#!/bin/bash # preproc_staff.sh # Preprocessor for staff keys in .md topic file # Replaces key fgabel by "Fabian Gabel, M.Sc." according to Webpage content # If Build_dir is passed, titles are appended to corresponding files # Note: mat.tuhh.de uses WINDOWS-1252 character encoding # processing command line arguments if [ $# -lt 2 ] then RESEARCH_BUILD=`pwd`"/build" echo -e "No building directory was specified." else RESEARCH_BUILD="$2" fi mkdir -p $RESEARCH_BUILD echo -e "Building into directory $RESEARCH_BUILD ..." filename=$(basename -- "$1") baseurl='https://www.mat.tuhh.de' #make output copy echo "Preprocessing $filename..." if [ -f "$RESEARCH_BUILD/$filename" ] then echo "File exists, performing preprocessing in place" else cp $1 $RESEARCH_BUILD/$filename fi #extract title of topic echo "Extracting title of $1..." title=`$RESEARCH_ROOT/bin/extract_title.sh $RESEARCH_BUILD/$filename` echo "Title of Topic: $title" # start preprocessing # -- leave original untouched, only work with copy in $RESEARCH_BUILD # replace working group echo "Replacing working group keys..." namelist=`grep -h -i -m 1 -r "###\s*Working Groups:" $RESEARCH_BUILD/$filename | sed -e 's/^###\s*Working Groups:\s*//I' -e 's/\s*,\s*/\n/g' | sort -u` echo "Found the following keys: " $namelist for wg in $namelist do #extract full name from tuhh-webpage (deprecated) # new version uses files in static #fullname=`wget -qO- $baseurl/forschung/$wg | grep h1 | sed -e "s;<h1>\s*\(.*\)</h1>.*$;\1;g" | sed -e 's;^[ \t]*;;'` fullname=`cat $RESEARCH_ROOT/static/forschung/$wg.html | grep h1 | sed -e "s;<h1>\s*\(.*\)</h1>.*$;\1;g" | sed -e 's;^[ \t]*;;'` echo "Found working group $fullname" # append research to wg-file if [ $# -gt 1 ] then echo -e "\n$title\n" >> $RESEARCH_BUILD/$wg.md fi sed -i "s;### Working Groups:\(.*\)$wg\(.*\);### Working Groups:\1\[$fullname\]($wg.html)\2;g" $RESEARCH_BUILD/$filename #sed -i "s;### Working Groups:\(.*\)$wg\(.*\);### Working Groups:\1\[$fullname\]($baseurl/forschung/$wg)\2;g" $RESEARCH_BUILD/$filename #uncomment this line for linking the mat.tuhh.de webpage done