igor@38: #!/bin/bash
igor@38: 
igor@55: cat <<EOF
igor@55: Please, use the script no more.
igor@55: You can execute new-words.py directly.
igor@55: EOF
igor@55: exit 1
igor@55: 
igor@38: show_usage()
igor@38: {
igor@38: cat <<HELP > /dev/stderr
igor@38: 
igor@38: USAGE: 
igor@38: 
igor@38:     new-words [ -l lang ] [ -s ] [ ARG ] 
igor@38: 
igor@38: SWITCHES: 
igor@38: 
igor@38:     -h          print this screen
igor@38:     -c          show compressed wordlist: one word per group
igor@50:     -f file     show only words related to the words in the file
igor@45:     -G          turn off word grouping
igor@38:     -k          put higher words that are similar to the known words (only for English)
igor@38:     -l lang     override language settings
igor@38:     -n          non-interactive mode (don't run vi)
igor@38:     -N          turn off known words filtering
igor@38:     -a          don't add marks (and don't save marks added by user)
igor@38:     -p pages    work with specified pages only (pages = start-stop/total )
igor@38:     -s          show the text statistics (percentage of known words and so on) and exit
igor@38:     -S          show your vocabulary statistics (number of words and word groups)
igor@38:     -t tag      tag known words with tag
igor@38:     -T          show list of active tags
igor@38:     -m tag      merge the words tagged with "tag" into the main vocabulary
igor@38:     -M          merge the words tagged with any tag into the main vocabulary 
igor@48:     -d tag      delete subvocabulary for the "tag"
igor@48:     -r RANGE    show only first RANGE words
igor@48:     -R RANGE    show only words lower than RANGE percent
igor@38:     -2 -3       find 2 and 3 words' sequences
igor@38: 
igor@38: The language of the text can be specified also
igor@38: by name of the program new-words (correspondent link must be created before).
igor@38: For example, these calls are equivalent:
igor@38: 
igor@38:     de-words URL
igor@38:     new-words -l de URL
igor@38: 
igor@38: HELP
igor@38: }
igor@38: 
igor@38: if [ "$1" = "-h" ]
igor@38: then
igor@38:     show_usage
igor@38:     exit 0
igor@38: fi
igor@38: 
igor@38: NEW_WORDS_PY=/home/igor/hg/new-words/new-words.py
igor@38: WORK_DIR=~/.new-words/
igor@38: editor=${EDITOR:-vim}
igor@38: 
igor@38: # language detection
igor@38: 
igor@38: LANGUAGE=en
igor@38: my_name="`echo $0 | sed s@.*/@@ | sed s/-.*// `"
igor@38: for arg
igor@38: do
igor@38:     if echo "$arg" | grep -q http://...wikipedia.org/wiki/
igor@38:     then
igor@38:     LANGUAGE="`echo $arg | sed s@http://@@ | sed s@.wikipedia.*@@`"
igor@38:     fi
igor@38: done
igor@38: [ "${my_name}" = "new" ] || LANGUAGE="$my_name"
igor@38: 
igor@38: #----------------------------------------------------
igor@38: # command line options processing
igor@38: 
igor@38: STAT_ONLY=NO
igor@38: NEED_TO_USE_VOCABULARY_WHEN_SORT=NO
igor@38: DONT_ADD_MARKS=NO
igor@38: NON_INTERACTIVE_MODE=NO
igor@38: PART_TO_PROCESS=''
igor@38: GROUP_WORDS_BY_THREE=NO
igor@38: GROUP_WORDS_BY_TWO=NO
igor@38: TAG_NAME=''
igor@38: MERGE_THIS_TAGS=''
igor@38: TAGS_LIST_ONLY=NO
igor@38: MERGE_TAGGED_WORDS=NO
igor@38: MERGE_ALL_TAGGED=NO
igor@38: DONT_ADD_MARKLINES=NO
igor@38: FILTER_WORDS=YES
igor@38: SHOW_VOC_STAT=NO
igor@38: COMPRESSED_WORDLIST=NO
igor@45: WORDS_GROUPING=YES
igor@50: ALLOWED_WORDS_FILENAME=''
igor@50: while getopts Gcf:l:sSkanNp:t:Tm:Md:r:R:23 opt
igor@38: do
igor@38:     case "$opt" in
igor@38:       c)  COMPRESSED_WORDLIST=YES;;
igor@50:       f)  ALLOWED_WORDS_FILENAME="$OPTARG";;
igor@45:       G)  WORDS_GROUPING=NO;;
igor@38:       s)  STAT_ONLY=YES;;
igor@38:       S)  SHOW_VOC_STAT=YES;;
igor@38:       k)  NEED_TO_USE_VOCABULARY_WHEN_SORT=YES;;
igor@38:       l)  LANGUAGE="$OPTARG";;
igor@38:       a)  DONT_ADD_MARKS=YES;;
igor@38:       n)  NON_INTERACTIVE_MODE=YES;;
igor@38:       N)  FILTER_WORDS=NO;;
igor@38:       p)  PART_TO_PROCESS="$OPTARG";;
igor@38:       t)  TAG_NAME="$OPTARG";;
igor@38:       T)  TAGS_LIST_ONLY="YES";;
igor@38:       m)  DONT_ADD_MARKLINES="YES"; MERGE_TAGGED_WORDS="YES"; MERGE_THIS_TAGS="$TAG_NAME $OPTARG";;
igor@38:       M)  DONT_ADD_MARKLINES="YES"; MERGE_ALL_TAGGED="YES";;
igor@48:       d)  REMOVE_TAG="YES"; TAG_NAME="$TAG_NAME $OPTARG";;
igor@48:       r)  SHOW_RANGE="$OPTARG";;
igor@48:       R)  SHOW_RANGE_PERCENTAGE="$OPTARG";;
igor@38:       2)  GROUP_WORDS_BY_TWO=YES;;
igor@38:       3)  GROUP_WORDS_BY_THREE=YES;;
igor@38:       \?)       # unknown flag
igor@38:           show_usage
igor@38:           exit 1;;
igor@38:     esac
igor@38: done
igor@38: shift `expr $OPTIND - 1`
igor@38: 
igor@38: if [ "$1" = "-l" ]
igor@38: then
igor@38:     LANGUAGE="$2"
igor@38:     shift 2
igor@38: fi
igor@38: 
igor@38: VOCABULARY=${LANGUAGE}.txt
igor@38: NOTES_FILE=notes-${LANGUAGE}.txt
igor@38: 
igor@38: if [ "${SHOW_VOC_STAT}" = "YES" ]
igor@38: then
igor@38:   $0 -l "${LANGUAGE}" -N -n ${WORK_DIR}/${VOCABULARY} | head -1 | awk '{print $5}' | tr -d "<>"
igor@38:   exit 0
igor@38: fi
igor@38: 
igor@40: get_words_group_words_add_stat()
igor@38: {
igor@54:     [ "$PART_TO_PROCESS" == "" ] || PART_TO_PROCESS="-p $PART_TO_PROCESS"
igor@54:     [ "$ALLOWED_WORDS_FILENAME" = "" ] || ALLOWED_WORDS_FILENAME="-f $ALLOWED_WORDS_FILENAME"
igor@55:     [ "$SHOW_RANGE" = "" ] || SHOW_RANGE="-r $SHOW_RANGE"
igor@54:     [ "$SHOW_RANGE_PERCENTAGE" = "" ] || SHOW_RANGE_PERCENTAGE="-R $SHOW_RANGE_PERCENTAGE"
igor@54:     [ "$NON_INTERACTIVE_MODE" = YES ] && non_interactive="-n"
igor@54:     [ "$STAT_ONLY" = YES ] && stat_only="-s"
igor@54:     [ "$COMPRESSED_WORDLIST" = YES ] && compressed_wordlist="-c"
igor@54:     [ "$FILTER_WORDS" = NO ] && filter_words="-N"
igor@54:     [ "$GROUP_WORDS_BY_TWO" = YES ] && group_words_by_two="-2"
igor@54:     [ "$GROUP_WORDS_BY_THREE" = YES ] && group_words_by_three="-3"
igor@55:     [ "$WORDS_GROUPING" = NO ] && words_grouping="-G"
igor@54: 
igor@54:     $NEW_WORDS_PY -l "$LANGUAGE" \
igor@55:     $SHOW_RANGE \
igor@54:     $SHOW_RANGE_PERCENTAGE \
igor@54:     $PART_TO_PROCESS \
igor@54:     $ALLOWED_WORDS_FILENAME \
igor@54:     $non_interactive \
igor@54:     $stat_only \
igor@54:     $compressed_wordlist \
igor@54:     $filter_words \
igor@54:     $group_words_by_two \
igor@54:     $group_words_by_three \
igor@55:     $words_grouping \
igor@54:     -X get_words_group_words_add_stat "$1"
igor@38: }
igor@38: 
igor@38: if [ "$TAGS_LIST_ONLY" = "YES" ] 
igor@38: then
igor@38:     cd "${WORK_DIR}"
igor@38:     echo ${LANGUAGE}_*.txt | tr ' ' '\n'  | grep -v '*' | sed 's/[^_]*_//;s/.txt$//'
igor@38:     exit 0
igor@38: fi
igor@38: 
igor@38: tag_file_name()
igor@38: {
igor@38:     echo "${LANGUAGE}_${1}.txt"
igor@38: }
igor@38: 
igor@38: if [ "$REMOVE_TAG" = "YES" ]
igor@38: then
igor@38:     cd "${WORK_DIR}"
igor@38:     for i in $TAG_NAME 
igor@38:     do
igor@38:         echo "$TAGNAME" | grep -q '[/*?]' && continue
igor@38:         f="`tag_file_name $i`"
igor@38:         if [ -e "$f" ] 
igor@38:         then
igor@38:             rm -f "$f" && echo Tag "'$i'" removed
igor@38:         else
igor@38:             echo Unknown tag "'$i'"
igor@38:         fi
igor@38:     done
igor@38:     exit 0
igor@38: fi
igor@38: 
igor@54: get_words_group_words_add_stat "$1"
igor@38: 
igor@54: #mkdir -p $WORK_DIR
igor@54: #oldpwd="$PWD"
igor@54: #cd $WORK_DIR
igor@54: #if [ "$MERGE_TAGGED_WORDS" = "YES" ]
igor@54: #then
igor@54: #    VOC_FILES=''
igor@54: #    for i in $MERGE_THIS_TAGS
igor@54: #    do
igor@54: #        f=`tag_file_name $i`
igor@54: #        [ -e "$f" ] && VOC_FILES="${VOC_FILES} $f"
igor@54: #    done
igor@54: #    if [ -z "$VOC_FILES" ]
igor@54: #    then 
igor@54: #        echo Unknown tags: $MERGE_THIS_TAGS > /dev/stderr
igor@54: #    else
igor@54: #        cat $VOC_FILES
igor@54: #    fi
igor@54: #elif [ "$MERGE_ALL_TAGGED" = "YES" ]
igor@54: #then
igor@54: #    cat ${LANGUAGE}_*.txt
igor@54: #else 
igor@54: #    cat
igor@54: #fi 
igor@38: 
igor@38: