spring-clean.sh

#!/bin/bash
# spring cleaning v0.0.1
# written and otherwise cobbled together by:
# killermist (killermist@gmail.com)

#Imports
SELFLINK="$0"
FILENAME="$1"

#Constants
DEBUG="true"  #should probably turn this into a "-v" command-line option
HASHDIR=".hashes"
NESTINGCODE="NeStCoDe..nEsTcOdE--blah"  #Should be something fairly long and unlikely to be used as a filename

if [ "$FILENAME" = "" ]; then  #Let's get this out of the way early.
  echo "Base path necessary.  Aborting."
  exit
fi

#too many if-degug tests.  Let's clean that up.
function debug {
if [ $DEBUG = "true" ]; then
  if [ "$1" = "sleep" ]; then sleep 1; else echo -e $*; fi
fi
}

function trymkdir {
if [ ! -d "$1" ]; then mkdir "$1"; fi
if [ ! -d "$1" ]; then echo "$1" did not exist and failed to be created.  Aborting; exit; fi
}

#filename disambiguation suite
# as acquired from http://stackoverflow.com/questions/965053/extract-filename-and-extension-in-bash
# included both to cite source and for reference later, if needed
: <<'extract-filename-and-extension-in-bash'
#======================================================================================================================
#!/bin/bash
for fullpath in "$@"
do
    filename="${fullpath##*/}"                      # Strip longest match of */ from start
    dir="${fullpath:0:${#fullpath} - ${#filename}}" # Substring from 0 thru pos of filename
    base="${filename%.[^.]*}"                       # Strip shortest match of . plus at least one non-dot char from end
    ext="${filename:${#base} + 1}"                  # Substring from len of base thru end
    if [[ -z "$base" && -n "$ext" ]]; then          # If we have an extension and no base, it's really the base
        base=".$ext"
        ext=""
    fi
    echo -e "$fullpath:\n\tdir  = \"$dir\"\n\tbase = \"$base\"\n\text  = \"$ext\""
done
#======================================================================================================================
extract-filename-and-extension-in-bash

debug
debug "========================"
debug "$FILENAME:\n\tdir  = \"$FDSDIR\"\n\tbase = \"$FDSBASE\"\n\text  = \"$FDSEXT\""

trymkdir "$HASHDIR"

debug sleep
debug "$FILENAME" should be filename or target directory

if [ "$2" = "$NESTINGCODE" ] #testing to see if self-summoned
then
  if [ -d "$FILENAME" ]; then debug echo running nested, skip directories; exit; fi

  #For safety/sanity, verify that filename passed does actually qualify as a file.
  if [ ! -f "$FILENAME" ]; then echo "$FILENAME" is not a file.  Skipping.; exit; fi

  FDSFN="${FILENAME##*/}"                          # Strip longest match of */ from start
  FDSDIR="${FILENAME:0:${#FILENAME} - ${#FDSFN}}"  # Substring from 0 thru pos of filename
  FDSBASE="${FDSFN%.[^.]*}"                        # Strip shortest match of . plus at least one non-dot char from end
  FDSEXT="${FDSFN:${#FDSBASE} + 1}"                # Substring from len of base thru end
  if [[ -z "$FDSBASE" && -n "$FDSEXT" ]]; then     # If we have an extension and no base, it's really the base
      FDSBASE=".$FDSEXT"
      FDSEXT =""
  fi

  MD5=($(md5sum "$FILENAME"))
  SHA256=($(sha256sum "$FILENAME"))
  SIZE=`stat --printf="%s" "$FILENAME"`

  if [ "$FDSEXT" = "" ]; then #things with no extension at all shouldn't have a dot following
    TARGET="$HASHDIR"/"$SIZE"."$MD5"-md5".$SHA256"-sha256
  else
    TARGET="$HASHDIR"/"$SIZE"."$MD5"-md5".$SHA256"-sha256."$FDSEXT"
  fi

  debug "$FILENAME" tested negative as directory
  debug md5sum=\""$MD5"\"
  debug sha256=\""$SHA256"\"
  debug size=\""$SIZE"\"
  debug linking \""$FILENAME"\" to \""$TARGET"\"

  ln "$FILENAME" "$TARGET"

  #It can be useful to know what filenames things with multiple links have used and/or are using
  trymkdir "$HASHDIR"/"Seen-filenames"
  TARGET="$HASHDIR"/"Seen-filenames"/"$FDSFN"
  debug linking \""$FILENAME"\" to \""$TARGET"\"
  ln "$FILENAME" "$TARGET"

  #job done for this file, dismiss thyself
  exit
else
  debug not running nested, so executing find routine
  find "$FILENAME" -exec "$SELFLINK" {} "$NESTINGCODE" \; #  The workhorse of the whole operation
fi