this is testing

Committer:
pmallick
Date:
Thu Jan 14 19:12:57 2021 +0530
Revision:
0:e8a1ba50c46b
this is testing

Who changed what in which revision?

UserRevisionLine numberNew contents of line
pmallick 0:e8a1ba50c46b 1 #!/bin/sh
pmallick 0:e8a1ba50c46b 2 #
pmallick 0:e8a1ba50c46b 3 # Copyright (c) 2006, 2008 Junio C Hamano
pmallick 0:e8a1ba50c46b 4 #
pmallick 0:e8a1ba50c46b 5 # The "pre-rebase" hook is run just before "git rebase" starts doing
pmallick 0:e8a1ba50c46b 6 # its job, and can prevent the command from running by exiting with
pmallick 0:e8a1ba50c46b 7 # non-zero status.
pmallick 0:e8a1ba50c46b 8 #
pmallick 0:e8a1ba50c46b 9 # The hook is called with the following parameters:
pmallick 0:e8a1ba50c46b 10 #
pmallick 0:e8a1ba50c46b 11 # $1 -- the upstream the series was forked from.
pmallick 0:e8a1ba50c46b 12 # $2 -- the branch being rebased (or empty when rebasing the current branch).
pmallick 0:e8a1ba50c46b 13 #
pmallick 0:e8a1ba50c46b 14 # This sample shows how to prevent topic branches that are already
pmallick 0:e8a1ba50c46b 15 # merged to 'next' branch from getting rebased, because allowing it
pmallick 0:e8a1ba50c46b 16 # would result in rebasing already published history.
pmallick 0:e8a1ba50c46b 17
pmallick 0:e8a1ba50c46b 18 publish=next
pmallick 0:e8a1ba50c46b 19 basebranch="$1"
pmallick 0:e8a1ba50c46b 20 if test "$#" = 2
pmallick 0:e8a1ba50c46b 21 then
pmallick 0:e8a1ba50c46b 22 topic="refs/heads/$2"
pmallick 0:e8a1ba50c46b 23 else
pmallick 0:e8a1ba50c46b 24 topic=`git symbolic-ref HEAD` ||
pmallick 0:e8a1ba50c46b 25 exit 0 ;# we do not interrupt rebasing detached HEAD
pmallick 0:e8a1ba50c46b 26 fi
pmallick 0:e8a1ba50c46b 27
pmallick 0:e8a1ba50c46b 28 case "$topic" in
pmallick 0:e8a1ba50c46b 29 refs/heads/??/*)
pmallick 0:e8a1ba50c46b 30 ;;
pmallick 0:e8a1ba50c46b 31 *)
pmallick 0:e8a1ba50c46b 32 exit 0 ;# we do not interrupt others.
pmallick 0:e8a1ba50c46b 33 ;;
pmallick 0:e8a1ba50c46b 34 esac
pmallick 0:e8a1ba50c46b 35
pmallick 0:e8a1ba50c46b 36 # Now we are dealing with a topic branch being rebased
pmallick 0:e8a1ba50c46b 37 # on top of master. Is it OK to rebase it?
pmallick 0:e8a1ba50c46b 38
pmallick 0:e8a1ba50c46b 39 # Does the topic really exist?
pmallick 0:e8a1ba50c46b 40 git show-ref -q "$topic" || {
pmallick 0:e8a1ba50c46b 41 echo >&2 "No such branch $topic"
pmallick 0:e8a1ba50c46b 42 exit 1
pmallick 0:e8a1ba50c46b 43 }
pmallick 0:e8a1ba50c46b 44
pmallick 0:e8a1ba50c46b 45 # Is topic fully merged to master?
pmallick 0:e8a1ba50c46b 46 not_in_master=`git rev-list --pretty=oneline ^master "$topic"`
pmallick 0:e8a1ba50c46b 47 if test -z "$not_in_master"
pmallick 0:e8a1ba50c46b 48 then
pmallick 0:e8a1ba50c46b 49 echo >&2 "$topic is fully merged to master; better remove it."
pmallick 0:e8a1ba50c46b 50 exit 1 ;# we could allow it, but there is no point.
pmallick 0:e8a1ba50c46b 51 fi
pmallick 0:e8a1ba50c46b 52
pmallick 0:e8a1ba50c46b 53 # Is topic ever merged to next? If so you should not be rebasing it.
pmallick 0:e8a1ba50c46b 54 only_next_1=`git rev-list ^master "^$topic" ${publish} | sort`
pmallick 0:e8a1ba50c46b 55 only_next_2=`git rev-list ^master ${publish} | sort`
pmallick 0:e8a1ba50c46b 56 if test "$only_next_1" = "$only_next_2"
pmallick 0:e8a1ba50c46b 57 then
pmallick 0:e8a1ba50c46b 58 not_in_topic=`git rev-list "^$topic" master`
pmallick 0:e8a1ba50c46b 59 if test -z "$not_in_topic"
pmallick 0:e8a1ba50c46b 60 then
pmallick 0:e8a1ba50c46b 61 echo >&2 "$topic is already up to date with master"
pmallick 0:e8a1ba50c46b 62 exit 1 ;# we could allow it, but there is no point.
pmallick 0:e8a1ba50c46b 63 else
pmallick 0:e8a1ba50c46b 64 exit 0
pmallick 0:e8a1ba50c46b 65 fi
pmallick 0:e8a1ba50c46b 66 else
pmallick 0:e8a1ba50c46b 67 not_in_next=`git rev-list --pretty=oneline ^${publish} "$topic"`
pmallick 0:e8a1ba50c46b 68 /usr/bin/perl -e '
pmallick 0:e8a1ba50c46b 69 my $topic = $ARGV[0];
pmallick 0:e8a1ba50c46b 70 my $msg = "* $topic has commits already merged to public branch:\n";
pmallick 0:e8a1ba50c46b 71 my (%not_in_next) = map {
pmallick 0:e8a1ba50c46b 72 /^([0-9a-f]+) /;
pmallick 0:e8a1ba50c46b 73 ($1 => 1);
pmallick 0:e8a1ba50c46b 74 } split(/\n/, $ARGV[1]);
pmallick 0:e8a1ba50c46b 75 for my $elem (map {
pmallick 0:e8a1ba50c46b 76 /^([0-9a-f]+) (.*)$/;
pmallick 0:e8a1ba50c46b 77 [$1 => $2];
pmallick 0:e8a1ba50c46b 78 } split(/\n/, $ARGV[2])) {
pmallick 0:e8a1ba50c46b 79 if (!exists $not_in_next{$elem->[0]}) {
pmallick 0:e8a1ba50c46b 80 if ($msg) {
pmallick 0:e8a1ba50c46b 81 print STDERR $msg;
pmallick 0:e8a1ba50c46b 82 undef $msg;
pmallick 0:e8a1ba50c46b 83 }
pmallick 0:e8a1ba50c46b 84 print STDERR " $elem->[1]\n";
pmallick 0:e8a1ba50c46b 85 }
pmallick 0:e8a1ba50c46b 86 }
pmallick 0:e8a1ba50c46b 87 ' "$topic" "$not_in_next" "$not_in_master"
pmallick 0:e8a1ba50c46b 88 exit 1
pmallick 0:e8a1ba50c46b 89 fi
pmallick 0:e8a1ba50c46b 90
pmallick 0:e8a1ba50c46b 91 <<\DOC_END
pmallick 0:e8a1ba50c46b 92
pmallick 0:e8a1ba50c46b 93 This sample hook safeguards topic branches that have been
pmallick 0:e8a1ba50c46b 94 published from being rewound.
pmallick 0:e8a1ba50c46b 95
pmallick 0:e8a1ba50c46b 96 The workflow assumed here is:
pmallick 0:e8a1ba50c46b 97
pmallick 0:e8a1ba50c46b 98 * Once a topic branch forks from "master", "master" is never
pmallick 0:e8a1ba50c46b 99 merged into it again (either directly or indirectly).
pmallick 0:e8a1ba50c46b 100
pmallick 0:e8a1ba50c46b 101 * Once a topic branch is fully cooked and merged into "master",
pmallick 0:e8a1ba50c46b 102 it is deleted. If you need to build on top of it to correct
pmallick 0:e8a1ba50c46b 103 earlier mistakes, a new topic branch is created by forking at
pmallick 0:e8a1ba50c46b 104 the tip of the "master". This is not strictly necessary, but
pmallick 0:e8a1ba50c46b 105 it makes it easier to keep your history simple.
pmallick 0:e8a1ba50c46b 106
pmallick 0:e8a1ba50c46b 107 * Whenever you need to test or publish your changes to topic
pmallick 0:e8a1ba50c46b 108 branches, merge them into "next" branch.
pmallick 0:e8a1ba50c46b 109
pmallick 0:e8a1ba50c46b 110 The script, being an example, hardcodes the publish branch name
pmallick 0:e8a1ba50c46b 111 to be "next", but it is trivial to make it configurable via
pmallick 0:e8a1ba50c46b 112 $GIT_DIR/config mechanism.
pmallick 0:e8a1ba50c46b 113
pmallick 0:e8a1ba50c46b 114 With this workflow, you would want to know:
pmallick 0:e8a1ba50c46b 115
pmallick 0:e8a1ba50c46b 116 (1) ... if a topic branch has ever been merged to "next". Young
pmallick 0:e8a1ba50c46b 117 topic branches can have stupid mistakes you would rather
pmallick 0:e8a1ba50c46b 118 clean up before publishing, and things that have not been
pmallick 0:e8a1ba50c46b 119 merged into other branches can be easily rebased without
pmallick 0:e8a1ba50c46b 120 affecting other people. But once it is published, you would
pmallick 0:e8a1ba50c46b 121 not want to rewind it.
pmallick 0:e8a1ba50c46b 122
pmallick 0:e8a1ba50c46b 123 (2) ... if a topic branch has been fully merged to "master".
pmallick 0:e8a1ba50c46b 124 Then you can delete it. More importantly, you should not
pmallick 0:e8a1ba50c46b 125 build on top of it -- other people may already want to
pmallick 0:e8a1ba50c46b 126 change things related to the topic as patches against your
pmallick 0:e8a1ba50c46b 127 "master", so if you need further changes, it is better to
pmallick 0:e8a1ba50c46b 128 fork the topic (perhaps with the same name) afresh from the
pmallick 0:e8a1ba50c46b 129 tip of "master".
pmallick 0:e8a1ba50c46b 130
pmallick 0:e8a1ba50c46b 131 Let's look at this example:
pmallick 0:e8a1ba50c46b 132
pmallick 0:e8a1ba50c46b 133 o---o---o---o---o---o---o---o---o---o "next"
pmallick 0:e8a1ba50c46b 134 / / / /
pmallick 0:e8a1ba50c46b 135 / a---a---b A / /
pmallick 0:e8a1ba50c46b 136 / / / /
pmallick 0:e8a1ba50c46b 137 / / c---c---c---c B /
pmallick 0:e8a1ba50c46b 138 / / / \ /
pmallick 0:e8a1ba50c46b 139 / / / b---b C \ /
pmallick 0:e8a1ba50c46b 140 / / / / \ /
pmallick 0:e8a1ba50c46b 141 ---o---o---o---o---o---o---o---o---o---o---o "master"
pmallick 0:e8a1ba50c46b 142
pmallick 0:e8a1ba50c46b 143
pmallick 0:e8a1ba50c46b 144 A, B and C are topic branches.
pmallick 0:e8a1ba50c46b 145
pmallick 0:e8a1ba50c46b 146 * A has one fix since it was merged up to "next".
pmallick 0:e8a1ba50c46b 147
pmallick 0:e8a1ba50c46b 148 * B has finished. It has been fully merged up to "master" and "next",
pmallick 0:e8a1ba50c46b 149 and is ready to be deleted.
pmallick 0:e8a1ba50c46b 150
pmallick 0:e8a1ba50c46b 151 * C has not merged to "next" at all.
pmallick 0:e8a1ba50c46b 152
pmallick 0:e8a1ba50c46b 153 We would want to allow C to be rebased, refuse A, and encourage
pmallick 0:e8a1ba50c46b 154 B to be deleted.
pmallick 0:e8a1ba50c46b 155
pmallick 0:e8a1ba50c46b 156 To compute (1):
pmallick 0:e8a1ba50c46b 157
pmallick 0:e8a1ba50c46b 158 git rev-list ^master ^topic next
pmallick 0:e8a1ba50c46b 159 git rev-list ^master next
pmallick 0:e8a1ba50c46b 160
pmallick 0:e8a1ba50c46b 161 if these match, topic has not merged in next at all.
pmallick 0:e8a1ba50c46b 162
pmallick 0:e8a1ba50c46b 163 To compute (2):
pmallick 0:e8a1ba50c46b 164
pmallick 0:e8a1ba50c46b 165 git rev-list master..topic
pmallick 0:e8a1ba50c46b 166
pmallick 0:e8a1ba50c46b 167 if this is empty, it is fully merged to "master".
pmallick 0:e8a1ba50c46b 168
pmallick 0:e8a1ba50c46b 169 DOC_END