ok/jj
1
0
Fork 0
forked from mirrors/jj
jj/cli/tests/test_workspaces.rs

758 lines
28 KiB
Rust
Raw Normal View History

// Copyright 2022 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 std::path::Path;
use crate::common::TestEnvironment;
/// Test adding a second workspace
#[test]
fn test_workspaces_add_second_workspace() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "initial"]);
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: rlvkpnrz e0e6d567 (empty) (no description set)
"###);
let (stdout, stderr) = test_env.jj_cmd_ok(
&main_path,
&["workspace", "add", "--name", "second", "../secondary"],
);
insta::assert_snapshot!(stdout.replace('\\', "/"), @"");
insta::assert_snapshot!(stderr.replace('\\', "/"), @r###"
Created workspace in "../secondary"
Working copy now at: rzvqmyuk 397eac93 (empty) (no description set)
Parent commit : qpvuntsm 7d308bc9 initial
Added 1 files, modified 0 files, removed 0 files
"###);
// Can see the working-copy commit in each workspace in the log output. The "@"
// node in the graph indicates the current workspace's working-copy commit.
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
397eac932ad3 second@
@ e0e6d5672858 default@
7d308bc9d934
000000000000
"###);
insta::assert_snapshot!(get_log_output(&test_env, &secondary_path), @r###"
@ 397eac932ad3 second@
e0e6d5672858 default@
7d308bc9d934
000000000000
"###);
// Both workspaces show up when we list them
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: rlvkpnrz e0e6d567 (empty) (no description set)
second: rzvqmyuk 397eac93 (empty) (no description set)
"###);
}
/// Test how sparse patterns are inherited
#[test]
fn test_workspaces_sparse_patterns() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "ws1"]);
let ws1_path = test_env.env_root().join("ws1");
let ws2_path = test_env.env_root().join("ws2");
let ws3_path = test_env.env_root().join("ws3");
test_env.jj_cmd_ok(&ws1_path, &["sparse", "set", "--clear", "--add=foo"]);
test_env.jj_cmd_ok(&ws1_path, &["workspace", "add", "../ws2"]);
let stdout = test_env.jj_cmd_success(&ws2_path, &["sparse", "list"]);
insta::assert_snapshot!(stdout, @r###"
foo
"###);
test_env.jj_cmd_ok(&ws2_path, &["sparse", "set", "--add=bar"]);
test_env.jj_cmd_ok(&ws2_path, &["workspace", "add", "../ws3"]);
let stdout = test_env.jj_cmd_success(&ws3_path, &["sparse", "list"]);
insta::assert_snapshot!(stdout, @r###"
bar
foo
"###);
}
/// Test adding a second workspace while the current workspace is editing a
/// merge
#[test]
fn test_workspaces_add_second_workspace_on_merge() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
test_env.jj_cmd_ok(&main_path, &["describe", "-m=left"]);
test_env.jj_cmd_ok(&main_path, &["new", "@-", "-m=right"]);
test_env.jj_cmd_ok(&main_path, &["new", "all:@-+", "-m=merge"]);
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: zsuskuln 21a0ea6d (empty) merge
"###);
test_env.jj_cmd_ok(
&main_path,
&["workspace", "add", "--name", "second", "../secondary"],
);
// The new workspace's working-copy commit shares all parents with the old one.
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
6d4c2b8ab610 second@
@ 21a0ea6d1c86 default@
09ba8d9dfa21
1694f2ddf8ec
000000000000
"###);
}
/// Test adding a workspace, but at a specific revision using '-r'
#[test]
fn test_workspaces_add_workspace_at_revision() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file-1"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "first"]);
std::fs::write(main_path.join("file-2"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "second"]);
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: kkmpptxz 2801c219 (empty) (no description set)
"###);
let (_, stderr) = test_env.jj_cmd_ok(
&main_path,
&[
"workspace",
"add",
"--name",
"second",
"../secondary",
"-r",
"@--",
],
);
insta::assert_snapshot!(stderr.replace('\\', "/"), @r###"
Created workspace in "../secondary"
Working copy now at: zxsnswpr e6baf9d9 (empty) (no description set)
Parent commit : qpvuntsm e7d7dbb9 first
Added 1 files, modified 0 files, removed 0 files
"###);
// Can see the working-copy commit in each workspace in the log output. The "@"
// node in the graph indicates the current workspace's working-copy commit.
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
e6baf9d9cfd0 second@
@ 2801c219094d default@
4ec5df5a189c
e7d7dbb91c5a
000000000000
"###);
insta::assert_snapshot!(get_log_output(&test_env, &secondary_path), @r###"
@ e6baf9d9cfd0 second@
2801c219094d default@
4ec5df5a189c
e7d7dbb91c5a
000000000000
"###);
}
/// Test multiple `-r` flags to `workspace add` to create a workspace
/// working-copy commit with multiple parents.
#[test]
fn test_workspaces_add_workspace_multiple_revisions() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
std::fs::write(main_path.join("file-1"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "first"]);
test_env.jj_cmd_ok(&main_path, &["new", "-r", "root()"]);
std::fs::write(main_path.join("file-2"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "second"]);
test_env.jj_cmd_ok(&main_path, &["new", "-r", "root()"]);
std::fs::write(main_path.join("file-3"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "third"]);
test_env.jj_cmd_ok(&main_path, &["new", "-r", "root()"]);
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
@ 5b36783cd11c
23881f07b53c
1f6a15f0af2a
e7d7dbb91c5a
000000000000
"###);
let (_, stderr) = test_env.jj_cmd_ok(
&main_path,
&[
"workspace",
"add",
"--name=merge",
"../merged",
"-r=238",
"-r=1f6",
"-r=e7d",
],
);
insta::assert_snapshot!(stderr.replace('\\', "/"), @r###"
Created workspace in "../merged"
Working copy now at: wmwvqwsz fa8fdc28 (empty) (no description set)
Parent commit : mzvwutvl 23881f07 third
Parent commit : kkmpptxz 1f6a15f0 second
Parent commit : qpvuntsm e7d7dbb9 first
Added 3 files, modified 0 files, removed 0 files
"###);
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
fa8fdc28af12 merge@
e7d7dbb91c5a
1f6a15f0af2a
23881f07b53c
@ 5b36783cd11c default@
000000000000
"###);
}
/// Test making changes to the working copy in a workspace as it gets rewritten
/// from another workspace
#[test]
fn test_workspaces_conflicting_edits() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file"), "contents\n").unwrap();
test_env.jj_cmd_ok(&main_path, &["new"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../secondary"]);
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
265af0cdbcc7 secondary@
@ 351099fa72cf default@
cf911c223d3e
000000000000
"###);
// Make changes in both working copies
std::fs::write(main_path.join("file"), "changed in main\n").unwrap();
std::fs::write(secondary_path.join("file"), "changed in second\n").unwrap();
// Squash the changes from the main workspace into the initial commit (before
// running any command in the secondary workspace
let (stdout, stderr) = test_env.jj_cmd_ok(&main_path, &["squash"]);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @r###"
Rebased 1 descendant commits
Working copy now at: mzvwutvl fe8f41ed (empty) (no description set)
Parent commit : qpvuntsm c0d4a99e (no description set)
"###);
// The secondary workspace's working-copy commit was updated
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
@ fe8f41ed01d6 default@
a1896a17282f secondary@
c0d4a99ef98a
000000000000
"###);
let stderr = test_env.jj_cmd_failure(&secondary_path, &["st"]);
insta::assert_snapshot!(stderr, @r###"
Error: The working copy is stale (not updated since operation 58b580b12eee).
Hint: Run `jj workspace update-stale` to update it.
See https://github.com/martinvonz/jj/blob/main/docs/working-copy.md#stale-working-copy for more information.
"###);
// Same error on second run, and from another command
let stderr = test_env.jj_cmd_failure(&secondary_path, &["log"]);
insta::assert_snapshot!(stderr, @r###"
Error: The working copy is stale (not updated since operation 58b580b12eee).
Hint: Run `jj workspace update-stale` to update it.
See https://github.com/martinvonz/jj/blob/main/docs/working-copy.md#stale-working-copy for more information.
"###);
let (stdout, stderr) = test_env.jj_cmd_ok(&secondary_path, &["workspace", "update-stale"]);
// It was detected that the working copy is now stale.
// Since there was an uncommitted change in the working copy, it should
// have been committed first (causing divergence)
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @r###"
Concurrent modification detected, resolving automatically.
Rebased 1 descendant commits onto commits rewritten by other operation
Working copy now at: pmmvwywv?? a1896a17 (empty) (no description set)
Added 0 files, modified 1 files, removed 0 files
"###);
insta::assert_snapshot!(get_log_output(&test_env, &secondary_path),
@r###"
b0b43f24d501 (divergent)
fe8f41ed01d6 default@
@ a1896a17282f secondary@ (divergent)
c0d4a99ef98a
000000000000
"###);
// The stale working copy should have been resolved by the previous command
let stdout = get_log_output(&test_env, &secondary_path);
assert!(!stdout.starts_with("The working copy is stale"));
insta::assert_snapshot!(stdout, @r###"
b0b43f24d501 (divergent)
fe8f41ed01d6 default@
@ a1896a17282f secondary@ (divergent)
c0d4a99ef98a
000000000000
"###);
}
/// Test a clean working copy that gets rewritten from another workspace
#[test]
fn test_workspaces_updated_by_other() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file"), "contents\n").unwrap();
test_env.jj_cmd_ok(&main_path, &["new"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../secondary"]);
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
265af0cdbcc7 secondary@
@ 351099fa72cf default@
cf911c223d3e
000000000000
"###);
// Rewrite the check-out commit in one workspace.
std::fs::write(main_path.join("file"), "changed in main\n").unwrap();
let (stdout, stderr) = test_env.jj_cmd_ok(&main_path, &["squash"]);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @r###"
Rebased 1 descendant commits
Working copy now at: mzvwutvl fe8f41ed (empty) (no description set)
Parent commit : qpvuntsm c0d4a99e (no description set)
"###);
// The secondary workspace's working-copy commit was updated.
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
@ fe8f41ed01d6 default@
a1896a17282f secondary@
c0d4a99ef98a
000000000000
"###);
let stderr = test_env.jj_cmd_failure(&secondary_path, &["st"]);
insta::assert_snapshot!(stderr, @r###"
Error: The working copy is stale (not updated since operation 58b580b12eee).
Hint: Run `jj workspace update-stale` to update it.
See https://github.com/martinvonz/jj/blob/main/docs/working-copy.md#stale-working-copy for more information.
"###);
let (stdout, stderr) = test_env.jj_cmd_ok(&secondary_path, &["workspace", "update-stale"]);
// It was detected that the working copy is now stale, but clean. So no
// divergent commit should be created.
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @r###"
Working copy now at: pmmvwywv a1896a17 (empty) (no description set)
Added 0 files, modified 1 files, removed 0 files
"###);
insta::assert_snapshot!(get_log_output(&test_env, &secondary_path),
@r###"
fe8f41ed01d6 default@
@ a1896a17282f secondary@
c0d4a99ef98a
000000000000
"###);
}
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
#[test]
fn test_workspaces_current_op_discarded_by_other() {
let test_env = TestEnvironment::default();
// Use the local backend because GitBackend::gc() depends on the git CLI.
test_env.jj_cmd_ok(
test_env.env_root(),
&["init", "main", "--config-toml=ui.allow-init-native=true"],
);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file"), "contents\n").unwrap();
test_env.jj_cmd_ok(&main_path, &["new"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../secondary"]);
// Create an op by abandoning the parent commit. Importantly, that commit also
// changes the target tree in the secondary workspace.
test_env.jj_cmd_ok(&main_path, &["abandon", "@-"]);
let stdout = test_env.jj_cmd_success(
&main_path,
&[
"operation",
"log",
"--template",
r#"id.short(10) ++ " " ++ description"#,
],
);
insta::assert_snapshot!(stdout, @r###"
@ f1dc1bf396 abandon commit acb4b92517b20aa4ee2f3dc58d7c2373754d0b29a3df310dbabda5813f13c3730d28d6a1b6dd37f3b0c8c5c9adaead5dab242ffe7ecc2e5a6a534fe4c6639f89
47f1ad5e1a Create initial working-copy commit in workspace secondary
fd918ce207 add workspace 'secondary'
5927500c4b new empty commit
c86cb4fdf4 snapshot working copy
17dbb2fe40 add workspace 'default'
cecfee9647 initialize repo
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
0000000000
"###);
// Abandon ops, including the one the secondary workspace is currently on.
test_env.jj_cmd_ok(&main_path, &["operation", "abandon", "..@-"]);
test_env.jj_cmd_ok(&main_path, &["util", "gc", "--expire=now"]);
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
@ 6dc8f254cd3c default@
4278b78fb503 secondary@
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
000000000000
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
"###);
let stderr = test_env.jj_cmd_failure(&secondary_path, &["st"]);
insta::assert_snapshot!(stderr, @r###"
Error: Could not read working copy's operation.
Hint: Run `jj workspace update-stale` to recover.
See https://github.com/martinvonz/jj/blob/main/docs/working-copy.md#stale-working-copy for more information.
"###);
let (stdout, stderr) = test_env.jj_cmd_ok(&secondary_path, &["workspace", "update-stale"]);
insta::assert_snapshot!(stderr, @r###"
Failed to read working copy's current operation; attempting recovery. Error message from read attempt: Object 47f1ad5e1adfaa1e9863181e6e45ae12b9db553e82bde82e278c0c6288053c833b4039a5baf2415871964bcdcd4b7875692979e66f2ec18bfc3896355091cf53 of type operation not found
Created and checked out recovery commit df3e46148439
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
"###);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
54400d0c58b7 secondary@
4278b78fb503
@ 6dc8f254cd3c default@
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
000000000000
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
"###);
let (stdout, stderr) = test_env.jj_cmd_ok(&secondary_path, &["st"]);
insta::assert_snapshot!(stderr, @"");
insta::assert_snapshot!(stdout, @r###"
Working copy changes:
A file
Working copy : znkkpsqq 54400d0c (no description set)
Parent commit: pmmvwywv 4278b78f (empty) (no description set)
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
"###);
let (stdout, stderr) = test_env.jj_cmd_ok(&secondary_path, &["obslog"]);
insta::assert_snapshot!(stderr, @"");
insta::assert_snapshot!(stdout, @r###"
@ znkkpsqq test.user@example.com 2001-02-03 04:05:16.000 +07:00 secondary@ 54400d0c
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
(no description set)
znkkpsqq hidden test.user@example.com 2001-02-03 04:05:16.000 +07:00 df3e4614
workspace: recover from missing operation If the operation corresponding to a workspace is missing for some reason (the specific situation in the test in this commit is that an operation was abandoned and garbage-collected from another workspace), currently, jj fails with a 255 error code. Teach jj a way to recover from this situation. When jj detects such a situation, it prints a message and stops operation, similar to when a workspace is stale. The message tells the user what command to run. When that command is run, jj loads the repo at the @ operation (instead of the operation of the workspace), creates a new commit on the @ commit with an empty tree, and then proceeds as usual - in particular, including the auto-snapshotting of the working tree, which creates another commit that obsoletes the newly created commit. There are several design points I considered. 1) Whether the recovery should be automatic, or (as in this commit) manual in that the user should be prompted to run a command. The user might prefer to recover in another way (e.g. by simply deleting the workspace) and this situation is (hopefully) rare enough that I think it's better to prompt the user. 2) Which command the user should be prompted to run (and thus, which command should be taught to perform the recovery). I chose "workspace update-stale" because the circumstances are very similar to it: it's symptom is that the regular jj operation is blocked somewhere at the beginning, and "workspace update-stale" already does some special work before the blockage (this commit adds more of such special work). But it might be better for something more explicitly named, or even a sequence of commands (e.g. "create a new operation that becomes @ that no workspace points to", "low-level command that makes a workspace point to the operation @") but I can see how this can be unnecessarily confusing for the user. 3) How we recover. I can think of several ways: a) Always create a commit, and allow the automatic snapshotting to create another commit that obsoletes this commit. b) Create a commit but somehow teach the automatic snapshotting to replace the created commit in-place (so it has no predecessor, as viewed in "obslog"). c) Do either a) or b), with the added improvement that if there is no diff between the newly created commit and the former @, to behave as if no new commit was created (@ remains as the former @). I chose a) since it was the simplest and most easily reasoned about, which I think is the best way to go when recovering from a rare situation.
2024-02-03 05:26:23 +00:00
(empty) (no description set)
"###);
}
#[test]
fn test_workspaces_update_stale_noop() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let (stdout, stderr) = test_env.jj_cmd_ok(&main_path, &["workspace", "update-stale"]);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @r###"
Nothing to do (the working copy is not stale).
"###);
let stderr = test_env.jj_cmd_failure(
&main_path,
&["workspace", "update-stale", "--ignore-working-copy"],
);
insta::assert_snapshot!(stderr, @r###"
Error: This command must be able to update the working copy.
Hint: Don't use --ignore-working-copy.
"###);
let stdout = test_env.jj_cmd_success(&main_path, &["op", "log", "-Tdescription"]);
insta::assert_snapshot!(stdout, @r###"
@ add workspace 'default'
initialize repo
"###);
}
/// Test "update-stale" in a dirty, but not stale working copy.
#[test]
fn test_workspaces_update_stale_snapshot() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file"), "changed in main\n").unwrap();
test_env.jj_cmd_ok(&main_path, &["new"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../secondary"]);
// Record new operation in one workspace.
test_env.jj_cmd_ok(&main_path, &["new"]);
// Snapshot the other working copy, which unfortunately results in concurrent
// operations, but should be resolved cleanly.
std::fs::write(secondary_path.join("file"), "changed in second\n").unwrap();
let (stdout, stderr) = test_env.jj_cmd_ok(&secondary_path, &["workspace", "update-stale"]);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @r###"
Concurrent modification detected, resolving automatically.
Nothing to do (the working copy is not stale).
"###);
insta::assert_snapshot!(get_log_output(&test_env, &secondary_path), @r###"
@ 4976dfa88529 secondary@
8357b22214ba default@
1a769966ed69
b4a6c25e7778
000000000000
"###);
}
/// Test forgetting workspaces
#[test]
fn test_workspaces_forget() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
std::fs::write(main_path.join("file"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["new"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../secondary"]);
let (stdout, stderr) = test_env.jj_cmd_ok(&main_path, &["workspace", "forget"]);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @"");
// When listing workspaces, only the secondary workspace shows up
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
secondary: pmmvwywv feda1c4e (empty) (no description set)
"###);
// `jj status` tells us that there's no working copy here
let (stdout, stderr) = test_env.jj_cmd_ok(&main_path, &["st"]);
insta::assert_snapshot!(stdout, @r###"
No working copy
"###);
insta::assert_snapshot!(stderr, @"");
// The old working copy doesn't get an "@" in the log output
// TODO: We should abandon the empty working copy commit
// TODO: It seems useful to still have the "secondary@" marker here even though
// there's only one workspace. We should show it when the command is not run
// from that workspace.
insta::assert_snapshot!(get_log_output(&test_env, &main_path), @r###"
feda1c4e5ffe
e949be04e93e
123ed18e4c4c
000000000000
"###);
// Revision "@" cannot be used
let stderr = test_env.jj_cmd_failure(&main_path, &["log", "-r", "@"]);
insta::assert_snapshot!(stderr, @r###"
Error: Workspace "default" doesn't have a working copy
"###);
// Try to add back the workspace
// TODO: We should make this just add it back instead of failing
let stderr = test_env.jj_cmd_failure(&main_path, &["workspace", "add", "."]);
insta::assert_snapshot!(stderr, @r###"
Error: Workspace already exists
"###);
// Add a third workspace...
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../third"]);
// ... and then forget it, and the secondary workspace too
let (stdout, stderr) =
test_env.jj_cmd_ok(&main_path, &["workspace", "forget", "secondary", "third"]);
insta::assert_snapshot!(stdout, @"");
insta::assert_snapshot!(stderr, @"");
// No workspaces left
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @"");
}
#[test]
fn test_workspaces_forget_multi_transaction() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
std::fs::write(main_path.join("file"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["new"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../second"]);
test_env.jj_cmd_ok(&main_path, &["workspace", "add", "../third"]);
// there should be three workspaces
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: rlvkpnrz e949be04 (empty) (no description set)
second: pmmvwywv feda1c4e (empty) (no description set)
third: rzvqmyuk 485853ed (empty) (no description set)
"###);
// delete two at once, in a single tx
test_env.jj_cmd_ok(&main_path, &["workspace", "forget", "second", "third"]);
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: rlvkpnrz e949be04 (empty) (no description set)
"###);
// the op log should have multiple workspaces forgotten in a single tx
let stdout = test_env.jj_cmd_success(&main_path, &["op", "log", "--limit", "1"]);
insta::assert_snapshot!(stdout, @r###"
@ c28e1481737d test-username@host.example.com 2001-02-03 04:05:12.000 +07:00 - 2001-02-03 04:05:12.000 +07:00
forget workspaces second, third
args: jj workspace forget second third
"###);
// now, undo, and that should restore both workspaces
test_env.jj_cmd_ok(&main_path, &["op", "undo"]);
// finally, there should be three workspaces at the end
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: rlvkpnrz e949be04 (empty) (no description set)
second: pmmvwywv feda1c4e (empty) (no description set)
third: rzvqmyuk 485853ed (empty) (no description set)
"###);
}
/// Test context of commit summary template
#[test]
fn test_list_workspaces_template() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
test_env.add_config(
r#"
templates.commit_summary = """commit_id.short() ++ " " ++ description.first_line() ++
if(current_working_copy, " (current)")"""
"#,
);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
std::fs::write(main_path.join("file"), "contents").unwrap();
test_env.jj_cmd_ok(&main_path, &["commit", "-m", "initial"]);
test_env.jj_cmd_ok(
&main_path,
&["workspace", "add", "--name", "second", "../secondary"],
);
// "current_working_copy" should point to the workspace we operate on
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: e0e6d5672858 (current)
second: f68da2d114f1
"###);
let stdout = test_env.jj_cmd_success(&secondary_path, &["workspace", "list"]);
insta::assert_snapshot!(stdout, @r###"
default: e0e6d5672858
second: f68da2d114f1 (current)
"###);
}
/// Test getting the workspace root from primary and secondary workspaces
#[test]
fn test_workspaces_root() {
let test_env = TestEnvironment::default();
test_env.jj_cmd_ok(test_env.env_root(), &["init", "--git", "main"]);
let main_path = test_env.env_root().join("main");
let secondary_path = test_env.env_root().join("secondary");
let stdout = test_env.jj_cmd_success(&main_path, &["workspace", "root"]);
insta::assert_snapshot!(stdout, @r###"
$TEST_ENV/main
"###);
let main_subdir_path = main_path.join("subdir");
std::fs::create_dir(&main_subdir_path).unwrap();
let stdout = test_env.jj_cmd_success(&main_subdir_path, &["workspace", "root"]);
insta::assert_snapshot!(stdout, @r###"
$TEST_ENV/main
"###);
test_env.jj_cmd_ok(
&main_path,
&["workspace", "add", "--name", "secondary", "../secondary"],
);
let stdout = test_env.jj_cmd_success(&secondary_path, &["workspace", "root"]);
insta::assert_snapshot!(stdout, @r###"
$TEST_ENV/secondary
"###);
let secondary_subdir_path = secondary_path.join("subdir");
std::fs::create_dir(&secondary_subdir_path).unwrap();
let stdout = test_env.jj_cmd_success(&secondary_subdir_path, &["workspace", "root"]);
insta::assert_snapshot!(stdout, @r###"
$TEST_ENV/secondary
"###);
}
fn get_log_output(test_env: &TestEnvironment, cwd: &Path) -> String {
let template = r#"
separate(" ",
commit_id.short(),
working_copies,
if(divergent, "(divergent)"),
)
"#;
test_env.jj_cmd_success(cwd, &["log", "-T", template, "-r", "all()"])
}