git-filter-branch.sh

   1 #!/bin/sh
   2 #
   3 # Rewrite revision history
   4 # Copyright (c) Petr Baudis, 2006
   5 # Minimal changes to "port" it to core-git (c) Johannes Schindelin, 2007
   6 #
   7 # Lets you rewrite the revision history of the current branch, creating
   8 # a new branch. You can specify a number of filters to modify the commits,
   9 # files and trees.
  10
  11 USAGE="git-filter-branch [-d TEMPDIR] [FILTERS] DESTBRANCH [REV-RANGE]"
  12 . git-sh-setup
  13
  14 warn () {
  15         echo "$*" >&2
  16 }
  17
  18 map()
  19 {
  20         # if it was not rewritten, take the original
  21         if test -r "$workdir/../map/$1"
  22         then
  23                 cat "$workdir/../map/$1"
  24         else
  25                 echo "$1"
  26         fi
  27 }
  28
  29 # override die(): this version puts in an extra line break, so that
  30 # the progress is still visible
  31
  32 die()
  33 {
  34         echo >&2
  35         echo "$*" >&2
  36         exit 1
  37 }
  38
  39 # When piped a commit, output a script to set the ident of either
  40 # "author" or "committer
  41
  42 set_ident () {
  43         lid="$(echo "$1" | tr "A-Z" "a-z")"
  44         uid="$(echo "$1" | tr "a-z" "A-Z")"
  45         pick_id_script='
  46                 /^'$lid' /{
  47                         s/'\''/'\''\\'\'\''/g
  48                         h
  49                         s/^'$lid' \([^<]*\) <[^>]*> .*$/\1/
  50                         s/'\''/'\''\'\'\''/g
  51                         s/.*/export GIT_'$uid'_NAME='\''&'\''/p
  52
  53                         g
  54                         s/^'$lid' [^<]* <\([^>]*\)> .*$/\1/
  55                         s/'\''/'\''\'\'\''/g
  56                         s/.*/export GIT_'$uid'_EMAIL='\''&'\''/p
  57
  58                         g
  59                         s/^'$lid' [^<]* <[^>]*> \(.*\)$/\1/
  60                         s/'\''/'\''\'\'\''/g
  61                         s/.*/export GIT_'$uid'_DATE='\''&'\''/p
  62
  63                         q
  64                 }
  65         '
  66
  67         LANG=C LC_ALL=C sed -ne "$pick_id_script"
  68         # Ensure non-empty id name.
  69         echo "[ -n \"\$GIT_${uid}_NAME\" ] || export GIT_${uid}_NAME=\"\${GIT_${uid}_EMAIL%%@*}\""
  70 }
  71
  72 tempdir=.git-rewrite
  73 filter_env=
  74 filter_tree=
  75 filter_index=
  76 filter_parent=
  77 filter_msg=cat
  78 filter_commit='git commit-tree "$@"'
  79 filter_tag_name=
  80 filter_subdir=
  81 orig_namespace=refs/original/
  82 force=
  83 while case "$#" in 0) usage;; esac
  84 do
  85         case "$1" in
  86         --)
  87                 shift
  88                 break
  89                 ;;
  90         --force|-f)
  91                 shift
  92                 force=t
  93                 continue
  94                 ;;
  95         -*)
  96                 ;;
  97         *)
  98                 break;
  99         esac
 100
 101         # all switches take one argument
 102         ARG="$1"
 103         case "$#" in 1) usage ;; esac
 104         shift
 105         OPTARG="$1"
 106         shift
 107
 108         case "$ARG" in
 109         -d)
 110                 tempdir="$OPTARG"
 111                 ;;
 112         --env-filter)
 113                 filter_env="$OPTARG"
 114                 ;;
 115         --tree-filter)
 116                 filter_tree="$OPTARG"
 117                 ;;
 118         --index-filter)
 119                 filter_index="$OPTARG"
 120                 ;;
 121         --parent-filter)
 122                 filter_parent="$OPTARG"
 123                 ;;
 124         --msg-filter)
 125                 filter_msg="$OPTARG"
 126                 ;;
 127         --commit-filter)
 128                 filter_commit="$OPTARG"
 129                 ;;
 130         --tag-name-filter)
 131                 filter_tag_name="$OPTARG"
 132                 ;;
 133         --subdirectory-filter)
 134                 filter_subdir="$OPTARG"
 135                 ;;
 136         --original)
 137                 orig_namespace="$OPTARG"
 138                 ;;
 139         *)
 140                 usage
 141                 ;;
 142         esac
 143 done
 144
 145 case "$force" in
 146 t)
 147         rm -rf "$tempdir"
 148 ;;
 149 '')
 150         test -d "$tempdir" &&
 151                 die "$tempdir already exists, please remove it"
 152 esac
 153 mkdir -p "$tempdir/t" &&
 154 tempdir="$(cd "$tempdir"; pwd)" &&
 155 cd "$tempdir/t" &&
 156 workdir="$(pwd)" ||
 157 die ""
 158
 159 # Make sure refs/original is empty
 160 git for-each-ref > "$tempdir"/backup-refs
 161 while read sha1 type name
 162 do
 163         case "$force,$name" in
 164         ,$orig_namespace*)
 165                 die "Namespace $orig_namespace not empty"
 166         ;;
 167         t,$orig_namespace*)
 168                 git update-ref -d "$name" $sha1
 169         ;;
 170         esac
 171 done < "$tempdir"/backup-refs
 172
 173 export GIT_DIR GIT_WORK_TREE=.
 174
 175 # These refs should be updated if their heads were rewritten
 176
 177 git rev-parse --revs-only --symbolic "$@" |
 178 while read ref
 179 do
 180         # normalize ref
 181         case "$ref" in
 182         HEAD)
 183                 ref="$(git symbolic-ref "$ref")"
 184         ;;
 185         refs/*)
 186         ;;
 187         *)
 188                 ref="$(git for-each-ref --format='%(refname)' |
 189                         grep /"$ref")"
 190         esac
 191
 192         git check-ref-format "$ref" && echo "$ref"
 193 done > "$tempdir"/heads
 194
 195 test -s "$tempdir"/heads ||
 196         die "Which ref do you want to rewrite?"
 197
 198 export GIT_INDEX_FILE="$(pwd)/../index"
 199 git read-tree || die "Could not seed the index"
 200
 201 ret=0
 202
 203 # map old->new commit ids for rewriting parents
 204 mkdir ../map || die "Could not create map/ directory"
 205
 206 case "$filter_subdir" in
 207 "")
 208         git rev-list --reverse --topo-order --default HEAD \
 209                 --parents "$@"
 210         ;;
 211 *)
 212         git rev-list --reverse --topo-order --default HEAD \
 213                 --parents --full-history "$@" -- "$filter_subdir"
 214 esac > ../revs || die "Could not get the commits"
 215 commits=$(wc -l <../revs | tr -d " ")
 216
 217 test $commits -eq 0 && die "Found nothing to rewrite"
 218
 219 # Rewrite the commits
 220
 221 i=0
 222 while read commit parents; do
 223         i=$(($i+1))
 224         printf "\rRewrite $commit ($i/$commits)"
 225
 226         case "$filter_subdir" in
 227         "")
 228                 git read-tree -i -m $commit
 229                 ;;
 230         *)
 231                 git read-tree -i -m $commit:"$filter_subdir"
 232         esac || die "Could not initialize the index"
 233
 234         export GIT_COMMIT=$commit
 235         git cat-file commit "$commit" >../commit ||
 236                 die "Cannot read commit $commit"
 237
 238         eval "$(set_ident AUTHOR <../commit)" ||
 239                 die "setting author failed for commit $commit"
 240         eval "$(set_ident COMMITTER <../commit)" ||
 241                 die "setting committer failed for commit $commit"
 242         eval "$filter_env" < /dev/null ||
 243                 die "env filter failed: $filter_env"
 244
 245         if [ "$filter_tree" ]; then
 246                 git checkout-index -f -u -a ||
 247                         die "Could not checkout the index"
 248                 # files that $commit removed are now still in the working tree;
 249                 # remove them, else they would be added again
 250                 git ls-files -z --others | xargs -0 rm -f
 251                 eval "$filter_tree" < /dev/null ||
 252                         die "tree filter failed: $filter_tree"
 253
 254                 git diff-index -r $commit | cut -f 2- | tr '\n' '\0' | \
 255                         xargs -0 git update-index --add --replace --remove
 256                 git ls-files -z --others | \
 257                         xargs -0 git update-index --add --replace --remove
 258         fi
 259
 260         eval "$filter_index" < /dev/null ||
 261                 die "index filter failed: $filter_index"
 262
 263         parentstr=
 264         for parent in $parents; do
 265                 for reparent in $(map "$parent"); do
 266                         parentstr="$parentstr -p $reparent"
 267                 done
 268         done
 269         if [ "$filter_parent" ]; then
 270                 parentstr="$(echo "$parentstr" | eval "$filter_parent")" ||
 271                                 die "parent filter failed: $filter_parent"
 272         fi
 273
 274         sed -e '1,/^$/d' <../commit | \
 275                 eval "$filter_msg" > ../message ||
 276                         die "msg filter failed: $filter_msg"
 277         sh -c "$filter_commit" "git commit-tree" \
 278                 $(git write-tree) $parentstr < ../message > ../map/$commit
 279 done <../revs
 280
 281 # In case of a subdirectory filter, it is possible that a specified head
 282 # is not in the set of rewritten commits, because it was pruned by the
 283 # revision walker.  Fix it by mapping these heads to the next rewritten
 284 # ancestor(s), i.e. the boundaries in the set of rewritten commits.
 285
 286 # NEEDSWORK: we should sort the unmapped refs topologically first
 287 while read ref
 288 do
 289         sha1=$(git rev-parse "$ref"^0)
 290         test -f "$workdir"/../map/$sha1 && continue
 291         # Assign the boundarie(s) in the set of rewritten commits
 292         # as the replacement commit(s).
 293         # (This would look a bit nicer if --not --stdin worked.)
 294         for p in $( (cd "$workdir"/../map; ls | sed "s/^/^/") |
 295                 git rev-list $ref --boundary --stdin |
 296                 sed -n "s/^-//p")
 297         do
 298                 map $p >> "$workdir"/../map/$sha1
 299         done
 300 done < "$tempdir"/heads
 301
 302 # Finally update the refs
 303
 304 _x40='[0-9a-f][0-9a-f][0-9a-f][0-9a-f][0-9a-f]'
 305 _x40="$_x40$_x40$_x40$_x40$_x40$_x40$_x40$_x40"
 306 count=0
 307 echo
 308 while read ref
 309 do
 310         # avoid rewriting a ref twice
 311         test -f "$orig_namespace$ref" && continue
 312
 313         sha1=$(git rev-parse "$ref"^0)
 314         rewritten=$(map $sha1)
 315
 316         test $sha1 = "$rewritten" &&
 317                 warn "WARNING: Ref '$ref' is unchanged" &&
 318                 continue
 319
 320         case "$rewritten" in
 321         '')
 322                 echo "Ref '$ref' was deleted"
 323                 git update-ref -m "filter-branch: delete" -d "$ref" $sha1 ||
 324                         die "Could not delete $ref"
 325         ;;
 326         $_x40)
 327                 echo "Ref '$ref' was rewritten"
 328                 git update-ref -m "filter-branch: rewrite" \
 329                                 "$ref" $rewritten $sha1 ||
 330                         die "Could not rewrite $ref"
 331         ;;
 332         *)
 333                 # NEEDSWORK: possibly add -Werror, making this an error
 334                 warn "WARNING: '$ref' was rewritten into multiple commits:"
 335                 warn "$rewritten"
 336                 warn "WARNING: Ref '$ref' points to the first one now."
 337                 rewritten=$(echo "$rewritten" | head -n 1)
 338                 git update-ref -m "filter-branch: rewrite to first" \
 339                                 "$ref" $rewritten $sha1 ||
 340                         die "Could not rewrite $ref"
 341         ;;
 342         esac
 343         git update-ref -m "filter-branch: backup" "$orig_namespace$ref" $sha1
 344         count=$(($count+1))
 345 done < "$tempdir"/heads
 346
 347 # TODO: This should possibly go, with the semantics that all positive given
 348 #       refs are updated, and their original heads stored in refs/original/
 349 # Filter tags
 350
 351 if [ "$filter_tag_name" ]; then
 352         git for-each-ref --format='%(objectname) %(objecttype) %(refname)' refs/tags |
 353         while read sha1 type ref; do
 354                 ref="${ref#refs/tags/}"
 355                 # XXX: Rewrite tagged trees as well?
 356                 if [ "$type" != "commit" -a "$type" != "tag" ]; then
 357                         continue;
 358                 fi
 359
 360                 if [ "$type" = "tag" ]; then
 361                         # Dereference to a commit
 362                         sha1t="$sha1"
 363                         sha1="$(git rev-parse "$sha1"^{commit} 2>/dev/null)" || continue
 364                 fi
 365
 366                 [ -f "../map/$sha1" ] || continue
 367                 new_sha1="$(cat "../map/$sha1")"
 368                 export GIT_COMMIT="$sha1"
 369                 new_ref="$(echo "$ref" | eval "$filter_tag_name")" ||
 370                         die "tag name filter failed: $filter_tag_name"
 371
 372                 echo "$ref -> $new_ref ($sha1 -> $new_sha1)"
 373
 374                 if [ "$type" = "tag" ]; then
 375                         # Warn that we are not rewriting the tag object itself.
 376                         warn "unreferencing tag object $sha1t"
 377                 fi
 378
 379                 git update-ref "refs/tags/$new_ref" "$new_sha1" ||
 380                         die "Could not write tag $new_ref"
 381         done
 382 fi
 383
 384 cd ../..
 385 rm -rf "$tempdir"
 386 echo
 387 test $count -gt 0 && echo "These refs were rewritten:"
 388 git show-ref | grep ^"$orig_namespace"
 389
 390 exit $ret