Mercurial > hg
view tests/test-push-warn.t @ 39764:e4e881572382
localrepo: iteratively derive local repository type
This commit implements the dynamic local repository type derivation
that was explained in the recent commit
bfeab472e3c0 "localrepo: create new function for instantiating a local
repo object."
Instead of a static localrepository class/type which must be customized
after construction, we now dynamically construct a type by building up
base classes/types to represent specific repository interfaces.
Conceptually, the end state is similar to what was happening when
various extensions would monkeypatch the __class__ of newly-constructed
repo instances. However, the approach is inverted. Instead of making
the instance then customizing it, we do the customization up front
by influencing the behavior of the type then we instantiate that
custom type.
This approach gives us much more flexibility. For example, we can
use completely separate classes for implementing different aspects
of the repository. For example, we could have one class representing
revlog-based file storage and another representing non-revlog based
file storage. When then choose which implementation to use based on
the presence of repo requirements.
A concern with this approach is that it creates a lot more types
and complexity and that complexity adds overhead. Yes, it is true that
this approach will result in more types being created. Yes, this is
more complicated than traditional "instantiate a static type." However,
I believe the alternatives to supporting alternate storage backends
are just as complicated. (Before I arrived at this solution, I had
patches storing factory functions on local repo instances for e.g.
constructing a file storage instance. We ended up having a handful
of these. And this was logically identical to assigning custom
methods. Since we were logically changing the type of the instance,
I figured it would be better to just use specialized types instead
of introducing levels of abstraction at run-time.)
On the performance front, I don't believe that having N base classes
has any significant performance overhead compared to just a single base
class. Intuition says that Python will need to iterate the base classes
to find an attribute. However, CPython caches method lookups: as long as
the __class__ or MRO isn't changing, method attribute lookup should be
constant time after first access. And non-method attributes are stored
in __dict__, of which there is only 1 per object, so the number of
base classes for __dict__ is irrelevant.
Anyway, this commit splits up the monolithic completelocalrepository
interface into sub-interfaces: 1 for file storage and 1 representing
everything else.
We've taught ``makelocalrepository()`` to call a series of factory
functions which will produce types implementing specific interfaces.
It then calls type() to create a new type from the built-up list of
base types.
This commit should be considered a start and not the end state. I
suspect we'll hit a number of problems as we start to implement
alternate storage backends:
* Passing custom arguments to __init__ and setting custom attributes
on __dict__.
* Customizing the set of interfaces that are needed. e.g. the
"readonly" intent could translate to not requesting an interface
providing methods related to writing.
* More ergonomic way for extensions to insert themselves so their
callbacks aren't unconditionally called.
* Wanting to modify vfs instances, other arguments passed to __init__.
That being said, this code is usable in its current state and I'm
convinced future commits will demonstrate the value in this approach.
Differential Revision: https://phab.mercurial-scm.org/D4642
author | Gregory Szorc <gregory.szorc@gmail.com> |
---|---|
date | Tue, 18 Sep 2018 15:29:42 -0700 |
parents | 5abc47d4ca6b |
children | 5997eabc7b85 |
line wrap: on
line source
$ hg init a $ cd a $ echo foo > t1 $ hg add t1 $ hg commit -m "1" $ cd .. $ hg clone a b updating to branch default 1 files updated, 0 files merged, 0 files removed, 0 files unresolved $ cd a $ echo foo > t2 $ hg add t2 $ hg commit -m "2" $ cd ../b $ echo foo > t3 $ hg add t3 $ hg commit -m "3" Specifying a revset that evaluates to null will abort $ hg push -r '0 & 1' ../a pushing to ../a abort: specified revisions evaluate to an empty set (use different revision arguments) [255] $ hg push ../a pushing to ../a searching for changes remote has heads on branch 'default' that are not known locally: 1c9246a22a0a abort: push creates new remote head 1e108cc5548c! (pull and merge or see 'hg help push' for details about pushing new heads) [255] $ hg push --debug ../a pushing to ../a query 1; heads searching for changes taking quick initial sample query 2; still undecided: 1, sample size is: 1 2 total queries in *.????s (glob) listing keys for "phases" checking for updated bookmarks listing keys for "bookmarks" listing keys for "bookmarks" remote has heads on branch 'default' that are not known locally: 1c9246a22a0a new remote heads on branch 'default': 1e108cc5548c abort: push creates new remote head 1e108cc5548c! (pull and merge or see 'hg help push' for details about pushing new heads) [255] $ hg pull ../a pulling from ../a searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files (+1 heads) new changesets 1c9246a22a0a (run 'hg heads' to see heads, 'hg merge' to merge) $ hg push ../a pushing to ../a searching for changes abort: push creates new remote head 1e108cc5548c! (merge or see 'hg help push' for details about pushing new heads) [255] $ hg merge 1 files updated, 0 files merged, 0 files removed, 0 files unresolved (branch merge, don't forget to commit) $ hg commit -m "4" $ hg push ../a pushing to ../a searching for changes adding changesets adding manifests adding file changes added 2 changesets with 1 changes to 1 files $ cd .. $ hg init c $ cd c $ for i in 0 1 2; do > echo $i >> foo > hg ci -Am $i > done adding foo $ cd .. $ hg clone c d updating to branch default 1 files updated, 0 files merged, 0 files removed, 0 files unresolved $ cd d $ for i in 0 1; do > hg co -C $i > echo d-$i >> foo > hg ci -m d-$i > done 1 files updated, 0 files merged, 0 files removed, 0 files unresolved created new head 1 files updated, 0 files merged, 0 files removed, 0 files unresolved created new head $ HGMERGE=true hg merge 3 merging foo 0 files updated, 1 files merged, 0 files removed, 0 files unresolved (branch merge, don't forget to commit) $ hg ci -m c-d $ hg push ../c pushing to ../c searching for changes abort: push creates new remote head 6346d66eb9f5! (merge or see 'hg help push' for details about pushing new heads) [255] $ hg push -r 2 ../c pushing to ../c searching for changes no changes found [1] $ hg push -r 3 ../c pushing to ../c searching for changes abort: push creates new remote head a5dda829a167! (merge or see 'hg help push' for details about pushing new heads) [255] $ hg push -v -r 3 -r 4 ../c pushing to ../c searching for changes new remote heads on branch 'default': a5dda829a167 ee8fbc7a0295 abort: push creates new remote head a5dda829a167! (merge or see 'hg help push' for details about pushing new heads) [255] $ hg push -v -f -r 3 -r 4 ../c pushing to ../c searching for changes 2 changesets found uncompressed size of bundle content: 352 (changelog) 326 (manifests) 25\d foo (re) adding changesets adding manifests adding file changes added 2 changesets with 2 changes to 1 files (+2 heads) $ hg push -r 5 ../c pushing to ../c searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files (-1 heads) $ hg in ../c comparing with ../c searching for changes no changes found [1] Issue450: push -r warns about remote head creation even if no heads will be created $ hg init ../e $ hg push -r 0 ../e pushing to ../e searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files $ hg push -r 1 ../e pushing to ../e searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files $ cd .. Issue736: named branches are not considered for detection of unmerged heads in "hg push" $ hg init f $ cd f $ hg -q branch a $ echo 0 > foo $ hg -q ci -Am 0 $ echo 1 > foo $ hg -q ci -m 1 $ hg -q up 0 $ echo 2 > foo $ hg -q ci -m 2 $ hg -q up 0 $ hg -q branch b $ echo 3 > foo $ hg -q ci -m 3 $ cd .. $ hg -q clone f g $ cd g Push on existing branch and new branch: $ hg -q up 1 $ echo 4 > foo $ hg -q ci -m 4 $ hg -q up 0 $ echo 5 > foo $ hg -q branch c $ hg -q ci -m 5 $ hg push ../f pushing to ../f searching for changes abort: push creates new remote branches: c! (use 'hg push --new-branch' to create new remote branches) [255] $ hg push -r 4 -r 5 ../f pushing to ../f searching for changes abort: push creates new remote branches: c! (use 'hg push --new-branch' to create new remote branches) [255] Multiple new branches: $ hg -q branch d $ echo 6 > foo $ hg -q ci -m 6 $ hg push ../f pushing to ../f searching for changes abort: push creates new remote branches: c, d! (use 'hg push --new-branch' to create new remote branches) [255] $ hg push -r 4 -r 6 ../f pushing to ../f searching for changes abort: push creates new remote branches: c, d! (use 'hg push --new-branch' to create new remote branches) [255] $ cd ../g Fail on multiple head push: $ hg -q up 1 $ echo 7 > foo $ hg -q ci -m 7 $ hg push -r 4 -r 7 ../f pushing to ../f searching for changes abort: push creates new remote head 0b715ef6ff8f on branch 'a'! (merge or see 'hg help push' for details about pushing new heads) [255] Push replacement head on existing branches: $ hg -q up 3 $ echo 8 > foo $ hg -q ci -m 8 $ hg push -r 7 -r 8 ../f pushing to ../f searching for changes adding changesets adding manifests adding file changes added 2 changesets with 2 changes to 1 files Merge of branch a to other branch b followed by unrelated push on branch a: $ hg -q up 7 $ HGMERGE=true hg -q merge 8 $ hg -q ci -m 9 $ hg -q up 8 $ echo 10 > foo $ hg -q ci -m 10 $ hg push -r 9 ../f pushing to ../f searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files (-1 heads) $ hg push -r 10 ../f pushing to ../f searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files (+1 heads) Cheating the counting algorithm: $ hg -q up 9 $ HGMERGE=true hg -q merge 2 $ hg -q ci -m 11 $ hg -q up 1 $ echo 12 > foo $ hg -q ci -m 12 $ hg push -r 11 -r 12 ../f pushing to ../f searching for changes adding changesets adding manifests adding file changes added 2 changesets with 2 changes to 1 files Failed push of new named branch: $ echo 12 > foo $ hg -q ci -m 12a [1] $ hg -q up 11 $ echo 13 > foo $ hg -q branch e $ hg -q ci -m 13d $ hg push -r 12 -r 13 ../f pushing to ../f searching for changes abort: push creates new remote branches: e! (use 'hg push --new-branch' to create new remote branches) [255] Using --new-branch to push new named branch: $ hg push --new-branch -r 12 -r 13 ../f pushing to ../f searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files Pushing multi headed new branch: $ echo 14 > foo $ hg -q branch f $ hg -q ci -m 14 $ echo 15 > foo $ hg -q ci -m 15 $ hg -q up 14 $ echo 16 > foo $ hg -q ci -m 16 $ hg push --branch f --new-branch ../f pushing to ../f searching for changes abort: push creates new branch 'f' with multiple heads (merge or see 'hg help push' for details about pushing new heads) [255] $ hg push --branch f --new-branch --force ../f pushing to ../f searching for changes adding changesets adding manifests adding file changes added 3 changesets with 3 changes to 1 files (+1 heads) Checking prepush logic does not allow silently pushing multiple new heads but also doesn't report too many heads: $ cd .. $ hg init h $ echo init > h/init $ hg -R h ci -Am init adding init $ echo a > h/a $ hg -R h ci -Am a adding a $ hg clone h i updating to branch default 2 files updated, 0 files merged, 0 files removed, 0 files unresolved $ hg -R h up 0 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo b > h/b $ hg -R h ci -Am b adding b created new head $ hg -R i up 0 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo c > i/c $ hg -R i ci -Am c adding c created new head $ for i in `"$PYTHON" $TESTDIR/seq.py 3`; do hg -R h up -q 0; echo $i > h/b; hg -R h ci -qAm$i; done $ hg -R i push h pushing to h searching for changes remote has heads on branch 'default' that are not known locally: 534543e22c29 764f8ec07b96 afe7cc7679f5 ce4212fc8847 abort: push creates new remote head 97bd0c84d346! (pull and merge or see 'hg help push' for details about pushing new heads) [255] $ hg -R h up -q 0; echo x > h/b; hg -R h ci -qAmx $ hg -R i push h pushing to h searching for changes remote has heads on branch 'default' that are not known locally: 18ddb72c4590 534543e22c29 764f8ec07b96 afe7cc7679f5 and 1 others abort: push creates new remote head 97bd0c84d346! (pull and merge or see 'hg help push' for details about pushing new heads) [255] $ hg -R i push h -v pushing to h searching for changes remote has heads on branch 'default' that are not known locally: 18ddb72c4590 534543e22c29 764f8ec07b96 afe7cc7679f5 ce4212fc8847 new remote heads on branch 'default': 97bd0c84d346 abort: push creates new remote head 97bd0c84d346! (pull and merge or see 'hg help push' for details about pushing new heads) [255] Check prepush logic with merged branches: $ hg init j $ hg -R j branch a marked working directory as branch a (branches are permanent and global, did you want a bookmark?) $ echo init > j/foo $ hg -R j ci -Am init adding foo $ hg clone j k updating to branch a 1 files updated, 0 files merged, 0 files removed, 0 files unresolved $ echo a1 > j/foo $ hg -R j ci -m a1 $ hg -R k branch b marked working directory as branch b $ echo b > k/foo $ hg -R k ci -m b $ hg -R k up 0 1 files updated, 0 files merged, 0 files removed, 0 files unresolved $ hg -R k merge b 1 files updated, 0 files merged, 0 files removed, 0 files unresolved (branch merge, don't forget to commit) $ hg -R k ci -m merge $ hg -R k push -r a j pushing to j searching for changes abort: push creates new remote branches: b! (use 'hg push --new-branch' to create new remote branches) [255] Prepush -r should not allow you to sneak in new heads: $ hg init l $ cd l $ echo a >> foo $ hg -q add foo $ hg -q branch a $ hg -q ci -ma $ hg -q up null $ echo a >> foo $ hg -q add foo $ hg -q branch b $ hg -q ci -mb $ cd .. $ hg -q clone l m -u a $ cd m $ hg -q merge b $ hg -q ci -mmb $ hg -q up 0 $ echo a >> foo $ hg -q ci -ma2 $ hg -q up 2 $ echo a >> foo $ hg -q branch -f b $ hg -q ci -mb2 $ hg -q merge 3 $ hg -q ci -mma $ hg push ../l -b b pushing to ../l searching for changes abort: push creates new remote head 451211cc22b0 on branch 'a'! (merge or see 'hg help push' for details about pushing new heads) [255] $ cd .. Check prepush with new branch head on former topo non-head: $ hg init n $ cd n $ hg branch A marked working directory as branch A (branches are permanent and global, did you want a bookmark?) $ echo a >a $ hg ci -Ama adding a $ hg branch B marked working directory as branch B $ echo b >b $ hg ci -Amb adding b b is now branch head of B, and a topological head a is now branch head of A, but not a topological head $ hg clone . inner updating to branch B 2 files updated, 0 files merged, 0 files removed, 0 files unresolved $ cd inner $ hg up B 0 files updated, 0 files merged, 0 files removed, 0 files unresolved $ echo b1 >b1 $ hg ci -Amb1 adding b1 in the clone b1 is now the head of B $ cd .. $ hg up 0 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo a2 >a2 $ hg ci -Ama2 adding a2 a2 is now the new branch head of A, and a new topological head it replaces a former inner branch head, so it should at most warn about A, not B glog of local: $ hg log -G --template "{rev}: {branches} {desc}\n" @ 2: A a2 | | o 1: B b |/ o 0: A a glog of remote: $ hg log -G -R inner --template "{rev}: {branches} {desc}\n" @ 2: B b1 | o 1: B b | o 0: A a outgoing: $ hg out inner --template "{rev}: {branches} {desc}\n" comparing with inner searching for changes 2: A a2 $ hg push inner pushing to inner searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files (+1 heads) $ cd .. Check prepush with new branch head on former topo head: $ hg init o $ cd o $ hg branch A marked working directory as branch A (branches are permanent and global, did you want a bookmark?) $ echo a >a $ hg ci -Ama adding a $ hg branch B marked working directory as branch B $ echo b >b $ hg ci -Amb adding b b is now branch head of B, and a topological head $ hg up 0 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo a1 >a1 $ hg ci -Ama1 adding a1 a1 is now branch head of A, and a topological head $ hg clone . inner updating to branch A 2 files updated, 0 files merged, 0 files removed, 0 files unresolved $ cd inner $ hg up B 1 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo b1 >b1 $ hg ci -Amb1 adding b1 in the clone b1 is now the head of B $ cd .. $ echo a2 >a2 $ hg ci -Ama2 adding a2 a2 is now the new branch head of A, and a topological head it replaces a former topological and branch head, so this should not warn glog of local: $ hg log -G --template "{rev}: {branches} {desc}\n" @ 3: A a2 | o 2: A a1 | | o 1: B b |/ o 0: A a glog of remote: $ hg log -G -R inner --template "{rev}: {branches} {desc}\n" @ 3: B b1 | | o 2: A a1 | | o | 1: B b |/ o 0: A a outgoing: $ hg out inner --template "{rev}: {branches} {desc}\n" comparing with inner searching for changes 3: A a2 $ hg push inner pushing to inner searching for changes adding changesets adding manifests adding file changes added 1 changesets with 1 changes to 1 files $ cd .. Check prepush with new branch head and new child of former branch head but child is on different branch: $ hg init p $ cd p $ hg branch A marked working directory as branch A (branches are permanent and global, did you want a bookmark?) $ echo a0 >a $ hg ci -Ama0 adding a $ echo a1 >a $ hg ci -ma1 $ hg up null 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ hg branch B marked working directory as branch B $ echo b0 >b $ hg ci -Amb0 adding b $ echo b1 >b $ hg ci -mb1 $ hg clone . inner updating to branch B 1 files updated, 0 files merged, 0 files removed, 0 files unresolved $ hg up A 1 files updated, 0 files merged, 1 files removed, 0 files unresolved $ hg branch -f B marked working directory as branch B $ echo a3 >a $ hg ci -ma3 created new head $ hg up 3 1 files updated, 0 files merged, 1 files removed, 0 files unresolved $ hg branch -f A marked working directory as branch A $ echo b3 >b $ hg ci -mb3 created new head glog of local: $ hg log -G --template "{rev}: {branches} {desc}\n" @ 5: A b3 | | o 4: B a3 | | o | 3: B b1 | | o | 2: B b0 / o 1: A a1 | o 0: A a0 glog of remote: $ hg log -G -R inner --template "{rev}: {branches} {desc}\n" @ 3: B b1 | o 2: B b0 o 1: A a1 | o 0: A a0 outgoing: $ hg out inner --template "{rev}: {branches} {desc}\n" comparing with inner searching for changes 4: B a3 5: A b3 $ hg push inner pushing to inner searching for changes abort: push creates new remote head 7d0f4fb6cf04 on branch 'A'! (merge or see 'hg help push' for details about pushing new heads) [255] $ hg push inner -r4 -r5 pushing to inner searching for changes abort: push creates new remote head 7d0f4fb6cf04 on branch 'A'! (merge or see 'hg help push' for details about pushing new heads) [255] $ hg in inner comparing with inner searching for changes no changes found [1] Test fail hook $ hg push inner --config hooks.fail-push="echo running fail-push hook" pushing to inner searching for changes running fail-push hook abort: push creates new remote head 7d0f4fb6cf04 on branch 'A'! (merge or see 'hg help push' for details about pushing new heads) [255] $ cd ..