xml_prettify_start xml_prettify_end xml_prettify_characters
 }
 
 xml_prettify_start () {
  case "$1" in
  *"/>") ;; # HACK: skip self terminating tags
  *) xml_prettify_print "$1" ;;
  esac
  stack_push "$2"
 }
 xml_prettify_end () {
  stack_pop
  xml_prettify_print "$1"
 }
 xml_prettify_characters () {
  xml_prettify_print "$1"
  return 0
 }
 
 if [ "$0" == "$BASH_SOURCE" ]; then
  xml_prettify
 fi
diff --git a/xml.bash b/xml.bash
@@ -1,121 0,0 @@
-#!/usr/bin/env bash
-
-set -o errtrace 
-set -o errexit
-set -o nounset
-
-
-# BASH STACK:
-
-declare -a stack
-stack_push () {
- [ $# -eq 1 ] || stack_err "stack_push takes one argument" || return 1
- stack[${#stack[@]}]="$1"
-}
-stack_pop () {
- index=$(stack_index) || stack_err "index out of range" || return 1
- unset stack[$index]
-}
-stack_item () {
- [ $# -ge 1 ] && item="$1" || item="0"
- index=$(stack_index $item) || stack_err "index out of range" || return 1
- echo ${stack[$index]}
-}
-stack_index () {
- [ $# -ge 1 ] && index="$1" || index="0" || true
- [ "$index" -ge 0 ] && [ "$index" -lt ${#stack[@]} ] || return 1
- expr ${#stack[@]} - "$index" - 1 || true
-}
-stack_count () {
- echo ${#stack[@]}
-}
-stack_list () {
- echo ${stack[@]}
-}
-stack_err () {
- echo "Stack error: $@" 1>&2
- return 1
-}
-
-
-# BASH XML PARSER:
-
-bash_sax_parse () {
- handle_element_start="$1"
- handle_element_end="$2"
- handle_characters="$3"
- # assumes each line contains one element
- cat "/dev/stdin" | bash_xml_split | while read line; do
- case "$line" in
- "<?"*) ;;
- "</"*) [ -z "$handle_element_end" ] || "$handle_element_end" "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
- "<"*"/>") [ -z "$handle_element_start" ] || "$handle_element_start" "$line" "$(expr "$line" : '</*\([^ />]*\)')"
- [ -z "$handle_element_end" ] || "$handle_element_end" "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
- "<"*) [ -z "$handle_element_start" ] || "$handle_element_start" "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
- *) [ -z "$handle_characters" ] || "$handle_characters" "$line" ;;
- esac
- done
-}
-
-# splits an XML document into a stream of lines containing one element each and removes blanks
-# TODO: make this more robust
-bash_xml_split () {
- sed -e 's/</\
-</g' -e 's/>/>\
-/g' | sed -e '/^ *$/d'
-}
-
-
-# EXAMPLE: XML prettifier
-
-xml_prettify () {
- bash_sax_parse xml_prettify_start xml_prettify_end xml_prettify_characters
-}
-xml_prettify_indent="2"
-xml_prettify_print () {
- printf "%*s%s\n" $(expr $(stack_count) \* $xml_prettify_indent) "" "$1"
-}
-xml_prettify_start () {
- case "$1" in
- *"/>") ;; # HACK: skip self terminating tags
- *) xml_prettify_print "$1" ;;
- esac
- stack_push "$2"
-}
-xml_prettify_end () {
- stack_pop
- xml_prettify_print "$1"
-}
-xml_prettify_characters () {
- xml_prettify_print "$1"
- return 0
-}
-
-
-# EXAMPLE: Gmail feed parser
-
-gmail_parse_feed () {
- bash_sax_parse gmail_parse_feed_element_start gmail_parse_feed_element_end gmail_parse_feed_characters
-}
-gmail_parse_feed_element_start () {
- [ "$2" != "entry" ] || (gmail_entry_name="" && gmail_entry_title="" && gmail_entry_date="")
- stack_push "$2"
-}
-gmail_parse_feed_element_end () {
- [ "$2" != "entry" ] || printf " s | %-25s | %s\n" "$gmail_entry_date" "$gmail_entry_name" "$gmail_entry_title"
- stack_pop
-}
-gmail_parse_feed_characters () {
- case "$(stack_list)" in
- "feed entry title") gmail_entry_title="$1" ;;
- "feed entry issued") gmail_entry_date="$1" ;;
- "feed entry author name") gmail_entry_name="$1" ;;
- esac
-}
-
-gmail_user='USERNAME'
-gmail_fetch_feed () {
- curl -u "$gmail_user" --silent 'https://mail.google.com/mail/feed/atom'
-}
-
-gmail_fetch_feed | gmail_parse_feed

diff --git a/xml.bash b/xml.bash
@@ -0,0 1,121 @@
 #!/usr/bin/env bash
 
 set -o errtrace 
 set -o errexit
 set -o nounset
 
 
 # BASH STACK:
 
 declare -a stack
 stack_push () {
  [ $# -eq 1 ] || stack_err "stack_push takes one argument" || return 1
  stack[${#stack[@]}]="$1"
 }
 stack_pop () {
  index=$(stack_index) || stack_err "index out of range" || return 1
  unset stack[$index]
 }
 stack_item () {
  [ $# -ge 1 ] && item="$1" || item="0"
  index=$(stack_index $item) || stack_err "index out of range" || return 1
  echo ${stack[$index]}
 }
 stack_index () {
  [ $# -ge 1 ] && index="$1" || index="0" || true
  [ "$index" -ge 0 ] && [ "$index" -lt ${#stack[@]} ] || return 1
  expr ${#stack[@]} - "$index" - 1 || true
 }
 stack_count () {
  echo ${#stack[@]}
 }
 stack_list () {
  echo ${stack[@]}
 }
 stack_err () {
  echo "Stack error: $@" 1>&2
  return 1
 }
 
 
 # BASH XML PARSER:
 
 bash_sax_parse () {
  handle_element_start="$1"
  handle_element_end="$2"
  handle_characters="$3"
  # assumes each line contains one element
  cat "/dev/stdin" | bash_xml_split | while read line; do
  case "$line" in
  "<?"*) ;;
  "</"*) [ -z "$handle_element_end" ] || "$handle_element_end" "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
  "<"*"/>") [ -z "$handle_element_start" ] || "$handle_element_start" "$line" "$(expr "$line" : '</*\([^ />]*\)')"
  [ -z "$handle_element_end" ] || "$handle_element_end" "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
  "<"*) [ -z "$handle_element_start" ] || "$handle_element_start" "$line" "$(expr "$line" : '</*\([^ />]*\)')" ;;
  *) [ -z "$handle_characters" ] || "$handle_characters" "$line" ;;
  esac
  done
 }
 
 # splits an XML document into a stream of lines containing one element each and removes blanks
 # TODO: make this more robust
 bash_xml_split () {
  sed -e 's/</\
 </g' -e 's/>/>\
 /g' | sed -e '/^ *$/d'
 }
 
 
 # EXAMPLE: XML prettifier
 
 xml_prettify () {
  bash_sax_parse xml_prettify_start xml_prettify_end xml_prettify_characters
 }
 xml_prettify_indent="2"
 xml_prettify_print () {
  printf "%*s%s\n" $(expr $(stack_count) \* $xml_prettify_indent) "" "$1"
 }
 xml_prettify_start () {
  case "$1" in
  *"/>") ;; # HACK: skip self terminating tags
  *) xml_prettify_print "$1" ;;
  esac
  stack_push "$2"
 }
 xml_prettify_end () {
  stack_pop
  xml_prettify_print "$1"
 }
 xml_prettify_characters () {
  xml_prettify_print "$1"
  return 0
 }
 
 
 # EXAMPLE: Gmail feed parser
 
 gmail_parse_feed () {
  bash_sax_parse gmail_parse_feed_element_start gmail_parse_feed_element_end gmail_parse_feed_characters
 }
 gmail_parse_feed_element_start () {
  [ "$2" != "entry" ] || (gmail_entry_name="" && gmail_entry_title="" && gmail_entry_date="")
  stack_push "$2"
 }
 gmail_parse_feed_element_end () {
  [ "$2" != "entry" ] || printf " s | %-25s | %s\n" "$gmail_entry_date" "$gmail_entry_name" "$gmail_entry_title"
  stack_pop
 }
 gmail_parse_feed_characters () {
  case "$(stack_list)" in
  "feed entry title") gmail_entry_title="$1" ;;
  "feed entry issued") gmail_entry_date="$1" ;;
  "feed entry author name") gmail_entry_name="$1" ;;
  esac
 }
 
 gmail_user='USERNAME'
 gmail_fetch_feed () {
  curl -u "$gmail_user" --silent 'https://mail.google.com/mail/feed/atom'
 }
 
 gmail_fetch_feed | gmail_parse_feed