ok/jj
1
0
Fork 0
forked from mirrors/jj
jj/lib/tests/test_conflicts.rs

753 lines
16 KiB
Rust
Raw Normal View History

// Copyright 2021 The Jujutsu Authors
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// https://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use jj_lib::backend::FileId;
use jj_lib::conflicts::{
extract_as_single_hunk, materialize_merge_result, parse_conflict, update_from_content,
};
use jj_lib::merge::Merge;
use jj_lib::repo::Repo;
use jj_lib::repo_path::RepoPath;
use jj_lib::store::Store;
use pollster::FutureExt;
use testutils::TestRepo;
#[test]
fn test_materialize_conflict_basic() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
let path = RepoPath::from_internal_string("file");
let base_id = testutils::write_file(
store,
path,
"line 1
line 2
line 3
line 4
line 5
",
);
let left_id = testutils::write_file(
store,
path,
"line 1
line 2
left 3.1
left 3.2
left 3.3
line 4
line 5
",
);
let right_id = testutils::write_file(
store,
path,
"line 1
line 2
right 3.1
line 4
line 5
",
);
// The left side should come first. The diff should be use the smaller (right)
// side, and the left side should be a snapshot.
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone())],
vec![Some(left_id.clone()), Some(right_id.clone())],
);
insta::assert_snapshot!(
&materialize_conflict_string(store, path, &conflict),
@r###"
line 1
line 2
<<<<<<<
+++++++
left 3.1
left 3.2
left 3.3
%%%%%%%
-line 3
+right 3.1
>>>>>>>
line 4
line 5
"###
);
// Swap the positive terms in the conflict. The diff should still use the right
// side, but now the right side should come first.
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone())],
vec![Some(right_id.clone()), Some(left_id.clone())],
);
insta::assert_snapshot!(
&materialize_conflict_string(store, path, &conflict),
@r###"
line 1
line 2
<<<<<<<
%%%%%%%
-line 3
+right 3.1
+++++++
left 3.1
left 3.2
left 3.3
>>>>>>>
line 4
line 5
"###
);
}
#[test]
fn test_materialize_conflict_multi_rebase_conflicts() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
// Create changes (a, b, c) on top of the base, and linearize them.
let path = RepoPath::from_internal_string("file");
let base_id = testutils::write_file(
store,
path,
"line 1
line 2 base
line 3
",
);
let a_id = testutils::write_file(
store,
path,
"line 1
line 2 a.1
line 2 a.2
line 2 a.3
line 3
",
);
let b_id = testutils::write_file(
store,
path,
"line 1
line 2 b.1
line 2 b.2
line 3
",
);
let c_id = testutils::write_file(
store,
path,
"line 1
line 2 c.1
line 3
",
);
// The order of (a, b, c) should be preserved. For all cases, the "a" side
// should be a snapshot.
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone()), Some(base_id.clone())],
vec![Some(a_id.clone()), Some(b_id.clone()), Some(c_id.clone())],
);
insta::assert_snapshot!(
&materialize_conflict_string(store, path, &conflict),
@r###"
line 1
<<<<<<<
+++++++
line 2 a.1
line 2 a.2
line 2 a.3
%%%%%%%
-line 2 base
+line 2 b.1
+line 2 b.2
%%%%%%%
-line 2 base
+line 2 c.1
>>>>>>>
line 3
"###
);
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone()), Some(base_id.clone())],
vec![Some(c_id.clone()), Some(b_id.clone()), Some(a_id.clone())],
);
insta::assert_snapshot!(
&materialize_conflict_string(store, path, &conflict),
@r###"
line 1
<<<<<<<
%%%%%%%
-line 2 base
+line 2 c.1
%%%%%%%
-line 2 base
+line 2 b.1
+line 2 b.2
+++++++
line 2 a.1
line 2 a.2
line 2 a.3
>>>>>>>
line 3
"###
);
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone()), Some(base_id.clone())],
vec![Some(c_id.clone()), Some(a_id.clone()), Some(b_id.clone())],
);
insta::assert_snapshot!(
&materialize_conflict_string(store, path, &conflict),
@r###"
line 1
<<<<<<<
%%%%%%%
-line 2 base
+line 2 c.1
+++++++
line 2 a.1
line 2 a.2
line 2 a.3
%%%%%%%
-line 2 base
+line 2 b.1
+line 2 b.2
>>>>>>>
line 3
"###
);
}
#[test]
fn test_materialize_parse_roundtrip() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
let path = RepoPath::from_internal_string("file");
let base_id = testutils::write_file(
store,
path,
"line 1
line 2
line 3
line 4
line 5
",
);
let left_id = testutils::write_file(
store,
path,
"line 1 left
line 2 left
line 3
line 4
line 5 left
",
);
let right_id = testutils::write_file(
store,
path,
"line 1 right
line 2
line 3
line 4 right
line 5 right
",
);
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone())],
vec![Some(left_id.clone()), Some(right_id.clone())],
);
let materialized = materialize_conflict_string(store, path, &conflict);
insta::assert_snapshot!(
materialized,
@r###"
<<<<<<<
+++++++
line 1 left
line 2 left
%%%%%%%
-line 1
+line 1 right
line 2
>>>>>>>
line 3
<<<<<<<
%%%%%%%
line 4
-line 5
+line 5 left
+++++++
line 4 right
line 5 right
>>>>>>>
"###
);
// The first add should always be from the left side
insta::assert_debug_snapshot!(
parse_conflict(materialized.as_bytes(), conflict.num_sides()),
@r###"
Some(
[
Conflicted(
[
"line 1 left\nline 2 left\n",
"line 1\nline 2\n",
"line 1 right\nline 2\n",
],
),
Resolved(
"line 3\n",
),
Conflicted(
[
"line 4\nline 5 left\n",
"line 4\nline 5\n",
"line 4 right\nline 5 right\n",
],
),
],
)
"###);
}
#[test]
fn test_materialize_conflict_modify_delete() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
let path = RepoPath::from_internal_string("file");
let base_id = testutils::write_file(
store,
path,
"line 1
line 2
line 3
line 4
line 5
",
);
let modified_id = testutils::write_file(
store,
path,
"line 1
line 2
modified
line 4
line 5
",
);
let deleted_id = testutils::write_file(
store,
path,
"line 1
line 2
line 4
line 5
",
);
// left modifies a line, right deletes the same line.
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone())],
vec![Some(modified_id.clone()), Some(deleted_id.clone())],
);
insta::assert_snapshot!(&materialize_conflict_string(store, path, &conflict), @r###"
line 1
line 2
<<<<<<<
+++++++
modified
%%%%%%%
-line 3
>>>>>>>
line 4
line 5
"###
);
// right modifies a line, left deletes the same line.
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone())],
vec![Some(deleted_id.clone()), Some(modified_id.clone())],
);
insta::assert_snapshot!(&materialize_conflict_string(store, path, &conflict), @r###"
line 1
line 2
<<<<<<<
%%%%%%%
-line 3
+++++++
modified
>>>>>>>
line 4
line 5
"###
);
// modify/delete conflict at the file level
let conflict = Merge::from_removes_adds(
vec![Some(base_id.clone())],
vec![Some(modified_id.clone()), None],
);
insta::assert_snapshot!(&materialize_conflict_string(store, path, &conflict), @r###"
<<<<<<<
%%%%%%%
line 1
line 2
-line 3
+modified
line 4
line 5
+++++++
>>>>>>>
"###
);
}
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
#[test]
fn test_materialize_conflict_two_forward_diffs() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
// Create conflict A-B+B-C+D-E+C. This is designed to tempt the algorithm to
// produce a negative snapshot at the end like this:
// <<<<
// ====
// A
// %%%%
// B
// ++++
// D
// %%%%
// C
// ----
// E
// >>>>
// TODO: Maybe we should never have negative snapshots
let path = RepoPath::from_internal_string("file");
let a_id = testutils::write_file(store, path, "A\n");
let b_id = testutils::write_file(store, path, "B\n");
let c_id = testutils::write_file(store, path, "C\n");
let d_id = testutils::write_file(store, path, "D\n");
let e_id = testutils::write_file(store, path, "E\n");
let conflict = Merge::from_removes_adds(
vec![Some(b_id.clone()), Some(c_id.clone()), Some(e_id.clone())],
vec![
Some(a_id.clone()),
Some(b_id.clone()),
Some(d_id.clone()),
Some(c_id.clone()),
],
);
insta::assert_snapshot!(
&materialize_conflict_string(store, path, &conflict),
@r###"
<<<<<<<
+++++++
A
%%%%%%%
B
+++++++
D
%%%%%%%
C
-------
E
>>>>>>>
"###
);
}
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
#[test]
fn test_parse_conflict_resolved() {
assert_eq!(
parse_conflict(
b"line 1
line 2
line 3
line 4
line 5
",
2
),
None
)
}
#[test]
fn test_parse_conflict_simple() {
insta::assert_debug_snapshot!(
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
parse_conflict(
b"line 1
<<<<<<<
%%%%%%%
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
line 2
-line 3
+left
line 4
+++++++
right
>>>>>>>
line 5
",
2
),
@r###"
Some(
[
Resolved(
"line 1\n",
),
Conflicted(
[
"line 2\nleft\nline 4\n",
"line 2\nline 3\nline 4\n",
"right\n",
],
),
Resolved(
"line 5\n",
),
],
)
"###
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
)
}
#[test]
fn test_parse_conflict_multi_way() {
insta::assert_debug_snapshot!(
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
parse_conflict(
b"line 1
<<<<<<<
%%%%%%%
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
line 2
-line 3
+left
line 4
+++++++
right
%%%%%%%
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
line 2
+forward
line 3
line 4
>>>>>>>
line 5
",
3
),
@r###"
Some(
[
Resolved(
"line 1\n",
),
Conflicted(
[
"line 2\nleft\nline 4\n",
"line 2\nline 3\nline 4\n",
"right\n",
"line 2\nline 3\nline 4\n",
"line 2\nforward\nline 3\nline 4\n",
],
),
Resolved(
"line 5\n",
),
],
)
"###
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
)
}
#[test]
fn test_parse_conflict_different_wrong_arity() {
assert_eq!(
parse_conflict(
b"line 1
<<<<<<<
%%%%%%%
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
line 2
-line 3
+left
line 4
+++++++
right
>>>>>>>
line 5
",
3
),
None
)
}
#[test]
fn test_parse_conflict_malformed_marker() {
// The conflict marker is missing `%%%%%%%`
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
assert_eq!(
parse_conflict(
b"line 1
<<<<<<<
line 2
-line 3
+left
line 4
+++++++
right
>>>>>>>
line 5
",
2
),
None
)
}
#[test]
fn test_parse_conflict_malformed_diff() {
// The diff part is invalid (missing space before "line 4")
assert_eq!(
parse_conflict(
b"line 1
<<<<<<<
%%%%%%%
conflicts: add a function for parsing a materialized conflict I initially made the working copy materialize conflicts in its `check_out()` method. Then I changed it later (exactly a year ago, on Halloween of 2020, actually) so that the working copy expected conflicts to already have been materalized, which happens in `MutableRepo::check_out`(). I think my reasoning then was that the file system cannot represent a conflict. While it's true that the file system itself doesn't have information to know whether a file represents a conflict, we can record that ourselves. We already record whether a file is executable or not and then preserve that if we're on a file system that isn't able to record it. It's not that different to do the same for conflicts if we're on a file system that doesn't understand conflicts (i.e. all file systems). The plan is to have the working copy remember whether a file represents a conflict. When we check if it has changed, we parse the file, including conflict markers, and recreate the conflict from it. We should be able to do that losslessly (and we should adjust formats to make it possible if we find cases where it's not). Having the working copy preserve conflict states has several advantages: * Because conflicts are not materialized in the working copy, you can rebase the conflicted commit and the working copy without causing more conflicts (that's currently a UX bug I run into every now and then). * If you don't change anything in the working copy, it will be unchanged compared to its parent, which means we'll automatically abandon it if you update away from it. * The user can choose to resolve only some of the conflicts in a file and squash those in, and it'll work they way you'd hope. * It should make it easier to implement support for external merge tools (#18) without having them treat the working copy differently. This patch prepares for that work by adding support for parsing materialized conflicts.
2021-10-31 18:57:12 +00:00
line 2
-line 3
+left
line 4
+++++++
right
>>>>>>>
line 5
",
2
),
None
)
}
#[test]
fn test_update_conflict_from_content() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
let path = RepoPath::from_internal_string("dir/file");
let base_file_id = testutils::write_file(store, path, "line 1\nline 2\nline 3\n");
let left_file_id = testutils::write_file(store, path, "left 1\nline 2\nleft 3\n");
let right_file_id = testutils::write_file(store, path, "right 1\nline 2\nright 3\n");
let conflict = Merge::from_removes_adds(
vec![Some(base_file_id.clone())],
vec![Some(left_file_id.clone()), Some(right_file_id.clone())],
);
// If the content is unchanged compared to the materialized value, we get the
// old conflict id back.
let materialized = materialize_conflict_string(store, path, &conflict);
let parse = |content| {
update_from_content(&conflict, store, path, content)
.block_on()
.unwrap()
};
assert_eq!(parse(materialized.as_bytes()), conflict);
// If the conflict is resolved, we get None back to indicate that.
let expected_file_id = testutils::write_file(store, path, "resolved 1\nline 2\nresolved 3\n");
assert_eq!(
parse(b"resolved 1\nline 2\nresolved 3\n"),
Merge::normal(expected_file_id)
);
// If the conflict is partially resolved, we get a new conflict back.
let new_conflict = parse(
b"resolved 1\nline 2\n<<<<<<<\n%%%%%%%\n-line 3\n+left 3\n+++++++\nright 3\n>>>>>>>\n",
);
assert_ne!(new_conflict, conflict);
// Calculate expected new FileIds
let new_base_file_id = testutils::write_file(store, path, "resolved 1\nline 2\nline 3\n");
let new_left_file_id = testutils::write_file(store, path, "resolved 1\nline 2\nleft 3\n");
let new_right_file_id = testutils::write_file(store, path, "resolved 1\nline 2\nright 3\n");
assert_eq!(
new_conflict,
Merge::from_removes_adds(
vec![Some(new_base_file_id.clone())],
vec![
Some(new_left_file_id.clone()),
Some(new_right_file_id.clone())
]
)
);
}
#[test]
fn test_update_conflict_from_content_modify_delete() {
let test_repo = TestRepo::init();
let store = test_repo.repo.store();
let path = RepoPath::from_internal_string("dir/file");
let before_file_id = testutils::write_file(store, path, "line 1\nline 2 before\nline 3\n");
let after_file_id = testutils::write_file(store, path, "line 1\nline 2 after\nline 3\n");
let conflict =
Merge::from_removes_adds(vec![Some(before_file_id)], vec![Some(after_file_id), None]);
// If the content is unchanged compared to the materialized value, we get the
// old conflict id back.
let materialized = materialize_conflict_string(store, path, &conflict);
let parse = |content| {
update_from_content(&conflict, store, path, content)
.block_on()
.unwrap()
};
assert_eq!(parse(materialized.as_bytes()), conflict);
// If the conflict is resolved, we get None back to indicate that.
let expected_file_id = testutils::write_file(store, path, "resolved\n");
assert_eq!(parse(b"resolved\n"), Merge::normal(expected_file_id));
// If the conflict is modified, we get a new conflict back.
let new_conflict = parse(
b"<<<<<<<\n%%%%%%%\n line 1\n-line 2 before\n+line 2 modified after\n line 3\n+++++++\n>>>>>>>\n",
);
// Calculate expected new FileIds
let new_base_file_id = testutils::write_file(store, path, "line 1\nline 2 before\nline 3\n");
let new_left_file_id =
testutils::write_file(store, path, "line 1\nline 2 modified after\nline 3\n");
assert_eq!(
new_conflict,
Merge::from_removes_adds(
vec![Some(new_base_file_id.clone())],
vec![Some(new_left_file_id.clone()), None]
)
);
}
fn materialize_conflict_string(
store: &Store,
path: &RepoPath,
conflict: &Merge<Option<FileId>>,
) -> String {
let mut result: Vec<u8> = vec![];
let contents = extract_as_single_hunk(conflict, store, path).block_on();
materialize_merge_result(&contents, &mut result).unwrap();
String::from_utf8(result).unwrap()
}