path: root/bin/rotcheck

#!/bin/sh
set -uf
IFS="$(printf '\n\t')"
LC_ALL="C"

# Copyright (C) 2019 Jamie Nguyen <j@jamielinux.com>
#
# A simple shell script to recursively generate, update and verify checksums
# for files you care about. It's useful for detecting bit rot.
#
# It's written in POSIX shell, but requires GNU coreutils, BusyBox or some
# other collection that includes similar checksum tools.

VERSION=1.1.2
COMMAND="sha512sum"
CHECKFILE="./.rotcheck"

APPEND_MODE=0
CHECK_MODE=0
DELETE_MODE=0
UPDATE_MODE=0

IGNORE_MISSING=0
FOLLOW_SYMLINKS=1
VERBOSE=0
WARN_FORMATTING=0
EXCLUDE_HIDDEN=0
FORCE_UPDATE=0

usage() {
    cat << EOF
rotcheck $VERSION
Usage: rotcheck MODE [OPTIONS]
   or: rotcheck MODE [OPTIONS] -- [DIRECTORY]... [ARBITRARY FIND OPTION]...
Recursively generate, update and verify checksums.

MODES:
 -a           APPEND mode: Record checksums for any files without a checksum
              already. Never modify existing checksums.
 -c           CHECK mode: Check that files checksums are the same.
 -d           DELETE mode: Remove checksums for files that don't exist.
 -u           APPEND-AND-UPDATE mode: Like append-only mode, but also update
              checksums for files with a modification date newer than the
              the checksum file. (NB: Also see \`-M\`.)

OPTIONS:
 -b COMMAND   Checksum command to use. Default: sha512sum
 -f FILE      File to store checksums. For relative paths, prefix with "./"
              or the checksum file will be checksummed. Default: ./.rotcheck
 -h           Display this help.
 -n           Don't follow symlinks. The default is to follow symlinks.
 -v           Be more verbose when adding, deleting, changing or verifying
              checksums.
 -w           Warn about improperly formatted checksum lines.
 -x           Exclude all hidden files and directories when generating
              checksums. The default is to include them.
 -M           Use with \`-u\` to update checksums regardless of modification
              time. This is very slow so avoid if possible; try \`touch\`
              instead to bump the modification time of specific files.
              WARNING: The checksums might have changed due to bit rot so
              use this option with care!

 (specific to GNU coreutils >= 8.25)
 -i           Ignore missing files when verifying checksums.


Supported commands:
  GNU coreutils:
    md5sum, sha1sum, sha224sum, sha256sum, sha384sum, sha512sum, b2sum

  BusyBox (applets must be symlinked):
    md5sum, sha1sum, sha256sum, sha512sum, sha3sum

  BSD & macOS (install GNU coreutils):
    gmd5sum, gsha1sum, gsha224sum, gsha256sum, gsha384sum, gsha512sum, gb2sum


Examples:
  # Create checksum file (located at "./.rotcheck"):
  rotcheck -a

  # You've added some new files and need to append some checksums:
  rotcheck -va

  # You've edited some files and need to update the checksums (for files with
  # a modification time newer than the checksum file):
  rotcheck -vu

  # Verify checksums:
  rotcheck -c

  # Search other directories instead of the current directory.
  # WARNING: checksums might get duplicated if mixing relative and absolute
  # paths, or if you change the way you specify directory paths!
  rotcheck -a -- /mnt/archive-2018/ /mnt/archive-2019/

  # Exclude .git folders (these arguments are passed directly to find):
  rotcheck -a -- ! -path '*/\\.git/*'

EOF
    exit 0
}

fail() {
    printf '%s\n' "$@"; exit 1
}

# Curiously, I stumbled across a bug in bash-3.0.16 (c. 2004) or older
# where \0177 (DEL) isn't handled properly. See the `find_safe` function below.
# bash-3.1 (c. 2005), dash-0.5.2 (c. 2005), and zsh-3.1 (c. 2000) all work
# and probably others too.
if [ -n ${BASH+x} ] && [ -n ${BASH_VERSION+x} ]; then
    if printf '%s' "${BASH_VERSION:-x}" | grep -qE '^[0-2]+|^3\.0'; then
        fail "bash-3.0.16 and older are broken." \
             "Try bash>=3.1, dash, zsh, or another POSIX shell."
    fi
fi

# Command-line arguments. `getopts` is POSIX, while `getopt` is not.
[ $# -gt 0 ] && [ "$1" = "--help" ] && usage
while getopts ":acdub:f:hinvwxM" opt; do
    case "$opt" in
        a)  APPEND_MODE=1;;
        c)  CHECK_MODE=1;;
        d)  DELETE_MODE=1;;
        u)  UPDATE_MODE=1;;
        b)  COMMAND="$OPTARG";;
        f)  CHECKFILE="$OPTARG";;
        h)  usage;;
        i)  IGNORE_MISSING=1;;
        n)  FOLLOW_SYMLINKS=0;;
        v)  VERBOSE=1;;
        w)  WARN_FORMATTING=1;;
        x)  EXCLUDE_HIDDEN=1;;
        M)  FORCE_UPDATE=1;;
        \?) fail "-$OPTARG: Invalid argument";;
        :)  fail "-$OPTARG requires an argument";;
    esac
done; shift $(($OPTIND - 1))



# A few sanity checks.
MODE=$(($APPEND_MODE + $CHECK_MODE + $DELETE_MODE + $UPDATE_MODE))
if [ $MODE -eq 0 ]; then
    fail "Please specify one of -a, -c, -d, or -u." \
         "See \`rotcheck -h\` for help with usage."
elif [ $MODE -gt 1 ]; then
    fail "You can only use one of -a, -c, -d, or -u options." \
         "See \`rotcheck -h\` for help with usage."
elif [ $CHECK_MODE -eq 1 ] || [ $DELETE_MODE -eq 1 ]; then
    if [ ! -f "$CHECKFILE" ]; then
        fail "$CHECKFILE: No such file." \
             "Try running \`rotcheck -a\` first, or see \`rotcheck -h\`."
    fi
elif ! command -v "$COMMAND" >/dev/null 2>/dev/null; then
    fail "$COMMAND: command not found" \
         "Try specifying a supported command using \`rotcheck -b COMMAND\`." \
         "You may need to install GNU coreutils or BusyBox." \
         "On *BSD, GNU coreutils commands begin with 'g', like 'gsha512sum'." \
         "See \`rotcheck -h\` for help with usage."
fi

# When printing text to terminal, make sure it won't do anything unexpected.
printf_sanitized() {
    printf '%s' "$@" | tr -d '[:cntrl:]' | iconv -cs -f UTF-8 -t UTF-8
    printf '\n'
}

verify_checksums() {
    IGNORE="" ; [ $IGNORE_MISSING -eq 1 ]  && IGNORE="--ignore-missing"
    WARN=""   ; [ $WARN_FORMATTING -eq 1 ] && WARN="-w"
    $COMMAND -c $WARN $IGNORE -- "$CHECKFILE"
}

# Just verify checksums.
if [ $CHECK_MODE -eq 1 ]; then
    # Only GNU coreutils supports `--quiet`, so use `grep -v` instead.
    # Unfortunately, pipefail isn't POSIX so to return the exit status from the
    # checksum command, we have to be clever (aka crazy) with file descriptors
    # and subshells instead.
    if [ $VERBOSE -eq 1 ]; then
        verify_checksums
        exit $?
    else
        exec 4>&1
            (
                exec 3>&1
                    (
                        # 2>&1 preserves order of stdout/stderr.
                        verify_checksums 2>&1; printf '%d' $? 1>&3
                    ) | grep -Ev ': OK$' 1>&4
                exec 3>&-
            ) | ( read -r retval; exit $retval ); retval=$?
        exec 4>&-
        exit $retval
    fi
fi

# Delete checksums for files that no longer exist.
if [ $DELETE_MODE -eq 1 ]; then
    i=1
    for file in $(cut -d ' ' -f 3- -- "$CHECKFILE"); do
        # `sed -i` isn't POSIX (nor is `mktemp`), so use `ex` instead.
        if [ ! -f "$file" ]; then
            cat << EOF | ex -s -- "$CHECKFILE"
${i}d
x
EOF
            # Print what checksums were deleted.
            if [ $VERBOSE -eq 1 ]; then
                printf '%s' "DELETED: "
                printf_sanitized "$file"
            fi
        else
            # Only increment the line number if we didn't delete a line.
            i=$(($i + 1))
        fi
    done
    exit $?
fi

# For safety and sanity, ignore all filenames that have control characters
# like newline, tab, delete etc.
find_safe() {
    FIND_L=""
    FIND_FOLLOW=""
    if [ $FOLLOW_SYMLINKS -eq 1 ]; then
        # Old versions of findutils don't have -L. Use it if available.
        if find -L / -maxdepth 0 -type d >/dev/null 2>/dev/null; then
            FIND_L="-L"
        else
            FIND_FOLLOW="-follow"
        fi
    fi

    # POSIX find requires that you specify the search path either first
    # or immediately after -H/-L. Use current directory by default unless
    # user has specified a path.
    FIND_DOT="./"
    if [ $# -gt 0 ]; then
        first_char="$(printf '%s' "$1" | cut -c 1)"
        # Replace search path unless first arg is a non-path `find` option.
        if [ "$first_char" != "-" ] \
                && [ "$first_char" != "!" ] && [ "$first_char" != "(" ]; then
            FIND_DOT=""
        fi
    fi

    HIDDEN=""
    [ $EXCLUDE_HIDDEN -eq 1 ] && HIDDEN='*/\.*'

    find $FIND_L $FIND_DOT "$@" $FIND_FOLLOW \
        -type f ! -path "$CHECKFILE" ! -path "$HIDDEN" \
        ! -name "$(printf '*%b*' '\0001')" ! -name "$(printf '*%b*' '\0002')" \
        ! -name "$(printf '*%b*' '\0003')" ! -name "$(printf '*%b*' '\0004')" \
        ! -name "$(printf '*%b*' '\0005')" ! -name "$(printf '*%b*' '\0006')" \
        ! -name "$(printf '*%b*' '\0007')" ! -name "$(printf '*%b*' '\0010')" \
        ! -name "$(printf '*%b*' '\0011')" ! -name "$(printf '*%b*' '\0012')" \
        ! -name "$(printf '*%b*' '\0013')" ! -name "$(printf '*%b*' '\0014')" \
        ! -name "$(printf '*%b*' '\0015')" ! -name "$(printf '*%b*' '\0016')" \
        ! -name "$(printf '*%b*' '\0017')" ! -name "$(printf '*%b*' '\0020')" \
        ! -name "$(printf '*%b*' '\0021')" ! -name "$(printf '*%b*' '\0022')" \
        ! -name "$(printf '*%b*' '\0023')" ! -name "$(printf '*%b*' '\0024')" \
        ! -name "$(printf '*%b*' '\0025')" ! -name "$(printf '*%b*' '\0026')" \
        ! -name "$(printf '*%b*' '\0027')" ! -name "$(printf '*%b*' '\0030')" \
        ! -name "$(printf '*%b*' '\0031')" ! -name "$(printf '*%b*' '\0032')" \
        ! -name "$(printf '*%b*' '\0033')" ! -name "$(printf '*%b*' '\0034')" \
        ! -name "$(printf '*%b*' '\0035')" ! -name "$(printf '*%b*' '\0036')" \
        ! -name "$(printf '*%b*' '\0037')" ! -name "$(printf '*%b*' '\0177')"
}

find_updated_files() {
    if [ $FORCE_UPDATE -eq 1 ]; then
        find_safe "$@"
    else
        find_safe "$@" -newer "$CHECKFILE"
    fi
}

# This function could be replaced entirely with the much simpler:
#   cut -d ' ' -f 3- "$CHECKFILE" | grep -Fxn -- "$file" | cut -d ':' -f 1
# But this function is slightly faster as it avoids passing huge chunks of text
# (ie, the whole checksum file minus the first column) through a pipe.
get_line_number() {
    # Avoid `grep -E` as filename characters might get interpreted (eg, $).
    for l in $(grep -Fn -- "$file" "$CHECKFILE" | cut -d ':' -f 1); do
        if sed -n -e "${l}p" -- "$CHECKFILE" \
                | cut -d ' ' -f 3- | grep -Fxq -- "$file" >/dev/null; then
            printf '%d' "$l"
            return 0
        fi
    done
    printf '%d' "0"
}

umask 077
# For files with a modification date newer than the checksum file, if there's
# an existing checksum then update it. Otherwise append a new checksum.
if [ $UPDATE_MODE -eq 1 ] && [ -f "$CHECKFILE" ]; then
    for file in $(find_updated_files "$@"); do
        line_num="$(get_line_number)"
        if [ ${line_num:-0} -eq 0 ]; then
            # No checksum yet, so append one.
            $COMMAND -- "$file" >> "$CHECKFILE"
        else
            old="$(sed -n -e "${line_num}p" -- "$CHECKFILE" | cut -d ' ' -f 1)"
            new="$($COMMAND -- "$file")"
            # Should never happen, but double check these aren't empty:
            if [ -z ${old:+x} ] || [ -z ${new:+x} ]; then
                continue
            fi
            # `sed -i` isn't POSIX (nor is `mktemp`), so use `ex` instead.
            if [ "$old" != "${new%% *}" ]; then
                cat << EOF | ex -s -- "$CHECKFILE"
${line_num}c
$new
.
x
EOF
                # Bail immediately if something went wrong.
                [ $? -ne 0 ] && fail "Failed to update checksum file."

                # Print what checksums were changed.
                if [ $VERBOSE -eq 1 ]; then
                    printf '%s' "CHANGED: "
                    printf_sanitized "$file"
                fi
            fi
        fi
    done
fi

# Append checksums for files that have no checksum yet.
if [ $APPEND_MODE -eq 1 ] || [ $UPDATE_MODE -eq 1 ]; then
    for file in $(find_safe "$@"); do
        # Avoid `grep -E` as filename characters might get interpreted (eg, $).
        # The first grep isn't strictly needed, but grep+cut+grep is faster
        # than just cut+grep here.
        if [ ! -f "$CHECKFILE" ] || ! grep -- "$file" "$CHECKFILE" \
                            | cut -d ' ' -f 3- | grep -Fxq -- "$file"; then
            if ! $COMMAND -- "$file" >> "$CHECKFILE"; then
                fail "Failed to write to checksum file."
            fi

            # Print what checksums were appended.
            if [ $VERBOSE -eq 1 ]; then
                printf '%s' "ADDED: "
                printf_sanitized "$file"
            fi
        fi
    done
fi