#!/bin/bash

# findbl - find bad symbolic links (dangling, malformed etc.)
# Copyright © 2000-2009 by Pádraig Brady <P@draigBrady.com>.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# See the GNU General Public License for more details,
# which is available at www.gnu.org


# Notes:
#
# By default (-d) it reports dangling (stale) links. Note invalid
# pathnames are reported as stale by this script.
#
# To report redundant info in symbolic links, such as multiple
# consequtive ////'s or /./././ etc. you specify the -n option.
# Note that fixing this won't give you back any disk space as
# the link will take a disk block anyway, but malformed links
# like this could cause other programs problems.
#
# To report all absolute links in the specified path(s), use
# the -A option. This can help when you want to move a tree
# around in the filesystem.
#
# Likewise the -l option reports all relative links in the
# specified paths.
#
# The -s option reports suspect absolute links. I.E. absolute
# links to files in the same directory as the link or below.
# TODO: support finding suspect relative link also. For e.g.:
# /etc/file.lnk ->../etc/file (python and abspath may be needed?)
#
# Note if this output passed to xargs or rm etc.
# then filenames with spaces will cause problems
#
# Should insert list of filesystems that support links
# or specify -xdev or something
#
# auto tidying of messy links could use tr -s / to squeeze consequtive /'s
# and sed s//..///d to remove /../

script_dir=$(dirname "$0")              #directory of this script
script_dir=$(readlink -f "$script_dir") #Make sure absolute path

. "$script_dir"/supprt/fslver

Usage() {
	ProgName=$(basename "$0")
	echo "find \"Bad\" symbolic Links.
Usage: $ProgName [-d] [-s] [-l] [-s] [-n] [[-r] [-f] paths(s) ...]

These options are mutually exclusive (i.e. only the last one takes effect).
-d Dangling (or stale) links. This is the default mode
-s Suspect links (absolute links to paths within or below the link's directory)
-l all reLative links
-A all Absolute links
-n reduNdant info in links (/././. ///// /../ etc.)

If no path(s) specified, then the current directory is assumed.

e.g. find dangling links in "library directories":
findbl \$(getffl)"
	exit
}

mode="dangling" #default

# Note must be careful with expressions (which are currently
# wrong I think), as the following are valid file/dir names:
# " ...", "..." etc. Hmm the symlinks program probably doesn't
# handle this correctly either?

ASC_01=$(printf "\001")

for arg
do
	case "$arg" in
	-d)
		mode="dangling" ;;
	-s)
		mode="suspect" ;;
	-A)
		mode="absolute"
		search_expr="${ASC_01}/" ;;
	-l)
		mode="relative"
		search_expr="${ASC_01}[^/]+" ;;
	-n)
		mode="redundant"
search_expr="(${ASC_01}.*[/]{2,})" #finds 2 or more consequtive /'s
search_expr="$search_expr|([${ASC_01}/]\./)" # starting with ./ or /./ anywhere
search_expr="$search_expr|(${ASC_01}.*[^.]+/[.][.]/)" #finds x/../y
search_expr="$search_expr|(${ASC_01}/[.][.]/)" #finds /../y
search_expr="$search_expr|(${ASC_01}.*[.]{3,}/[.][.]/)" #finds .../../y etc
search_expr="$search_expr|(${ASC_01}.+/$)" ;; #finds ending with /
	-h|--help|-help)
		Usage ;;
	-v|--version)
		Version ;;
	*)
		argsToPassOn="$argsToPassOn $(shell_quote "$arg")" ;;
	esac
done

. "$script_dir"/supprt/getfpf "$argsToPassOn"

EPWD=$(pwd | sed 's#/#\\/#g')

case "$mode" in
suspect)
	FPF="$FPF ->\1%p\1%l" ;;
absolute|relative|redundant)
	FPF="$FPF ->\1%l" ;;
esac

find "$@" -type l ! -name "*$LF*" -printf "$FPF\n" |
sort -u |            #merge links (indirectly) specified multiple times

case "$mode" in
dangling)
	#perl -nle '-e || print' | #this is much faster than below
	#/usr/share/dict/words is 45424 lines which is 0.12s for
	#perl and 3.8s for shell while read! should be OK though?
        #Note also this doesn't distinguish EPERM and ENOENT
        while read; do file="$REPLY"; [ -e "$file" ] || echo $file; done |
	tr '\n' '\0' |
	xargs -r0 ls -lUd -- |
        sed 's/.*[0-9]\{2\} \(.*\)/\1/' ;; #only leave link -> target
suspect)
	grep "${ASC_01}.*${ASC_01}/" |               #get absolute links
	sed -e "s/${ASC_01}.\//${ASC_01}$EPWD\//g" | #change rel to abs paths
	grep -E "${ASC_01}(.*)/[^/]+${ASC_01}\1" |   #find suspect links
	sed -e "s/${ASC_01}.*${ASC_01}/ /g" ;;       #drop work column
absolute|relative|redundant)
	grep -E "$search_expr" |
	tr '\1' ' ' ;;
esac