#!/bin/sh REVISION=`head -1 ../../todo.txt | awk '{print $1}'` if [ $1 ] then RELEASE=$1 echo Creating Processing release $RELEASE... else RELEASE=$REVISION echo Creating Processing distribution for revision $REVISION... fi VERSIONED=`cat ../../app/src/processing/app/Base.java | grep $REVISION` if [ -z "$VERSIONED" ] then echo Fix the revision number in Base.java exit fi # remove any unfinished builds or old builds rm -rf processing rm -rf Processing* rm -rf processing-* rm -rf work ./make.sh if [ $1 ] then # write the release version number into the output directory echo $1 > work/Processing.app/Contents/Resources/Java/lib/version.txt fi echo Cleaning file boogers... # remove boogers find work -name "*~" -exec rm -f {} ';' # need to leave ds store stuff cuz one of those is important #find processing -name ".DS_Store" -exec rm -f {} ';' find work -name "._*" -exec rm -f {} ';' find work -name "Thumbs.db" -exec rm -f {} ';' # clean out the cvs entries find work -name "CVS" -exec rm -rf {} ';' 2> /dev/null find work -name ".cvsignore" -exec rm -rf {} ';' find work -name ".svn" -exec rm -rf {} 2> /dev/null ';' # the following was adopted from the makefile by Remko Troncon: # http://el-tramo.be/guides/fancy-dmg echo Creating disk image... SOURCE_DIR="work" SOURCE_FILES="Processing.app" OUTPUT_DMG="processing-$RELEASE" WORK_DMG="working.dmg" WORK_DIR="working_dir" gzip -cd template.dmg.gz > "$WORK_DMG" mkdir -p "$WORK_DIR" hdiutil attach "$WORK_DMG" -noautoopen -quiet -mountpoint "$WORK_DIR" for i in "$SOURCE_FILES"; do rm -rf "$WORK_DIR/$i" ditto -rsrc "$SOURCE_DIR/$i" "$WORK_DIR/$i" done WC_DEV=`hdiutil info | grep "$WORK_DIR" | awk '{print $1}'` && hdiutil detach $WC_DEV -quiet -force hdiutil convert "$WORK_DMG" -quiet -format UDZO -imagekey zlib-level=9 -o "$OUTPUT_DMG" rm -rf "$WORK_DIR" rm -f "$WORK_DMG" # for later, if we need to resize, etc #hdiutil resize -size 200mb -growonly -imageonly working.dmg echo Done.