Mercurial > hg
view tests/test-histedit-commute.t @ 40022:33eb670e2834
wireprotov2: define semantics for content redirects
When I implemented the clonebundles feature and deployed it on
hg.mozilla.org using Amazon S3 as a content server, server-side CPU
and bandwidth usage dropped off a cliff and a ton of server scaling
headaches went away pretty much the instant clients with support for
clonebundles were rolled out to Firefox CI.
An obvious takeaway from that experience was that offloading server
load to scalable file servers - potentially backed by a CDN - is a
really good idea. Another takeaway was that Mercurial's wire protocol
wasn't in a good position to support data offload generally.
In wire protocol version 1, there isn't a mechanism in the protocol to
say "grab the data from over here instead." For HTTP, we could teach
the client to follow HTTP redirects. Or we could invent a media type
that encoded redirects inline. But for SSH, we were pretty much out of
luck because that protocol wasn't very flexible.
Wire protocol version 2 offers the opportunity to do something better.
The recent generic server-side content caching layer in the wire
protocol version 2 server demonstrated that it is possible to have
drop-in caching of responses to command requests. This by itself
adds tons of value and already makes the built-in server much more
scalable. But I don't want to stop there.
The existing server-side caching implementation has a big weakness:
it requires the server to send data to the client. This means that
the Mercurial server is potentially sending gigabytes of data to
thousands of clients. This is problematic because compared to scaling
static file servers, scaling dynamic servers is *hard*.
A solution to this is to "offload" serving of content to something
that isn't the Mercurial server. By offloading content serving, you
turn the Mercurial server from a centralized monolithic service to
a distributed mostly-indexing service. Assuming high rates of content
offload, this should drastically reduce the total work performed by
the Mercurial server, both in terms of CPU and data transfer. This
will make Mercurial servers vastly easier to scale.
This commit defines the semantics for "content redirects" in wire
protocol version 2. Essentially:
* Servers advertise the set of locations a response could be served
from.
* When making requests, clients advertise the set of locations they
are willing to fetch content from.
* Servers can then replace the inline response with one that says
"get the response from over here instead."
This feature - when fully implemented - will allow extending the
server-side caching layer to facilitate such things as integrating
your server-side cache with a scalable blob store (such as S3 or
a CDN) and offloading most data transfer to that external service.
This feature could also be leveraged for load balancing. e.g.
requests could come into a central server and then get redirected
to an available mirror depending on server availability or locality.
There's tons of potential :)
Differential Revision: https://phab.mercurial-scm.org/D4774
author | Gregory Szorc <gregory.szorc@gmail.com> |
---|---|
date | Wed, 26 Sep 2018 18:02:06 -0700 |
parents | 89630d0b3e23 |
children | 11c076786d56 |
line wrap: on
line source
$ . "$TESTDIR/histedit-helpers.sh" $ cat >> $HGRCPATH <<EOF > [extensions] > histedit= > EOF $ initrepo () > { > hg init r > cd r > for x in a b c d e f ; do > echo $x > $x > hg add $x > hg ci -m $x > done > } $ initrepo log before edit $ hg log --graph @ changeset: 5:652413bf663e | tag: tip | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: f | o changeset: 4:e860deea161a | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 3:055a42cdd887 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: d | o changeset: 2:177f92b77385 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: c | o changeset: 1:d2ae7f538514 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: b | o changeset: 0:cb9a9f314b8b user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: a show the edit commands offered $ HGEDITOR=cat hg histedit 177f92b77385 pick 177f92b77385 2 c pick 055a42cdd887 3 d pick e860deea161a 4 e pick 652413bf663e 5 f # Edit history between 177f92b77385 and 652413bf663e # # Commits are listed from least to most recent # # You can reorder changesets by reordering the lines # # Commands: # # e, edit = use commit, but stop for amending # m, mess = edit commit message without changing commit content # p, pick = use commit # b, base = checkout changeset and apply further changesets from there # d, drop = remove commit from history # f, fold = use commit, but combine it with the one above # r, roll = like fold, but discard this commit's description and date # edit the history (use a hacky editor to check histedit-last-edit.txt backup) $ EDITED="$TESTTMP/editedhistory" $ cat > $EDITED <<EOF > edit 177f92b77385 c > pick e860deea161a e > pick 652413bf663e f > pick 055a42cdd887 d > EOF $ HGEDITOR="cat \"$EDITED\" > " hg histedit 177f92b77385 2>&1 | fixbundle 0 files updated, 0 files merged, 4 files removed, 0 files unresolved Editing (177f92b77385), you may commit or record as needed now. (hg histedit --continue to resume) rules should end up in .hg/histedit-last-edit.txt: $ cat .hg/histedit-last-edit.txt edit 177f92b77385 c pick e860deea161a e pick 652413bf663e f pick 055a42cdd887 d $ hg histedit --abort 4 files updated, 0 files merged, 0 files removed, 0 files unresolved $ cat > $EDITED <<EOF > pick 177f92b77385 c > pick e860deea161a e > pick 652413bf663e f > pick 055a42cdd887 d > EOF $ HGEDITOR="cat \"$EDITED\" > " hg histedit 177f92b77385 2>&1 | fixbundle log after edit $ hg log --graph @ changeset: 5:07114f51870f | tag: tip | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: d | o changeset: 4:8ade9693061e | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: f | o changeset: 3:d8249471110a | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 2:177f92b77385 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: c | o changeset: 1:d2ae7f538514 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: b | o changeset: 0:cb9a9f314b8b user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: a put things back $ hg histedit 177f92b77385 --commands - 2>&1 << EOF | fixbundle > pick 177f92b77385 c > pick 07114f51870f d > pick d8249471110a e > pick 8ade9693061e f > EOF $ hg log --graph @ changeset: 5:7eca9b5b1148 | tag: tip | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: f | o changeset: 4:915da888f2de | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 3:10517e47bbbb | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: d | o changeset: 2:177f92b77385 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: c | o changeset: 1:d2ae7f538514 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: b | o changeset: 0:cb9a9f314b8b user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: a slightly different this time $ hg histedit 177f92b77385 --commands - << EOF 2>&1 | fixbundle > pick 10517e47bbbb d > pick 7eca9b5b1148 f > pick 915da888f2de e > pick 177f92b77385 c > EOF $ hg log --graph @ changeset: 5:38b92f448761 | tag: tip | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: c | o changeset: 4:de71b079d9ce | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 3:be9ae3a309c6 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: f | o changeset: 2:799205341b6b | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: d | o changeset: 1:d2ae7f538514 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: b | o changeset: 0:cb9a9f314b8b user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: a keep prevents stripping dead revs $ hg histedit 799205341b6b --keep --commands - 2>&1 << EOF | fixbundle > pick 799205341b6b d > pick be9ae3a309c6 f > pick 38b92f448761 c > pick de71b079d9ce e > EOF $ hg log --graph @ changeset: 7:803ef1c6fcfd | tag: tip | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 6:ece0b8d93dda | parent: 3:be9ae3a309c6 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: c | | o changeset: 5:38b92f448761 | | user: test | | date: Thu Jan 01 00:00:00 1970 +0000 | | summary: c | | | o changeset: 4:de71b079d9ce |/ user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 3:be9ae3a309c6 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: f | o changeset: 2:799205341b6b | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: d | o changeset: 1:d2ae7f538514 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: b | o changeset: 0:cb9a9f314b8b user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: a try with --rev $ hg histedit --commands - --rev -2 2>&1 <<EOF | fixbundle > pick de71b079d9ce e > pick 38b92f448761 c > EOF hg: parse error: pick "de71b079d9ce" changeset was not a candidate (only use listed changesets) $ hg log --graph @ changeset: 7:803ef1c6fcfd | tag: tip | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 6:ece0b8d93dda | parent: 3:be9ae3a309c6 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: c | | o changeset: 5:38b92f448761 | | user: test | | date: Thu Jan 01 00:00:00 1970 +0000 | | summary: c | | | o changeset: 4:de71b079d9ce |/ user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: e | o changeset: 3:be9ae3a309c6 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: f | o changeset: 2:799205341b6b | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: d | o changeset: 1:d2ae7f538514 | user: test | date: Thu Jan 01 00:00:00 1970 +0000 | summary: b | o changeset: 0:cb9a9f314b8b user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: a Verify that revsetalias entries work with histedit: $ cat >> $HGRCPATH <<EOF > [revsetalias] > grandparent(ARG) = p1(p1(ARG)) > EOF $ echo extra commit >> c $ hg ci -m 'extra commit to c' $ HGEDITOR=cat hg histedit 'grandparent(.)' pick ece0b8d93dda 6 c pick 803ef1c6fcfd 7 e pick 9c863c565126 8 extra commit to c # Edit history between ece0b8d93dda and 9c863c565126 # # Commits are listed from least to most recent # # You can reorder changesets by reordering the lines # # Commands: # # e, edit = use commit, but stop for amending # m, mess = edit commit message without changing commit content # p, pick = use commit # b, base = checkout changeset and apply further changesets from there # d, drop = remove commit from history # f, fold = use commit, but combine it with the one above # r, roll = like fold, but discard this commit's description and date # should also work if a commit message is missing $ BUNDLE="$TESTDIR/missing-comment.hg" $ hg init missing $ cd missing $ hg unbundle $BUNDLE adding changesets adding manifests adding file changes added 3 changesets with 3 changes to 1 files new changesets 141947992243:bd22688093b3 (3 drafts) (run 'hg update' to get a working copy) $ hg co tip 1 files updated, 0 files merged, 0 files removed, 0 files unresolved $ hg log --graph @ changeset: 2:bd22688093b3 | tag: tip | user: Robert Altman <robert.altman@telventDTN.com> | date: Mon Nov 28 16:40:04 2011 +0000 | summary: Update file. | o changeset: 1:3b3e956f9171 | user: Robert Altman <robert.altman@telventDTN.com> | date: Mon Nov 28 16:37:57 2011 +0000 | o changeset: 0:141947992243 user: Robert Altman <robert.altman@telventDTN.com> date: Mon Nov 28 16:35:28 2011 +0000 summary: Checked in text file $ hg histedit 0 $ cd .. $ cd .. Test to make sure folding renames doesn't cause bogus conflicts (issue4251): $ hg init issue4251 $ cd issue4251 $ mkdir initial-dir $ echo foo > initial-dir/initial-file $ hg add initial-dir/initial-file $ hg commit -m "initial commit" Move the file to a new directory, and in the same commit, change its content: $ mkdir another-dir $ hg mv initial-dir/initial-file another-dir/ $ echo changed > another-dir/initial-file $ hg commit -m "moved and changed" Rename the file: $ hg mv another-dir/initial-file another-dir/renamed-file $ hg commit -m "renamed" Now, let's try to fold the second commit into the first: $ cat > editor.sh <<EOF > #!/bin/sh > cat > \$1 <<ENDOF > pick b0f4233702ca 0 initial commit > fold 5e8704a8f2d2 1 moved and changed > pick 40e7299e8fa7 2 renamed > ENDOF > EOF $ HGEDITOR="sh ./editor.sh" hg histedit 0 saved backup bundle to $TESTTMP/issue4251/.hg/strip-backup/b0f4233702ca-4cf5af69-histedit.hg $ hg --config diff.git=yes export 0 # HG changeset patch # User test # Date 0 0 # Thu Jan 01 00:00:00 1970 +0000 # Node ID fffadc26f8f85623ce60b028a3f1ccc3730f8530 # Parent 0000000000000000000000000000000000000000 pick b0f4233702ca 0 initial commit fold 5e8704a8f2d2 1 moved and changed pick 40e7299e8fa7 2 renamed diff --git a/another-dir/initial-file b/another-dir/initial-file new file mode 100644 --- /dev/null +++ b/another-dir/initial-file @@ -0,0 +1,1 @@ +changed $ hg --config diff.git=yes export 1 # HG changeset patch # User test # Date 0 0 # Thu Jan 01 00:00:00 1970 +0000 # Node ID 9b730d82b00af8a2766facebfa47cc124405a118 # Parent fffadc26f8f85623ce60b028a3f1ccc3730f8530 renamed diff --git a/another-dir/initial-file b/another-dir/renamed-file rename from another-dir/initial-file rename to another-dir/renamed-file $ cd .. Test that branches are preserved and stays active ------------------------------------------------- $ hg init repo-with-branch $ cd repo-with-branch $ echo a > a $ hg add a $ hg commit -m A $ hg branch foo marked working directory as branch foo (branches are permanent and global, did you want a bookmark?) $ echo a > b $ hg add b $ hg commit -m foo-B $ echo a > c $ hg add c $ hg commit -m foo-C $ hg branch foo $ echo "pick efefa76d6dc3 2 foo-C" >> cmd $ echo "pick 7336e7550422 1 foo-B" >> cmd $ HGEDITOR=cat hg histedit -r ".^" --commands cmd --quiet $ hg log --template '{rev} {branch}\n' 2 foo 1 foo 0 default $ hg branch foo $ cd ..