IEICEの2013年03月の総合大会のProceedingsを僕好みにするシェルスクリプト

ieice2013g_hacks.sh

#!/bin/sh

# ---- ここから設定箇所 ----

# Proceedingsの中身をHDDにコピーして，その中の"Settings"ディレクトリを以下に指定してください．
ROOT_DIR="/Users/river24/Desktop/EIC2013G/Settings"

# 処理中の一時ファイルを出力する先を指定してください．
TMP_DIR="/tmp"

# ---- ここまで設定箇所 ----

# 処理中に生成される一時ファイルです．
PDF_LIST="${TMP_DIR}/pdf.list"
PAGE_LIST="${TMP_DIR}/page.list"
LINK_LIST="${TMP_DIR}/link.list"
TMP_HTML="${TMP_DIR}/tmp.html"
SESSION_LIST="${TMP_DIR}/session.list"
PAPER_LIST="${TMP_DIR}/paper.list"

cd "${ROOT_DIR}"

# このスクリプトによる変更を施す前の './html' ディレクトリを，
# './html.original' としてバックアップします．
# このスクリプトによる変更を施す前に戻したい場合は，これから復元して下さい．
if [ -e "./html.original" ]
then
    sleep 0
else
    if [ -e "./html" ]
    then
        cp -r "./html" "./html.original"
    else
        echo "'./html' directory is not found."
        exit
    fi
fi

if [ -e "./html.original" ]
then
    if [ -e "./html" ]
    then
        rm -rf "./html"
    fi
    cp -r "./html.original" "./html"
else
    echo "failed to backup './html' directory as './html.original'."
    exit
fi

if [ -e "./html" ]
then
    sleep 0
else
    echo "'./html' directory is not found."
    exit
fi

if [ -e ${PDF_LIST} ]
then
    rm "${PDF_LIST}"
fi

find "./pdf" -type f | egrep "[^\/]+\.pdf" | perl -pe 's/^.*\/([^\/]+\.pdf).*$/$1/g' | sort | uniq > ${PDF_LIST}

find "./html" -type f | egrep "[^\/]+\.html" | xargs chmod 644

find "./html" -type f | egrep "[^\/]+\.html" |
while read HTML
do
    cat ${HTML} | perl -pe 's/<\/td>/<\/td>\n/g' | perl -pe 's/<tr>/<tr>\n/g' | perl -pe 's/<\/tr>/<\/tr>\n/g' > ${TMP_HTML}
    cp ${TMP_HTML} ${HTML}
done

for PDF in `cat ${PDF_LIST}`
do
    echo ${PDF}
    if [ -e ${PAGE_LIST} ]
    then
        rm "${PAGE_LIST}"
    fi
    find "./html" -type f | egrep "[^\/]+\.html" | xargs grep "${PDF}" | perl -pe 's/^([^:]+):.*(<a [^>]*href=\"[^\"]*${PDF}\"[^>]*>.+<\/a>).*$/$1:$2/g' > ${PAGE_LIST}
    if [ -e ${PAGE_LIST} ]
    then
        if [ -e ${LINK_LIST} ]
        then
            rm "${LINK_LIST}"
        fi
        touch ${LINK_LIST}
        echo '<form><select style="width: 60px; font-size: 8px;" onChange="location.href=value;"><option value="#">▼Link</option>' >> ${LINK_LIST}
        cat ${PAGE_LIST} |
        while read LINE
        do
            FLAG=`echo "${LINE}" | md5`
            PAGE=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$1/g'`
            TEXT=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$2/g' | perl -pe 's/^.*<a [^>]*>(.+)<\/a>.*$/$1/g' | perl -pe 's/<.*?>//g'`
            echo "<option value=\"../.${PAGE}#${FLAG}\">${TEXT}</option>" >> ${LINK_LIST}
        done
        echo '</select></form>' >> ${LINK_LIST}
        cat ${PAGE_LIST} |
        while read LINE
        do
            FLAG=`echo "${LINE}" | md5`
            PAGE=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$1/g'`
            TARGET=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$2/g'`
            LINKS=`cat ${LINK_LIST}`
            cat ${PAGE} | perl -pe "s|\Q${TARGET}\E|${TARGET}<a name='${FLAG}' id='${FLAG}'>&nbsp;</a><br />${LINKS}|g" > ${TMP_HTML}
            cp ${TMP_HTML} ${PAGE}
        done
    fi
done

cat "./html/program/schedule.html" | perl -pe 's|<br />\n|<br /><KAIGYOU>|g' | perl -pe 's|<hr />|\n            <hr />\n            |g' > ${TMP_HTML}
cp ${TMP_HTML} "./html/program/schedule.html"
egrep "[A-Z]+\-[0-9]+\. " "./html/program/schedule.html" > ${SESSION_LIST}

cat ${SESSION_LIST} |
while read SESSION
do
    if [ -e $PAPER_LIST ]
    then
        rm $PAPER_LIST
    fi
    touch $PAPER_LIST
    SESSION_ID=`echo " ${SESSION}" | perl -pe 's|^.*[^A-Z]+([A-Z]+\-[0-9]+)\..*$|$1|g'`
    FLAG=`echo ${SESSION} | md5`
    SESSION_TYPE1=`echo ${SESSION} | egrep "＜[0-9]+〜[0-9]+＞" | wc -l`
    if [ ${SESSION_TYPE1} -eq 1 ]
    then
        SESSION_START=`echo " ${SESSION}" | perl -pe 's|^.*[^A-Z]+([A-Z]+\-[0-9]+)\..*＜([0-9]+)〜([0-9]+)＞.*$|$2|g'`
        SESSION_END=`echo " ${SESSION}" | perl -pe 's|^.*[^A-Z]+([A-Z]+\-[0-9]+)\..*＜([0-9]+)〜([0-9]+)＞.*$|$3|g'`
        i=${SESSION_START}
        while [ ${i} -le ${SESSION_END} ]
        do
            echo "${SESSION_ID}-${i}" >> ${PAPER_LIST}
            i=`expr ${i} + 1`
        done
    else
        SESSION_TYPE2=`echo ${SESSION} | egrep "＜[0-9]+＞" | wc -l`
        if [ ${SESSION_TYPE2} -eq 1 ]
        then
            SESSION_START=`echo ${SESSION} | perl -pe 's|^(.)| $1|g' | perl -pe 's|^.*[^A-Z]+([A-Z]+\-[0-9]+)\..*＜([0-9]+)＞.*$|$2|g'`
            echo "${SESSION_ID}-${SESSION_START}" >> ${PAPER_LIST}
        else
            echo ${SESSION_ID} >> ${PAPER_LIST}
        fi
    fi
    cat ${PAPER_LIST} |
    while read PAPER_ID
    do
        find "./html" -type f | egrep "[^\/]+\.html" | xargs egrep "<option .*>${PAPER_ID}</option>" |
        while read LINE
        do
            PAGE=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$1/g'`
            TARGET=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$2/g'`
            cat ${PAGE} | perl -pe "s|${TARGET}$|${TARGET}<option value='../../html/program/schedule.html#${FLAG}'>■スケジュール一覧で確認■</option>|g" > ${TMP_HTML}
            cp ${TMP_HTML} ${PAGE}
        done
    done
    SESSION_TOP=`cat ${PAPER_LIST} | head -n 1`
    find "./html" -type f | egrep "[^\/]+\.html" | xargs egrep "<option .*>${SESSION_TOP}</option>" | head -n 1 |
    while read LINE
    do
        TARGET=`echo "${LINE}" | perl -pe 's/^([^:]+):(.*)$/$2/g'`
        LINK=`echo ${TARGET} | perl -pe 's|^.* value="([^"]+)".*$|$1|g'`
        NEW_SESSION=`echo "${SESSION}" | perl -pe "s|${SESSION_ID}|<a href='${LINK}' name='${FLAG}' id='${FLAG}'>${SESSION_ID}</a>|g"`
        cat "./html/program/schedule.html" | perl -pe "s|${SESSION}|${NEW_SESSION}|g" > ${TMP_HTML}
        cp ${TMP_HTML} "./html/program/schedule.html"
    done
done

cat "./html/program/schedule.html" | perl -pe 's|<KAIGYOU>|\n|g' | perl -pe 's|^<td|          <td|g' | perl -pe 's|^</tr>|        </tr>|g' | perl -pe 's|^(        <td class="bumon">&nbsp;</td>)$|  $1|g' > ${TMP_HTML}
cp ${TMP_HTML} "./html/program/schedule.html"