view tests/test-merge-no-file-change.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

  $ cat <<'EOF' >> "$HGRCPATH"
  > [extensions]
  > convert =
  > [templates]
  > l = '{rev}:{node|short} p={p1rev},{p2rev} m={manifest} f={files|json}'
  > EOF

  $ check_convert_identity () {
  >     hg convert -q "$1" "$1.converted"
  >     hg outgoing -q -R "$1.converted" "$1"
  >     if [ "$?" != 1 ]; then
  >         echo '*** BUG: hash changes on convert ***'
  >         hg log -R "$1.converted" -GTl
  >     fi
  > }

Files added at both parents:

  $ hg init added-both
  $ cd added-both
  $ touch a b c
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ hg ci -qAm2 c

  $ hg merge
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  committing files:
  b
  not reusing manifest (no file change in changelog, but manifest differs)
  committing manifest
  committing changelog
  updating the branch cache
  committed changeset 3:7aa8a293f5d97377037afc21e871e036e718d659
  $ hg log -GTl
  @    3:7aa8a293f5d9 p=2,1 m=3:8667461869a1 f=[]
  |\
  | o  2:e0ea47086fce p=0,-1 m=2:b2e5b07f9374 f=["c"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  

  $ cd ..
  $ check_convert_identity added-both

Files added at both parents, but the other removed at the merge:
(In this case, ctx.files() after the commit contains the removed file "b", but
its manifest does not differ from p1.)

  $ hg init added-both-removed-at-merge
  $ cd added-both-removed-at-merge
  $ touch a b c
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ hg ci -qAm2 c

  $ hg merge
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg rm -f b
  $ hg ci --debug -m merge
  committing files:
  committing manifest
  committing changelog
  updating the branch cache
  committed changeset 3:915745f3ca3d9d699925269474c2d0a9526e8dfa
  $ hg log -GTl
  @    3:915745f3ca3d p=2,1 m=3:8e9cf3456921 f=["b"]
  |\
  | o  2:e0ea47086fce p=0,-1 m=2:b2e5b07f9374 f=["c"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  

  $ cd ..
  $ check_convert_identity added-both

An identical file added at both parents:

  $ hg init added-identical
  $ cd added-identical
  $ touch a b
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ touch b
  $ hg ci -qAm2 b

  $ hg merge
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  reusing manifest from p1 (no file change)
  committing changelog
  updating the branch cache
  committed changeset 3:de26182cd210f0c3fb175ca7616704ab963d3024
  $ hg log -GTl
  @    3:de26182cd210 p=2,1 m=1:686dbf0aeca4 f=[]
  |\
  | o  2:f00991f11eca p=0,-1 m=1:686dbf0aeca4 f=["b"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  

  $ cd ..
  $ check_convert_identity added-identical

#if execbit

An identical file added at both parents, but the flag differs. Take local:

  $ hg init flag-change-take-p1
  $ cd flag-change-take-p1
  $ touch a b
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ touch b
  $ chmod +x b
  $ hg ci -qAm2 b

  $ hg merge
  warning: cannot merge flags for b without common ancestor - keeping local flags
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ chmod +x b
  $ hg ci --debug -m merge
  committing files:
  b
  reusing manifest form p1 (listed files actually unchanged)
  committing changelog
  updating the branch cache
  committed changeset 3:c8d50407916ef8a5a97cb6e36ca9bc844a6ee13e
  $ hg log -GTl
  @    3:c8d50407916e p=2,1 m=2:36b69ba4b24b f=[]
  |\
  | o  2:99451f16b3f5 p=0,-1 m=2:36b69ba4b24b f=["b"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  
  $ hg files -vr3
           0   a
           0 x b

  $ cd ..
  $ check_convert_identity flag-change-take-p1

An identical file added at both parents, but the flag differs. Take other:

  $ hg init flag-change-take-p2
  $ cd flag-change-take-p2
  $ touch a b
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ touch b
  $ chmod +x b
  $ hg ci -qAm2 b

  $ hg merge
  warning: cannot merge flags for b without common ancestor - keeping local flags
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ chmod -x b
  $ hg ci --debug -m merge
  committing files:
  b
  committing manifest
  committing changelog
  updating the branch cache
  committed changeset 3:06a62a687d87c7d8944743dee1ee9d8c66b3f6e3
  $ hg log -GTl
  @    3:06a62a687d87 p=2,1 m=3:2a315ba1aa45 f=["b"]
  |\
  | o  2:99451f16b3f5 p=0,-1 m=2:36b69ba4b24b f=["b"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  
  $ hg files -vr3
           0   a
           0   b

  $ cd ..
  $ check_convert_identity flag-change-take-p2

#endif

An identical file added at both parents, one more file added at p2:

  $ hg init added-some-p2
  $ cd added-some-p2
  $ touch a b c
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg ci -qAm2 c
  $ hg up -q 0
  $ touch b
  $ hg ci -qAm3 b

  $ hg merge
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  committing files:
  c
  not reusing manifest (no file change in changelog, but manifest differs)
  committing manifest
  committing changelog
  updating the branch cache
  committed changeset 4:f7fbc4e4d9a8fde03ba475adad675578c8bf472d
  $ hg log -GTl
  @    4:f7fbc4e4d9a8 p=3,2 m=3:92acd5bfd716 f=[]
  |\
  | o  3:e9d9f3cc981f p=0,-1 m=1:686dbf0aeca4 f=["b"]
  | |
  o |  2:93c5529a4ec7 p=1,-1 m=2:ae25a31b30b3 f=["c"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  

  $ cd ..
  $ check_convert_identity added-some-p2

An identical file added at both parents, one more file added at p1:
(In this case, p1 manifest is reused at the merge commit, which means the
manifest DAG does not have the same shape as the changelog.)

  $ hg init added-some-p1
  $ cd added-some-p1
  $ touch a b
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ touch b c
  $ hg ci -qAm2 b
  $ hg ci -qAm3 c

  $ hg merge
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  reusing manifest from p1 (no file change)
  committing changelog
  updating the branch cache
  committed changeset 4:a9f0f589a913f5a149dc10dfbd5af726977c36c4
  $ hg log -GTl
  @    4:a9f0f589a913 p=3,1 m=2:ae25a31b30b3 f=[]
  |\
  | o  3:b8dc385241b5 p=2,-1 m=2:ae25a31b30b3 f=["c"]
  | |
  | o  2:f00991f11eca p=0,-1 m=1:686dbf0aeca4 f=["b"]
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"]
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"]
  

  $ cd ..
  $ check_convert_identity added-some-p1

A file added at p2, a named branch created at p1:

  $ hg init named-branch-p1
  $ cd named-branch-p1
  $ touch a b
  $ hg ci -qAm0 a
  $ hg ci -qAm1 b
  $ hg up -q 0
  $ hg branch -q foo
  $ hg ci -m2

  $ hg merge default
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  committing files:
  b
  not reusing manifest (no file change in changelog, but manifest differs)
  committing manifest
  committing changelog
  updating the branch cache
  committed changeset 3:fb97d83b02fd072295cfc2171f21b7d38509bfd7
  $ hg log -GT'{l} branch={branch}'
  @    3:fb97d83b02fd p=2,1 m=2:9091c64f4ea1 f=[] branch=foo
  |\
  | o  2:a3a9fa6587e5 p=0,-1 m=0:8515d4bfda76 f=[] branch=foo
  | |
  o |  1:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"] branch=default
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"] branch=default
  

  $ cd ..
  $ check_convert_identity named-branch-p1

A file added at p1, a named branch created at p2:
(In this case, p1 manifest is reused at the merge commit, which means the
manifest DAG does not have the same shape as the changelog.)

  $ hg init named-branch-p2
  $ cd named-branch-p2
  $ touch a b
  $ hg ci -qAm0 a
  $ hg branch -q foo
  $ hg ci -m1
  $ hg up -q 0
  $ hg ci -qAm1 b

  $ hg merge foo
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  reusing manifest from p1 (no file change)
  committing changelog
  updating the branch cache
  committed changeset 3:036823e24692218324d4af43b07ff89f8a000096
  $ hg log -GT'{l} branch={branch}'
  @    3:036823e24692 p=2,1 m=1:686dbf0aeca4 f=[] branch=default
  |\
  | o  2:64d01526d4c2 p=0,-1 m=1:686dbf0aeca4 f=["b"] branch=default
  | |
  o |  1:da38c8e00727 p=0,-1 m=0:8515d4bfda76 f=[] branch=foo
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"] branch=default
  

  $ cd ..
  $ check_convert_identity named-branch-p2

A file changed once at both parents, but amended to have identical content:

  $ hg init amend-p1
  $ cd amend-p1
  $ touch a
  $ hg ci -qAm0 a
  $ echo foo > a
  $ hg ci -m1
  $ hg up -q 0
  $ echo bar > a
  $ hg ci -qm2
  $ echo foo > a
  $ hg ci -qm3 --amend

  $ hg merge
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci --debug -m merge
  reusing manifest from p1 (no file change)
  committing changelog
  updating the branch cache
  committed changeset 3:314e5bc5adf5c58ea571efabe33eedba20a201aa
  $ hg log -GT'{l} branch={branch}'
  @    3:314e5bc5adf5 p=2,1 m=1:d33ea248bd73 f=[] branch=default
  |\
  | o  2:de9c64f226a3 p=0,-1 m=1:d33ea248bd73 f=["a"] branch=default
  | |
  o |  1:6a74aec01b3c p=0,-1 m=1:d33ea248bd73 f=["a"] branch=default
  |/
  o  0:487a0a245cea p=-1,-1 m=0:8515d4bfda76 f=["a"] branch=default
  

  $ cd ..
  $ check_convert_identity amend-p1