Show More
@@ -1,376 +1,484 b'' | |||||
1 | The *dirstate* is what Mercurial uses internally to track |
|
1 | The *dirstate* is what Mercurial uses internally to track | |
2 | the state of files in the working directory, |
|
2 | the state of files in the working directory, | |
3 | such as set by commands like `hg add` and `hg rm`. |
|
3 | such as set by commands like `hg add` and `hg rm`. | |
4 | It also contains some cached data that help make `hg status` faster. |
|
4 | It also contains some cached data that help make `hg status` faster. | |
5 | The name refers both to `.hg/dirstate` on the filesystem |
|
5 | The name refers both to `.hg/dirstate` on the filesystem | |
6 | and the corresponding data structure in memory while a Mercurial process |
|
6 | and the corresponding data structure in memory while a Mercurial process | |
7 | is running. |
|
7 | is running. | |
8 |
|
8 | |||
9 | The original file format, retroactively dubbed `dirstate-v1`, |
|
9 | The original file format, retroactively dubbed `dirstate-v1`, | |
10 | is described at https://www.mercurial-scm.org/wiki/DirState. |
|
10 | is described at https://www.mercurial-scm.org/wiki/DirState. | |
11 | It is made of a flat sequence of unordered variable-size entries, |
|
11 | It is made of a flat sequence of unordered variable-size entries, | |
12 | so accessing any information in it requires parsing all of it. |
|
12 | so accessing any information in it requires parsing all of it. | |
13 | Similarly, saving changes requires rewriting the entire file. |
|
13 | Similarly, saving changes requires rewriting the entire file. | |
14 |
|
14 | |||
15 | The newer `dirsate-v2` file format is designed to fix these limitations |
|
15 | The newer `dirsate-v2` file format is designed to fix these limitations | |
16 | and make `hg status` faster. |
|
16 | and make `hg status` faster. | |
17 |
|
17 | |||
18 | User guide |
|
18 | User guide | |
19 | ========== |
|
19 | ========== | |
20 |
|
20 | |||
21 | Compatibility |
|
21 | Compatibility | |
22 | ------------- |
|
22 | ------------- | |
23 |
|
23 | |||
24 | The file format is experimental and may still change. |
|
24 | The file format is experimental and may still change. | |
25 | Different versions of Mercurial may not be compatible with each other |
|
25 | Different versions of Mercurial may not be compatible with each other | |
26 | when working on a local repository that uses this format. |
|
26 | when working on a local repository that uses this format. | |
27 | When using an incompatible version with the experimental format, |
|
27 | When using an incompatible version with the experimental format, | |
28 | anything can happen including data corruption. |
|
28 | anything can happen including data corruption. | |
29 |
|
29 | |||
30 | Since the dirstate is entirely local and not relevant to the wire protocol, |
|
30 | Since the dirstate is entirely local and not relevant to the wire protocol, | |
31 | `dirstate-v2` does not affect compatibility with remote Mercurial versions. |
|
31 | `dirstate-v2` does not affect compatibility with remote Mercurial versions. | |
32 |
|
32 | |||
33 | When `share-safe` is enabled, different repositories sharing the same store |
|
33 | When `share-safe` is enabled, different repositories sharing the same store | |
34 | can use different dirstate formats. |
|
34 | can use different dirstate formats. | |
35 |
|
35 | |||
36 | Enabling `dirsate-v2` for new local repositories |
|
36 | Enabling `dirsate-v2` for new local repositories | |
37 | ------------------------------------------------ |
|
37 | ------------------------------------------------ | |
38 |
|
38 | |||
39 | When creating a new local repository such as with `hg init` or `hg clone`, |
|
39 | When creating a new local repository such as with `hg init` or `hg clone`, | |
40 | the `exp-dirstate-v2` boolean in the `format` configuration section |
|
40 | the `exp-dirstate-v2` boolean in the `format` configuration section | |
41 | controls whether to use this file format. |
|
41 | controls whether to use this file format. | |
42 | This is disabled by default as of this writing. |
|
42 | This is disabled by default as of this writing. | |
43 | To enable it for a single repository, run for example:: |
|
43 | To enable it for a single repository, run for example:: | |
44 |
|
44 | |||
45 | $ hg init my-project --config format.exp-dirstate-v2=1 |
|
45 | $ hg init my-project --config format.exp-dirstate-v2=1 | |
46 |
|
46 | |||
47 | Checking the format of an existing local repsitory |
|
47 | Checking the format of an existing local repsitory | |
48 | -------------------------------------------------- |
|
48 | -------------------------------------------------- | |
49 |
|
49 | |||
50 | The `debugformat` commands prints information about |
|
50 | The `debugformat` commands prints information about | |
51 | which of multiple optional formats are used in the current repository, |
|
51 | which of multiple optional formats are used in the current repository, | |
52 | including `dirstate-v2`:: |
|
52 | including `dirstate-v2`:: | |
53 |
|
53 | |||
54 | $ hg debugformat |
|
54 | $ hg debugformat | |
55 | format-variant repo |
|
55 | format-variant repo | |
56 | fncache: yes |
|
56 | fncache: yes | |
57 | dirstate-v2: yes |
|
57 | dirstate-v2: yes | |
58 | [β¦] |
|
58 | [β¦] | |
59 |
|
59 | |||
60 | Upgrading or downgrading an existing local repository |
|
60 | Upgrading or downgrading an existing local repository | |
61 | ----------------------------------------------------- |
|
61 | ----------------------------------------------------- | |
62 |
|
62 | |||
63 | The `debugupgrade` command does various upgrades or downgrades |
|
63 | The `debugupgrade` command does various upgrades or downgrades | |
64 | on a local repository |
|
64 | on a local repository | |
65 | based on the current Mercurial version and on configuration. |
|
65 | based on the current Mercurial version and on configuration. | |
66 | The same `format.exp-dirstate-v2` configuration is used again. |
|
66 | The same `format.exp-dirstate-v2` configuration is used again. | |
67 |
|
67 | |||
68 | Example to upgrade:: |
|
68 | Example to upgrade:: | |
69 |
|
69 | |||
70 | $ hg debugupgrade --config format.exp-dirstate-v2=1 |
|
70 | $ hg debugupgrade --config format.exp-dirstate-v2=1 | |
71 |
|
71 | |||
72 | Example to downgrade to `dirstate-v1`:: |
|
72 | Example to downgrade to `dirstate-v1`:: | |
73 |
|
73 | |||
74 | $ hg debugupgrade --config format.exp-dirstate-v2=0 |
|
74 | $ hg debugupgrade --config format.exp-dirstate-v2=0 | |
75 |
|
75 | |||
76 | Both of this commands do nothing but print a list of proposed changes, |
|
76 | Both of this commands do nothing but print a list of proposed changes, | |
77 | which may include changes unrelated to the dirstate. |
|
77 | which may include changes unrelated to the dirstate. | |
78 | Those other changes are controlled by their own configuration keys. |
|
78 | Those other changes are controlled by their own configuration keys. | |
79 | Add `--run` to a command to actually apply the proposed changes. |
|
79 | Add `--run` to a command to actually apply the proposed changes. | |
80 |
|
80 | |||
81 | Backups of `.hg/requires` and `.hg/dirstate` are created |
|
81 | Backups of `.hg/requires` and `.hg/dirstate` are created | |
82 | in a `.hg/upgradebackup.*` directory. |
|
82 | in a `.hg/upgradebackup.*` directory. | |
83 | If something goes wrong, restoring those files should undo the change. |
|
83 | If something goes wrong, restoring those files should undo the change. | |
84 |
|
84 | |||
85 | Note that upgrading affects compatibility with older versions of Mercurial |
|
85 | Note that upgrading affects compatibility with older versions of Mercurial | |
86 | as noted above. |
|
86 | as noted above. | |
87 | This can be relevant when a repositoryβs files are on a USB drive |
|
87 | This can be relevant when a repositoryβs files are on a USB drive | |
88 | or some other removable media, or shared over the network, etc. |
|
88 | or some other removable media, or shared over the network, etc. | |
89 |
|
89 | |||
90 | Internal filesystem representation |
|
90 | Internal filesystem representation | |
91 | ================================== |
|
91 | ================================== | |
92 |
|
92 | |||
93 | Requirements file |
|
93 | Requirements file | |
94 | ----------------- |
|
94 | ----------------- | |
95 |
|
95 | |||
96 | The `.hg/requires` file indicates which of various optional file formats |
|
96 | The `.hg/requires` file indicates which of various optional file formats | |
97 | are used by a given repository. |
|
97 | are used by a given repository. | |
98 | Mercurial aborts when seeing a requirement it does not know about, |
|
98 | Mercurial aborts when seeing a requirement it does not know about, | |
99 | which avoids older version accidentally messing up a respository |
|
99 | which avoids older version accidentally messing up a respository | |
100 | that uses a format that was introduced later. |
|
100 | that uses a format that was introduced later. | |
101 | For versions that do support a format, the presence or absence of |
|
101 | For versions that do support a format, the presence or absence of | |
102 | the corresponding requirement indicates whether to use that format. |
|
102 | the corresponding requirement indicates whether to use that format. | |
103 |
|
103 | |||
104 | When the file contains a `exp-dirstate-v2` line, |
|
104 | When the file contains a `exp-dirstate-v2` line, | |
105 | the `dirstate-v2` format is used. |
|
105 | the `dirstate-v2` format is used. | |
106 | With no such line `dirstate-v1` is used. |
|
106 | With no such line `dirstate-v1` is used. | |
107 |
|
107 | |||
108 | High level description |
|
108 | High level description | |
109 | ---------------------- |
|
109 | ---------------------- | |
110 |
|
110 | |||
111 | Whereas `dirstate-v1` uses a single `.hg/disrtate` file, |
|
111 | Whereas `dirstate-v1` uses a single `.hg/disrtate` file, | |
112 | in `dirstate-v2` that file is a "docket" file |
|
112 | in `dirstate-v2` that file is a "docket" file | |
113 | that only contains some metadata |
|
113 | that only contains some metadata | |
114 | and points to separate data file named `.hg/dirstate.{ID}`, |
|
114 | and points to separate data file named `.hg/dirstate.{ID}`, | |
115 | where `{ID}` is a random identifier. |
|
115 | where `{ID}` is a random identifier. | |
116 |
|
116 | |||
117 | This separation allows making data files append-only |
|
117 | This separation allows making data files append-only | |
118 | and therefore safer to memory-map. |
|
118 | and therefore safer to memory-map. | |
119 | Creating a new data file (occasionally to clean up unused data) |
|
119 | Creating a new data file (occasionally to clean up unused data) | |
120 | can be done with a different ID |
|
120 | can be done with a different ID | |
121 | without disrupting another Mercurial process |
|
121 | without disrupting another Mercurial process | |
122 | that could still be using the previous data file. |
|
122 | that could still be using the previous data file. | |
123 |
|
123 | |||
124 | Both files have a format designed to reduce the need for parsing, |
|
124 | Both files have a format designed to reduce the need for parsing, | |
125 | by using fixed-size binary components as much as possible. |
|
125 | by using fixed-size binary components as much as possible. | |
126 | For data that is not fixed-size, |
|
126 | For data that is not fixed-size, | |
127 | references to other parts of a file can be made by storing "pseudo-pointers": |
|
127 | references to other parts of a file can be made by storing "pseudo-pointers": | |
128 | integers counted in bytes from the start of a file. |
|
128 | integers counted in bytes from the start of a file. | |
129 | For read-only access no data structure is needed, |
|
129 | For read-only access no data structure is needed, | |
130 | only a bytes buffer (possibly memory-mapped directly from the filesystem) |
|
130 | only a bytes buffer (possibly memory-mapped directly from the filesystem) | |
131 | with specific parts read on demand. |
|
131 | with specific parts read on demand. | |
132 |
|
132 | |||
133 | The data file contains "nodes" organized in a tree. |
|
133 | The data file contains "nodes" organized in a tree. | |
134 | Each node represents a file or directory inside the working directory |
|
134 | Each node represents a file or directory inside the working directory | |
135 | or its parent changeset. |
|
135 | or its parent changeset. | |
136 | This tree has the same structure as the filesystem, |
|
136 | This tree has the same structure as the filesystem, | |
137 | so a node representing a directory has child nodes representing |
|
137 | so a node representing a directory has child nodes representing | |
138 | the files and subdirectories contained directly in that directory. |
|
138 | the files and subdirectories contained directly in that directory. | |
139 |
|
139 | |||
140 | The docket file format |
|
140 | The docket file format | |
141 | ---------------------- |
|
141 | ---------------------- | |
142 |
|
142 | |||
143 | This is implemented in `rust/hg-core/src/dirstate_tree/on_disk.rs` |
|
143 | This is implemented in `rust/hg-core/src/dirstate_tree/on_disk.rs` | |
144 | and `mercurial/dirstateutils/docket.py`. |
|
144 | and `mercurial/dirstateutils/docket.py`. | |
145 |
|
145 | |||
146 | Components of the docket file are found at fixed offsets, |
|
146 | Components of the docket file are found at fixed offsets, | |
147 | counted in bytes from the start of the file: |
|
147 | counted in bytes from the start of the file: | |
148 |
|
148 | |||
149 | * Offset 0: |
|
149 | * Offset 0: | |
150 | The 12-bytes marker string "dirstate-v2\n" ending with a newline character. |
|
150 | The 12-bytes marker string "dirstate-v2\n" ending with a newline character. | |
151 | This makes it easier to tell a dirstate-v2 file from a dirstate-v1 file, |
|
151 | This makes it easier to tell a dirstate-v2 file from a dirstate-v1 file, | |
152 | although it is not strictly necessary |
|
152 | although it is not strictly necessary | |
153 | since `.hg/requires` determines which format to use. |
|
153 | since `.hg/requires` determines which format to use. | |
154 |
|
154 | |||
155 | * Offset 12: |
|
155 | * Offset 12: | |
156 | The changeset node ID on the first parent of the working directory, |
|
156 | The changeset node ID on the first parent of the working directory, | |
157 | as up to 32 binary bytes. |
|
157 | as up to 32 binary bytes. | |
158 | If a node ID is shorter (20 bytes for SHA-1), |
|
158 | If a node ID is shorter (20 bytes for SHA-1), | |
159 | it is start-aligned and the rest of the bytes are set to zero. |
|
159 | it is start-aligned and the rest of the bytes are set to zero. | |
160 |
|
160 | |||
161 | * Offset 44: |
|
161 | * Offset 44: | |
162 | The changeset node ID on the second parent of the working directory, |
|
162 | The changeset node ID on the second parent of the working directory, | |
163 | or all zeros if there isnβt one. |
|
163 | or all zeros if there isnβt one. | |
164 | Also 32 binary bytes. |
|
164 | Also 32 binary bytes. | |
165 |
|
165 | |||
166 | * Offset 76: |
|
166 | * Offset 76: | |
167 | Tree metadata on 44 bytes, described below. |
|
167 | Tree metadata on 44 bytes, described below. | |
168 | Its separation in this documentation from the rest of the docket |
|
168 | Its separation in this documentation from the rest of the docket | |
169 | reflects a detail of the current implementation. |
|
169 | reflects a detail of the current implementation. | |
170 | Since tree metadata is also made of fields at fixed offsets, those could |
|
170 | Since tree metadata is also made of fields at fixed offsets, those could | |
171 | be inlined here by adding 76 bytes to each offset. |
|
171 | be inlined here by adding 76 bytes to each offset. | |
172 |
|
172 | |||
173 | * Offset 120: |
|
173 | * Offset 120: | |
174 | The used size of the data file, as a 32-bit big-endian integer. |
|
174 | The used size of the data file, as a 32-bit big-endian integer. | |
175 | The actual size of the data file may be larger |
|
175 | The actual size of the data file may be larger | |
176 | (if another Mercurial processis in appending to it |
|
176 | (if another Mercurial processis in appending to it | |
177 | but has not updated the docket yet). |
|
177 | but has not updated the docket yet). | |
178 | That extra data must be ignored. |
|
178 | That extra data must be ignored. | |
179 |
|
179 | |||
180 | * Offset 124: |
|
180 | * Offset 124: | |
181 | The length of the data file identifier, as a 8-bit integer. |
|
181 | The length of the data file identifier, as a 8-bit integer. | |
182 |
|
182 | |||
183 | * Offset 125: |
|
183 | * Offset 125: | |
184 | The data file identifier. |
|
184 | The data file identifier. | |
185 |
|
185 | |||
186 | * Any additional data is current ignored, and dropped when updating the file. |
|
186 | * Any additional data is current ignored, and dropped when updating the file. | |
187 |
|
187 | |||
188 | Tree metadata in the docket file |
|
188 | Tree metadata in the docket file | |
189 | -------------------------------- |
|
189 | -------------------------------- | |
190 |
|
190 | |||
191 | Tree metadata is similarly made of components at fixed offsets. |
|
191 | Tree metadata is similarly made of components at fixed offsets. | |
192 | These offsets are counted in bytes from the start of tree metadata, |
|
192 | These offsets are counted in bytes from the start of tree metadata, | |
193 | which is 76 bytes after the start of the docket file. |
|
193 | which is 76 bytes after the start of the docket file. | |
194 |
|
194 | |||
195 | This metadata can be thought of as the singular root of the tree |
|
195 | This metadata can be thought of as the singular root of the tree | |
196 | formed by nodes in the data file. |
|
196 | formed by nodes in the data file. | |
197 |
|
197 | |||
198 | * Offset 0: |
|
198 | * Offset 0: | |
199 | Pseudo-pointer to the start of root nodes, |
|
199 | Pseudo-pointer to the start of root nodes, | |
200 | counted in bytes from the start of the data file, |
|
200 | counted in bytes from the start of the data file, | |
201 | as a 32-bit big-endian integer. |
|
201 | as a 32-bit big-endian integer. | |
202 | These nodes describe files and directories found directly |
|
202 | These nodes describe files and directories found directly | |
203 | at the root of the working directory. |
|
203 | at the root of the working directory. | |
204 |
|
204 | |||
205 | * Offset 4: |
|
205 | * Offset 4: | |
206 | Number of root nodes, as a 32-bit big-endian integer. |
|
206 | Number of root nodes, as a 32-bit big-endian integer. | |
207 |
|
207 | |||
208 | * Offset 8: |
|
208 | * Offset 8: | |
209 | Total number of nodes in the entire tree that "have a dirstate entry", |
|
209 | Total number of nodes in the entire tree that "have a dirstate entry", | |
210 | as a 32-bit big-endian integer. |
|
210 | as a 32-bit big-endian integer. | |
211 | Those nodes represent files that would be present at all in `dirstate-v1`. |
|
211 | Those nodes represent files that would be present at all in `dirstate-v1`. | |
212 | This is typically less than the total number of nodes. |
|
212 | This is typically less than the total number of nodes. | |
213 | This counter is used to implement `len(dirstatemap)`. |
|
213 | This counter is used to implement `len(dirstatemap)`. | |
214 |
|
214 | |||
215 | * Offset 12: |
|
215 | * Offset 12: | |
216 | Number of nodes in the entire tree that have a copy source, |
|
216 | Number of nodes in the entire tree that have a copy source, | |
217 | as a 32-bit big-endian integer. |
|
217 | as a 32-bit big-endian integer. | |
218 | At the next commit, these files are recorded |
|
218 | At the next commit, these files are recorded | |
219 | as having been copied or moved/renamed from that source. |
|
219 | as having been copied or moved/renamed from that source. | |
220 | (A move is recorded as a copy and separate removal of the source.) |
|
220 | (A move is recorded as a copy and separate removal of the source.) | |
221 | This counter is used to implement `len(dirstatemap.copymap)`. |
|
221 | This counter is used to implement `len(dirstatemap.copymap)`. | |
222 |
|
222 | |||
223 | * Offset 16: |
|
223 | * Offset 16: | |
224 | An estimation of how many bytes of the data file |
|
224 | An estimation of how many bytes of the data file | |
225 | (within its used size) are unused, as a 32-bit big-endian integer. |
|
225 | (within its used size) are unused, as a 32-bit big-endian integer. | |
226 | When appending to an existing data file, |
|
226 | When appending to an existing data file, | |
227 | some existing nodes or paths can be unreachable from the new root |
|
227 | some existing nodes or paths can be unreachable from the new root | |
228 | but they still take up space. |
|
228 | but they still take up space. | |
229 | This counter is used to decide when to write a new data file from scratch |
|
229 | This counter is used to decide when to write a new data file from scratch | |
230 | instead of appending to an existing one, |
|
230 | instead of appending to an existing one, | |
231 | in order to get rid of that unreachable data |
|
231 | in order to get rid of that unreachable data | |
232 | and avoid unbounded file size growth. |
|
232 | and avoid unbounded file size growth. | |
233 |
|
233 | |||
234 | * Offset 20: |
|
234 | * Offset 20: | |
235 | These four bytes are currently ignored |
|
235 | These four bytes are currently ignored | |
236 | and reset to zero when updating a docket file. |
|
236 | and reset to zero when updating a docket file. | |
237 | This is an attempt at forward compatibility: |
|
237 | This is an attempt at forward compatibility: | |
238 | future Mercurial versions could use this as a bit field |
|
238 | future Mercurial versions could use this as a bit field | |
239 | to indicate that a dirstate has additional data or constraints. |
|
239 | to indicate that a dirstate has additional data or constraints. | |
240 | Finding a dirstate file with the relevant bit unset indicates that |
|
240 | Finding a dirstate file with the relevant bit unset indicates that | |
241 | it was written by a then-older version |
|
241 | it was written by a then-older version | |
242 | which is not aware of that future change. |
|
242 | which is not aware of that future change. | |
243 |
|
243 | |||
244 | * Offset 24: |
|
244 | * Offset 24: | |
245 | Either 20 zero bytes, or a SHA-1 hash as 20 binary bytes. |
|
245 | Either 20 zero bytes, or a SHA-1 hash as 20 binary bytes. | |
246 | When present, the hash is of ignore patterns |
|
246 | When present, the hash is of ignore patterns | |
247 | that were used for some previous run of the `status` algorithm. |
|
247 | that were used for some previous run of the `status` algorithm. | |
248 |
|
248 | |||
249 | * (Offset 44: end of tree metadata) |
|
249 | * (Offset 44: end of tree metadata) | |
250 |
|
250 | |||
251 | Optional hash of ignore patterns |
|
251 | Optional hash of ignore patterns | |
252 | -------------------------------- |
|
252 | -------------------------------- | |
253 |
|
253 | |||
254 | The implementation of `status` at `rust/hg-core/src/dirstate_tree/status.rs` |
|
254 | The implementation of `status` at `rust/hg-core/src/dirstate_tree/status.rs` | |
255 | has been optimized such that its run time is dominated by calls |
|
255 | has been optimized such that its run time is dominated by calls | |
256 | to `stat` for reading the filesystem metadata of a file or directory, |
|
256 | to `stat` for reading the filesystem metadata of a file or directory, | |
257 | and to `readdir` for listing the contents of a directory. |
|
257 | and to `readdir` for listing the contents of a directory. | |
258 | In some cases the algorithm can skip calls to `readdir` |
|
258 | In some cases the algorithm can skip calls to `readdir` | |
259 | (saving significant time) |
|
259 | (saving significant time) | |
260 | because the dirstate already contains enough of the relevant information |
|
260 | because the dirstate already contains enough of the relevant information | |
261 | to build the correct `status` results. |
|
261 | to build the correct `status` results. | |
262 |
|
262 | |||
263 | The default configuration of `hg status` is to list unknown files |
|
263 | The default configuration of `hg status` is to list unknown files | |
264 | but not ignored files. |
|
264 | but not ignored files. | |
265 | In this case, it matters for the `readdir`-skipping optimization |
|
265 | In this case, it matters for the `readdir`-skipping optimization | |
266 | if a given file used to be ignored but became unknown |
|
266 | if a given file used to be ignored but became unknown | |
267 | because `.hgignore` changed. |
|
267 | because `.hgignore` changed. | |
268 | To detect the possibility of such a change, |
|
268 | To detect the possibility of such a change, | |
269 | the tree metadata contains an optional hash of all ignore patterns. |
|
269 | the tree metadata contains an optional hash of all ignore patterns. | |
270 |
|
270 | |||
271 | We define: |
|
271 | We define: | |
272 |
|
272 | |||
273 | * "Root" ignore files as: |
|
273 | * "Root" ignore files as: | |
274 |
|
274 | |||
275 | - `.hgignore` at the root of the repository if it exists |
|
275 | - `.hgignore` at the root of the repository if it exists | |
276 | - And all files from `ui.ignore.*` config. |
|
276 | - And all files from `ui.ignore.*` config. | |
277 |
|
277 | |||
278 | This set of files is sorted by the string representation of their path. |
|
278 | This set of files is sorted by the string representation of their path. | |
279 |
|
279 | |||
280 | * The "expanded contents" of an ignore files is the byte string made |
|
280 | * The "expanded contents" of an ignore files is the byte string made | |
281 | by the concatenation of its contents followed by the "expanded contents" |
|
281 | by the concatenation of its contents followed by the "expanded contents" | |
282 | of other files included with `include:` or `subinclude:` directives, |
|
282 | of other files included with `include:` or `subinclude:` directives, | |
283 | in inclusion order. This definition is recursive, as included files can |
|
283 | in inclusion order. This definition is recursive, as included files can | |
284 | themselves include more files. |
|
284 | themselves include more files. | |
285 |
|
285 | |||
286 | This hash is defined as the SHA-1 of the concatenation (in sorted |
|
286 | This hash is defined as the SHA-1 of the concatenation (in sorted | |
287 | order) of the "expanded contents" of each "root" ignore file. |
|
287 | order) of the "expanded contents" of each "root" ignore file. | |
288 | (Note that computing this does not require actually concatenating |
|
288 | (Note that computing this does not require actually concatenating | |
289 | into a single contiguous byte sequence. |
|
289 | into a single contiguous byte sequence. | |
290 | Instead a SHA-1 hasher object can be created |
|
290 | Instead a SHA-1 hasher object can be created | |
291 | and fed separate chunks one by one.) |
|
291 | and fed separate chunks one by one.) | |
292 |
|
292 | |||
293 | The data file format |
|
293 | The data file format | |
294 | -------------------- |
|
294 | -------------------- | |
295 |
|
295 | |||
296 | This is implemented in `rust/hg-core/src/dirstate_tree/on_disk.rs` |
|
296 | This is implemented in `rust/hg-core/src/dirstate_tree/on_disk.rs` | |
297 | and `mercurial/dirstateutils/v2.py`. |
|
297 | and `mercurial/dirstateutils/v2.py`. | |
298 |
|
298 | |||
299 | The data file contains two types of data: paths and nodes. |
|
299 | The data file contains two types of data: paths and nodes. | |
300 |
|
300 | |||
301 | Paths and nodes can be organized in any order in the file, except that sibling |
|
301 | Paths and nodes can be organized in any order in the file, except that sibling | |
302 | nodes must be next to each other and sorted by their path. |
|
302 | nodes must be next to each other and sorted by their path. | |
303 | Contiguity lets the parent refer to them all |
|
303 | Contiguity lets the parent refer to them all | |
304 | by their count and a single pseudo-pointer, |
|
304 | by their count and a single pseudo-pointer, | |
305 | instead of storing one pseudo-pointer per child node. |
|
305 | instead of storing one pseudo-pointer per child node. | |
306 | Sorting allows using binary seach to find a child node with a given name |
|
306 | Sorting allows using binary seach to find a child node with a given name | |
307 | in `O(log(n))` byte sequence comparisons. |
|
307 | in `O(log(n))` byte sequence comparisons. | |
308 |
|
308 | |||
309 | The current implemention writes paths and child node before a given node |
|
309 | The current implemention writes paths and child node before a given node | |
310 | for ease of figuring out the value of pseudo-pointers by the time the are to be |
|
310 | for ease of figuring out the value of pseudo-pointers by the time the are to be | |
311 | written, but this is not an obligation and readers must not rely on it. |
|
311 | written, but this is not an obligation and readers must not rely on it. | |
312 |
|
312 | |||
313 | A path is stored as a byte string anywhere in the file, without delimiter. |
|
313 | A path is stored as a byte string anywhere in the file, without delimiter. | |
314 | It is refered to by one or more node by a pseudo-pointer to its start, and its |
|
314 | It is refered to by one or more node by a pseudo-pointer to its start, and its | |
315 | length in bytes. Since there is no delimiter, |
|
315 | length in bytes. Since there is no delimiter, | |
316 | when a path is a substring of another the same bytes could be reused, |
|
316 | when a path is a substring of another the same bytes could be reused, | |
317 | although the implementation does not exploit this as of this writing. |
|
317 | although the implementation does not exploit this as of this writing. | |
318 |
|
318 | |||
319 | A node is stored on 43 bytes with components at fixed offsets. Paths and |
|
319 | A node is stored on 43 bytes with components at fixed offsets. Paths and | |
320 | child nodes relevant to a node are stored externally and referenced though |
|
320 | child nodes relevant to a node are stored externally and referenced though | |
321 | pseudo-pointers. |
|
321 | pseudo-pointers. | |
322 |
|
322 | |||
323 | All integers are stored in big-endian. All pseudo-pointers are 32-bit integers |
|
323 | All integers are stored in big-endian. All pseudo-pointers are 32-bit integers | |
324 | counting bytes from the start of the data file. Path lengths and positions |
|
324 | counting bytes from the start of the data file. Path lengths and positions | |
325 | are 16-bit integers, also counted in bytes. |
|
325 | are 16-bit integers, also counted in bytes. | |
326 |
|
326 | |||
327 | Node components are: |
|
327 | Node components are: | |
328 |
|
328 | |||
329 | * Offset 0: |
|
329 | * Offset 0: | |
330 | Pseudo-pointer to the full path of this node, |
|
330 | Pseudo-pointer to the full path of this node, | |
331 | from the working directory root. |
|
331 | from the working directory root. | |
332 |
|
332 | |||
333 | * Offset 4: |
|
333 | * Offset 4: | |
334 | Length of the full path. |
|
334 | Length of the full path. | |
335 |
|
335 | |||
336 | * Offset 6: |
|
336 | * Offset 6: | |
337 | Position of the last `/` path separator within the full path, |
|
337 | Position of the last `/` path separator within the full path, | |
338 | in bytes from the start of the full path, |
|
338 | in bytes from the start of the full path, | |
339 | or zero if there isnβt one. |
|
339 | or zero if there isnβt one. | |
340 | The part of the full path after this position is the "base name". |
|
340 | The part of the full path after this position is the "base name". | |
341 | Since sibling nodes have the same parent, only their base name vary |
|
341 | Since sibling nodes have the same parent, only their base name vary | |
342 | and needs to be considered when doing binary search to find a given path. |
|
342 | and needs to be considered when doing binary search to find a given path. | |
343 |
|
343 | |||
344 | * Offset 8: |
|
344 | * Offset 8: | |
345 | Pseudo-pointer to the "copy source" path for this node, |
|
345 | Pseudo-pointer to the "copy source" path for this node, | |
346 | or zero if there is no copy source. |
|
346 | or zero if there is no copy source. | |
347 |
|
347 | |||
348 | * Offset 12: |
|
348 | * Offset 12: | |
349 | Length of the copy source path, or zero if there isnβt one. |
|
349 | Length of the copy source path, or zero if there isnβt one. | |
350 |
|
350 | |||
351 | * Offset 14: |
|
351 | * Offset 14: | |
352 | Pseudo-pointer to the start of child nodes. |
|
352 | Pseudo-pointer to the start of child nodes. | |
353 |
|
353 | |||
354 | * Offset 18: |
|
354 | * Offset 18: | |
355 | Number of child nodes, as a 32-bit integer. |
|
355 | Number of child nodes, as a 32-bit integer. | |
356 | They occupy 43 times this number of bytes |
|
356 | They occupy 43 times this number of bytes | |
357 | (not counting space for paths, and further descendants). |
|
357 | (not counting space for paths, and further descendants). | |
358 |
|
358 | |||
359 | * Offset 22: |
|
359 | * Offset 22: | |
360 | Number as a 32-bit integer of descendant nodes in this subtree, |
|
360 | Number as a 32-bit integer of descendant nodes in this subtree, | |
361 | not including this node itself, |
|
361 | not including this node itself, | |
362 | that "have a dirstate entry". |
|
362 | that "have a dirstate entry". | |
363 | Those nodes represent files that would be present at all in `dirstate-v1`. |
|
363 | Those nodes represent files that would be present at all in `dirstate-v1`. | |
364 | This is typically less than the total number of descendants. |
|
364 | This is typically less than the total number of descendants. | |
365 | This counter is used to implement `has_dir`. |
|
365 | This counter is used to implement `has_dir`. | |
366 |
|
366 | |||
367 | * Offset 26: |
|
367 | * Offset 26: | |
368 | Number as a 32-bit integer of descendant nodes in this subtree, |
|
368 | Number as a 32-bit integer of descendant nodes in this subtree, | |
369 | not including this node itself, |
|
369 | not including this node itself, | |
370 | that represent files tracked in the working directory. |
|
370 | that represent files tracked in the working directory. | |
371 | (For example, `hg rm` makes a file untracked.) |
|
371 | (For example, `hg rm` makes a file untracked.) | |
372 | This counter is used to implement `has_tracked_dir`. |
|
372 | This counter is used to implement `has_tracked_dir`. | |
373 |
|
373 | |||
374 |
* Offset 30 |
|
374 | * Offset 30: | |
375 | **TODO:** docs not written yet |
|
375 | Some boolean values packed as bits of a single byte. | |
376 | as this part of the format might be changing soon. |
|
376 | Starting from least-significant, bit masks are:: | |
|
377 | ||||
|
378 | WDIR_TRACKED = 1 << 0 | |||
|
379 | P1_TRACKED = 1 << 1 | |||
|
380 | P2_INFO = 1 << 2 | |||
|
381 | HAS_MODE_AND_SIZE = 1 << 3 | |||
|
382 | HAS_MTIME = 1 << 4 | |||
|
383 | ||||
|
384 | Other bits are unset. The meaning of these bits are: | |||
|
385 | ||||
|
386 | `WDIR_TRACKED` | |||
|
387 | Set if the working directory contains a tracked file at this nodeβs path. | |||
|
388 | This is typically set and unset by `hg add` and `hg rm`. | |||
|
389 | ||||
|
390 | `P1_TRACKED` | |||
|
391 | set if the working directoryβs first parent changeset | |||
|
392 | (whose node identifier is found in tree metadata) | |||
|
393 | contains a tracked file at this nodeβs path. | |||
|
394 | This is a cache to reduce manifest lookups. | |||
|
395 | ||||
|
396 | `P2_INFO` | |||
|
397 | Set if the file has been involved in some merge operation. | |||
|
398 | Either because it was actually merged, | |||
|
399 | or because the version in the second parent p2 version was ahead, | |||
|
400 | or because some rename moved it there. | |||
|
401 | In either case `hg status` will want it displayed as modified. | |||
|
402 | ||||
|
403 | Files that would be mentioned at all in the `dirstate-v1` file format | |||
|
404 | have a node with at least one of the above three bits set in `dirstate-v2`. | |||
|
405 | Letβs call these files "tracked anywhere", | |||
|
406 | and "untracked" the nodes with all three of these bits unset. | |||
|
407 | Untracked nodes are typically for directories: | |||
|
408 | they hold child nodes and form the tree structure. | |||
|
409 | Additional untracked nodes may also exist. | |||
|
410 | Although implementations should strive to clean up nodes | |||
|
411 | that are entirely unused, other untracked nodes may also exist. | |||
|
412 | For example, a future version of Mercurial might in some cases | |||
|
413 | add nodes for untracked files or/and ignored files in the working directory | |||
|
414 | in order to optimize `hg status` | |||
|
415 | by enabling it to skip `readdir` in more cases. | |||
|
416 | ||||
|
417 | When a node is for a file tracked anywhere, | |||
|
418 | the rest of the node data is three fields: | |||
|
419 | ||||
|
420 | * Offset 31: | |||
|
421 | If `HAS_MODE_AND_SIZE` is unset, four zero bytes. | |||
|
422 | Otherwise, a 32-bit integer for the Unix mode (as in `stat_result.st_mode`) | |||
|
423 | expected for this file to be considered clean. | |||
|
424 | Only the `S_IXUSR` bit (owner has execute permission) is considered. | |||
|
425 | ||||
|
426 | * Offset 35: | |||
|
427 | If `HAS_MTIME` is unset, four zero bytes. | |||
|
428 | Otherwise, a 32-bit integer for expected modified time of the file | |||
|
429 | (as in `stat_result.st_mtime`), | |||
|
430 | truncated to its 31 least-significant bits. | |||
|
431 | Unlike in dirstate-v1, negative values are not used. | |||
|
432 | ||||
|
433 | * Offset 39: | |||
|
434 | If `HAS_MODE_AND_SIZE` is unset, four zero bytes. | |||
|
435 | Otherwise, a 32-bit integer for expected size of the file | |||
|
436 | truncated to its 31 least-significant bits. | |||
|
437 | Unlike in dirstate-v1, negative values are not used. | |||
|
438 | ||||
|
439 | If an untracked node `HAS_MTIME` *unset*, this space is unused: | |||
|
440 | ||||
|
441 | * Offset 31: | |||
|
442 | 12 bytes set to zero | |||
|
443 | ||||
|
444 | If an untracked node `HAS_MTIME` *set*, | |||
|
445 | what follows is the modification time of a directory | |||
|
446 | represented with separated second and sub-second components | |||
|
447 | since the Unix epoch: | |||
|
448 | ||||
|
449 | * Offset 31: | |||
|
450 | The number of seconds as a signed (twoβs complement) 64-bit integer. | |||
|
451 | ||||
|
452 | * Offset 39: | |||
|
453 | The number of nanoseconds as 32-bit integer. | |||
|
454 | Always greater than or equal to zero, and strictly less than a billion. | |||
|
455 | Increasing this component makes the modification time | |||
|
456 | go forward or backward in time dependening | |||
|
457 | on the sign of the integral seconds components. | |||
|
458 | (Note: this is buggy because there is no negative zero integer, | |||
|
459 | but will be changed soon.) | |||
|
460 | ||||
|
461 | The presence of a directory modification time means that at some point, | |||
|
462 | this path in the working directory was observed: | |||
|
463 | ||||
|
464 | - To be a directory | |||
|
465 | - With the given modification time | |||
|
466 | - That time was already strictly in the past when observed, | |||
|
467 | meaning that later changes cannot happen in the same clock tick | |||
|
468 | and must cause a different modification time | |||
|
469 | (unless the system clock jumps back and we get unlucky, | |||
|
470 | which is not impossible but deemed unlikely enough). | |||
|
471 | - All direct children of this directory | |||
|
472 | (as returned by `std::fs::read_dir`) | |||
|
473 | either have a corresponding dirstate node, | |||
|
474 | or are ignored by ignore patterns whose hash is in tree metadata. | |||
|
475 | ||||
|
476 | This means that if `std::fs::symlink_metadata` later reports | |||
|
477 | the same modification time | |||
|
478 | and ignored patterns havenβt changed, | |||
|
479 | a run of status that is not listing ignored files | |||
|
480 | can skip calling `std::fs::read_dir` again for this directory, | |||
|
481 | and iterate child dirstate nodes instead. | |||
|
482 | ||||
|
483 | ||||
|
484 | * (Offset 43: end of this node) |
@@ -1,736 +1,736 b'' | |||||
1 | # parsers.py - Python implementation of parsers.c |
|
1 | # parsers.py - Python implementation of parsers.c | |
2 | # |
|
2 | # | |
3 | # Copyright 2009 Olivia Mackall <olivia@selenic.com> and others |
|
3 | # Copyright 2009 Olivia Mackall <olivia@selenic.com> and others | |
4 | # |
|
4 | # | |
5 | # This software may be used and distributed according to the terms of the |
|
5 | # This software may be used and distributed according to the terms of the | |
6 | # GNU General Public License version 2 or any later version. |
|
6 | # GNU General Public License version 2 or any later version. | |
7 |
|
7 | |||
8 | from __future__ import absolute_import |
|
8 | from __future__ import absolute_import | |
9 |
|
9 | |||
10 | import struct |
|
10 | import struct | |
11 | import zlib |
|
11 | import zlib | |
12 |
|
12 | |||
13 | from ..node import ( |
|
13 | from ..node import ( | |
14 | nullrev, |
|
14 | nullrev, | |
15 | sha1nodeconstants, |
|
15 | sha1nodeconstants, | |
16 | ) |
|
16 | ) | |
17 | from ..thirdparty import attr |
|
17 | from ..thirdparty import attr | |
18 | from .. import ( |
|
18 | from .. import ( | |
19 | error, |
|
19 | error, | |
20 | pycompat, |
|
20 | pycompat, | |
21 | revlogutils, |
|
21 | revlogutils, | |
22 | util, |
|
22 | util, | |
23 | ) |
|
23 | ) | |
24 |
|
24 | |||
25 | from ..revlogutils import nodemap as nodemaputil |
|
25 | from ..revlogutils import nodemap as nodemaputil | |
26 | from ..revlogutils import constants as revlog_constants |
|
26 | from ..revlogutils import constants as revlog_constants | |
27 |
|
27 | |||
28 | stringio = pycompat.bytesio |
|
28 | stringio = pycompat.bytesio | |
29 |
|
29 | |||
30 |
|
30 | |||
31 | _pack = struct.pack |
|
31 | _pack = struct.pack | |
32 | _unpack = struct.unpack |
|
32 | _unpack = struct.unpack | |
33 | _compress = zlib.compress |
|
33 | _compress = zlib.compress | |
34 | _decompress = zlib.decompress |
|
34 | _decompress = zlib.decompress | |
35 |
|
35 | |||
36 |
|
36 | |||
37 | # a special value used internally for `size` if the file come from the other parent |
|
37 | # a special value used internally for `size` if the file come from the other parent | |
38 | FROM_P2 = -2 |
|
38 | FROM_P2 = -2 | |
39 |
|
39 | |||
40 | # a special value used internally for `size` if the file is modified/merged/added |
|
40 | # a special value used internally for `size` if the file is modified/merged/added | |
41 | NONNORMAL = -1 |
|
41 | NONNORMAL = -1 | |
42 |
|
42 | |||
43 | # a special value used internally for `time` if the time is ambigeous |
|
43 | # a special value used internally for `time` if the time is ambigeous | |
44 | AMBIGUOUS_TIME = -1 |
|
44 | AMBIGUOUS_TIME = -1 | |
45 |
|
45 | |||
46 |
|
46 | |||
47 | @attr.s(slots=True, init=False) |
|
47 | @attr.s(slots=True, init=False) | |
48 | class DirstateItem(object): |
|
48 | class DirstateItem(object): | |
49 | """represent a dirstate entry |
|
49 | """represent a dirstate entry | |
50 |
|
50 | |||
51 | It hold multiple attributes |
|
51 | It hold multiple attributes | |
52 |
|
52 | |||
53 | # about file tracking |
|
53 | # about file tracking | |
54 | - wc_tracked: is the file tracked by the working copy |
|
54 | - wc_tracked: is the file tracked by the working copy | |
55 | - p1_tracked: is the file tracked in working copy first parent |
|
55 | - p1_tracked: is the file tracked in working copy first parent | |
56 | - p2_info: the file has been involved in some merge operation. Either |
|
56 | - p2_info: the file has been involved in some merge operation. Either | |
57 | because it was actually merged, or because the p2 version was |
|
57 | because it was actually merged, or because the p2 version was | |
58 |
ahead, or because some rename |
|
58 | ahead, or because some rename moved it there. In either case | |
59 | `hg status` will want it displayed as modified. |
|
59 | `hg status` will want it displayed as modified. | |
60 |
|
60 | |||
61 | # about the file state expected from p1 manifest: |
|
61 | # about the file state expected from p1 manifest: | |
62 | - mode: the file mode in p1 |
|
62 | - mode: the file mode in p1 | |
63 | - size: the file size in p1 |
|
63 | - size: the file size in p1 | |
64 |
|
64 | |||
65 | These value can be set to None, which mean we don't have a meaningful value |
|
65 | These value can be set to None, which mean we don't have a meaningful value | |
66 | to compare with. Either because we don't really care about them as there |
|
66 | to compare with. Either because we don't really care about them as there | |
67 | `status` is known without having to look at the disk or because we don't |
|
67 | `status` is known without having to look at the disk or because we don't | |
68 | know these right now and a full comparison will be needed to find out if |
|
68 | know these right now and a full comparison will be needed to find out if | |
69 | the file is clean. |
|
69 | the file is clean. | |
70 |
|
70 | |||
71 | # about the file state on disk last time we saw it: |
|
71 | # about the file state on disk last time we saw it: | |
72 | - mtime: the last known clean mtime for the file. |
|
72 | - mtime: the last known clean mtime for the file. | |
73 |
|
73 | |||
74 | This value can be set to None if no cachable state exist. Either because we |
|
74 | This value can be set to None if no cachable state exist. Either because we | |
75 | do not care (see previous section) or because we could not cache something |
|
75 | do not care (see previous section) or because we could not cache something | |
76 | yet. |
|
76 | yet. | |
77 | """ |
|
77 | """ | |
78 |
|
78 | |||
79 | _wc_tracked = attr.ib() |
|
79 | _wc_tracked = attr.ib() | |
80 | _p1_tracked = attr.ib() |
|
80 | _p1_tracked = attr.ib() | |
81 | _p2_info = attr.ib() |
|
81 | _p2_info = attr.ib() | |
82 | _mode = attr.ib() |
|
82 | _mode = attr.ib() | |
83 | _size = attr.ib() |
|
83 | _size = attr.ib() | |
84 | _mtime = attr.ib() |
|
84 | _mtime = attr.ib() | |
85 |
|
85 | |||
86 | def __init__( |
|
86 | def __init__( | |
87 | self, |
|
87 | self, | |
88 | wc_tracked=False, |
|
88 | wc_tracked=False, | |
89 | p1_tracked=False, |
|
89 | p1_tracked=False, | |
90 | p2_info=False, |
|
90 | p2_info=False, | |
91 | has_meaningful_data=True, |
|
91 | has_meaningful_data=True, | |
92 | has_meaningful_mtime=True, |
|
92 | has_meaningful_mtime=True, | |
93 | parentfiledata=None, |
|
93 | parentfiledata=None, | |
94 | ): |
|
94 | ): | |
95 | self._wc_tracked = wc_tracked |
|
95 | self._wc_tracked = wc_tracked | |
96 | self._p1_tracked = p1_tracked |
|
96 | self._p1_tracked = p1_tracked | |
97 | self._p2_info = p2_info |
|
97 | self._p2_info = p2_info | |
98 |
|
98 | |||
99 | self._mode = None |
|
99 | self._mode = None | |
100 | self._size = None |
|
100 | self._size = None | |
101 | self._mtime = None |
|
101 | self._mtime = None | |
102 | if parentfiledata is None: |
|
102 | if parentfiledata is None: | |
103 | has_meaningful_mtime = False |
|
103 | has_meaningful_mtime = False | |
104 | has_meaningful_data = False |
|
104 | has_meaningful_data = False | |
105 | if has_meaningful_data: |
|
105 | if has_meaningful_data: | |
106 | self._mode = parentfiledata[0] |
|
106 | self._mode = parentfiledata[0] | |
107 | self._size = parentfiledata[1] |
|
107 | self._size = parentfiledata[1] | |
108 | if has_meaningful_mtime: |
|
108 | if has_meaningful_mtime: | |
109 | self._mtime = parentfiledata[2] |
|
109 | self._mtime = parentfiledata[2] | |
110 |
|
110 | |||
111 | @classmethod |
|
111 | @classmethod | |
112 | def from_v1_data(cls, state, mode, size, mtime): |
|
112 | def from_v1_data(cls, state, mode, size, mtime): | |
113 | """Build a new DirstateItem object from V1 data |
|
113 | """Build a new DirstateItem object from V1 data | |
114 |
|
114 | |||
115 | Since the dirstate-v1 format is frozen, the signature of this function |
|
115 | Since the dirstate-v1 format is frozen, the signature of this function | |
116 | is not expected to change, unlike the __init__ one. |
|
116 | is not expected to change, unlike the __init__ one. | |
117 | """ |
|
117 | """ | |
118 | if state == b'm': |
|
118 | if state == b'm': | |
119 | return cls(wc_tracked=True, p1_tracked=True, p2_info=True) |
|
119 | return cls(wc_tracked=True, p1_tracked=True, p2_info=True) | |
120 | elif state == b'a': |
|
120 | elif state == b'a': | |
121 | return cls(wc_tracked=True) |
|
121 | return cls(wc_tracked=True) | |
122 | elif state == b'r': |
|
122 | elif state == b'r': | |
123 | if size == NONNORMAL: |
|
123 | if size == NONNORMAL: | |
124 | p1_tracked = True |
|
124 | p1_tracked = True | |
125 | p2_info = True |
|
125 | p2_info = True | |
126 | elif size == FROM_P2: |
|
126 | elif size == FROM_P2: | |
127 | p1_tracked = False |
|
127 | p1_tracked = False | |
128 | p2_info = True |
|
128 | p2_info = True | |
129 | else: |
|
129 | else: | |
130 | p1_tracked = True |
|
130 | p1_tracked = True | |
131 | p2_info = False |
|
131 | p2_info = False | |
132 | return cls(p1_tracked=p1_tracked, p2_info=p2_info) |
|
132 | return cls(p1_tracked=p1_tracked, p2_info=p2_info) | |
133 | elif state == b'n': |
|
133 | elif state == b'n': | |
134 | if size == FROM_P2: |
|
134 | if size == FROM_P2: | |
135 | return cls(wc_tracked=True, p2_info=True) |
|
135 | return cls(wc_tracked=True, p2_info=True) | |
136 | elif size == NONNORMAL: |
|
136 | elif size == NONNORMAL: | |
137 | return cls(wc_tracked=True, p1_tracked=True) |
|
137 | return cls(wc_tracked=True, p1_tracked=True) | |
138 | elif mtime == AMBIGUOUS_TIME: |
|
138 | elif mtime == AMBIGUOUS_TIME: | |
139 | return cls( |
|
139 | return cls( | |
140 | wc_tracked=True, |
|
140 | wc_tracked=True, | |
141 | p1_tracked=True, |
|
141 | p1_tracked=True, | |
142 | has_meaningful_mtime=False, |
|
142 | has_meaningful_mtime=False, | |
143 | parentfiledata=(mode, size, 42), |
|
143 | parentfiledata=(mode, size, 42), | |
144 | ) |
|
144 | ) | |
145 | else: |
|
145 | else: | |
146 | return cls( |
|
146 | return cls( | |
147 | wc_tracked=True, |
|
147 | wc_tracked=True, | |
148 | p1_tracked=True, |
|
148 | p1_tracked=True, | |
149 | parentfiledata=(mode, size, mtime), |
|
149 | parentfiledata=(mode, size, mtime), | |
150 | ) |
|
150 | ) | |
151 | else: |
|
151 | else: | |
152 | raise RuntimeError(b'unknown state: %s' % state) |
|
152 | raise RuntimeError(b'unknown state: %s' % state) | |
153 |
|
153 | |||
154 | def set_possibly_dirty(self): |
|
154 | def set_possibly_dirty(self): | |
155 | """Mark a file as "possibly dirty" |
|
155 | """Mark a file as "possibly dirty" | |
156 |
|
156 | |||
157 | This means the next status call will have to actually check its content |
|
157 | This means the next status call will have to actually check its content | |
158 | to make sure it is correct. |
|
158 | to make sure it is correct. | |
159 | """ |
|
159 | """ | |
160 | self._mtime = None |
|
160 | self._mtime = None | |
161 |
|
161 | |||
162 | def set_clean(self, mode, size, mtime): |
|
162 | def set_clean(self, mode, size, mtime): | |
163 | """mark a file as "clean" cancelling potential "possibly dirty call" |
|
163 | """mark a file as "clean" cancelling potential "possibly dirty call" | |
164 |
|
164 | |||
165 | Note: this function is a descendant of `dirstate.normal` and is |
|
165 | Note: this function is a descendant of `dirstate.normal` and is | |
166 | currently expected to be call on "normal" entry only. There are not |
|
166 | currently expected to be call on "normal" entry only. There are not | |
167 | reason for this to not change in the future as long as the ccode is |
|
167 | reason for this to not change in the future as long as the ccode is | |
168 | updated to preserve the proper state of the non-normal files. |
|
168 | updated to preserve the proper state of the non-normal files. | |
169 | """ |
|
169 | """ | |
170 | self._wc_tracked = True |
|
170 | self._wc_tracked = True | |
171 | self._p1_tracked = True |
|
171 | self._p1_tracked = True | |
172 | self._mode = mode |
|
172 | self._mode = mode | |
173 | self._size = size |
|
173 | self._size = size | |
174 | self._mtime = mtime |
|
174 | self._mtime = mtime | |
175 |
|
175 | |||
176 | def set_tracked(self): |
|
176 | def set_tracked(self): | |
177 | """mark a file as tracked in the working copy |
|
177 | """mark a file as tracked in the working copy | |
178 |
|
178 | |||
179 | This will ultimately be called by command like `hg add`. |
|
179 | This will ultimately be called by command like `hg add`. | |
180 | """ |
|
180 | """ | |
181 | self._wc_tracked = True |
|
181 | self._wc_tracked = True | |
182 | # `set_tracked` is replacing various `normallookup` call. So we mark |
|
182 | # `set_tracked` is replacing various `normallookup` call. So we mark | |
183 | # the files as needing lookup |
|
183 | # the files as needing lookup | |
184 | # |
|
184 | # | |
185 | # Consider dropping this in the future in favor of something less broad. |
|
185 | # Consider dropping this in the future in favor of something less broad. | |
186 | self._mtime = None |
|
186 | self._mtime = None | |
187 |
|
187 | |||
188 | def set_untracked(self): |
|
188 | def set_untracked(self): | |
189 | """mark a file as untracked in the working copy |
|
189 | """mark a file as untracked in the working copy | |
190 |
|
190 | |||
191 | This will ultimately be called by command like `hg remove`. |
|
191 | This will ultimately be called by command like `hg remove`. | |
192 | """ |
|
192 | """ | |
193 | self._wc_tracked = False |
|
193 | self._wc_tracked = False | |
194 | self._mode = None |
|
194 | self._mode = None | |
195 | self._size = None |
|
195 | self._size = None | |
196 | self._mtime = None |
|
196 | self._mtime = None | |
197 |
|
197 | |||
198 | def drop_merge_data(self): |
|
198 | def drop_merge_data(self): | |
199 | """remove all "merge-only" from a DirstateItem |
|
199 | """remove all "merge-only" from a DirstateItem | |
200 |
|
200 | |||
201 | This is to be call by the dirstatemap code when the second parent is dropped |
|
201 | This is to be call by the dirstatemap code when the second parent is dropped | |
202 | """ |
|
202 | """ | |
203 | if self._p2_info: |
|
203 | if self._p2_info: | |
204 | self._p2_info = False |
|
204 | self._p2_info = False | |
205 | self._mode = None |
|
205 | self._mode = None | |
206 | self._size = None |
|
206 | self._size = None | |
207 | self._mtime = None |
|
207 | self._mtime = None | |
208 |
|
208 | |||
209 | @property |
|
209 | @property | |
210 | def mode(self): |
|
210 | def mode(self): | |
211 | return self.v1_mode() |
|
211 | return self.v1_mode() | |
212 |
|
212 | |||
213 | @property |
|
213 | @property | |
214 | def size(self): |
|
214 | def size(self): | |
215 | return self.v1_size() |
|
215 | return self.v1_size() | |
216 |
|
216 | |||
217 | @property |
|
217 | @property | |
218 | def mtime(self): |
|
218 | def mtime(self): | |
219 | return self.v1_mtime() |
|
219 | return self.v1_mtime() | |
220 |
|
220 | |||
221 | @property |
|
221 | @property | |
222 | def state(self): |
|
222 | def state(self): | |
223 | """ |
|
223 | """ | |
224 | States are: |
|
224 | States are: | |
225 | n normal |
|
225 | n normal | |
226 | m needs merging |
|
226 | m needs merging | |
227 | r marked for removal |
|
227 | r marked for removal | |
228 | a marked for addition |
|
228 | a marked for addition | |
229 |
|
229 | |||
230 | XXX This "state" is a bit obscure and mostly a direct expression of the |
|
230 | XXX This "state" is a bit obscure and mostly a direct expression of the | |
231 | dirstatev1 format. It would make sense to ultimately deprecate it in |
|
231 | dirstatev1 format. It would make sense to ultimately deprecate it in | |
232 | favor of the more "semantic" attributes. |
|
232 | favor of the more "semantic" attributes. | |
233 | """ |
|
233 | """ | |
234 | if not self.any_tracked: |
|
234 | if not self.any_tracked: | |
235 | return b'?' |
|
235 | return b'?' | |
236 | return self.v1_state() |
|
236 | return self.v1_state() | |
237 |
|
237 | |||
238 | @property |
|
238 | @property | |
239 | def tracked(self): |
|
239 | def tracked(self): | |
240 | """True is the file is tracked in the working copy""" |
|
240 | """True is the file is tracked in the working copy""" | |
241 | return self._wc_tracked |
|
241 | return self._wc_tracked | |
242 |
|
242 | |||
243 | @property |
|
243 | @property | |
244 | def any_tracked(self): |
|
244 | def any_tracked(self): | |
245 | """True is the file is tracked anywhere (wc or parents)""" |
|
245 | """True is the file is tracked anywhere (wc or parents)""" | |
246 | return self._wc_tracked or self._p1_tracked or self._p2_info |
|
246 | return self._wc_tracked or self._p1_tracked or self._p2_info | |
247 |
|
247 | |||
248 | @property |
|
248 | @property | |
249 | def added(self): |
|
249 | def added(self): | |
250 | """True if the file has been added""" |
|
250 | """True if the file has been added""" | |
251 | return self._wc_tracked and not (self._p1_tracked or self._p2_info) |
|
251 | return self._wc_tracked and not (self._p1_tracked or self._p2_info) | |
252 |
|
252 | |||
253 | @property |
|
253 | @property | |
254 | def maybe_clean(self): |
|
254 | def maybe_clean(self): | |
255 | """True if the file has a chance to be in the "clean" state""" |
|
255 | """True if the file has a chance to be in the "clean" state""" | |
256 | if not self._wc_tracked: |
|
256 | if not self._wc_tracked: | |
257 | return False |
|
257 | return False | |
258 | elif not self._p1_tracked: |
|
258 | elif not self._p1_tracked: | |
259 | return False |
|
259 | return False | |
260 | elif self._p2_info: |
|
260 | elif self._p2_info: | |
261 | return False |
|
261 | return False | |
262 | return True |
|
262 | return True | |
263 |
|
263 | |||
264 | @property |
|
264 | @property | |
265 | def p1_tracked(self): |
|
265 | def p1_tracked(self): | |
266 | """True if the file is tracked in the first parent manifest""" |
|
266 | """True if the file is tracked in the first parent manifest""" | |
267 | return self._p1_tracked |
|
267 | return self._p1_tracked | |
268 |
|
268 | |||
269 | @property |
|
269 | @property | |
270 | def p2_info(self): |
|
270 | def p2_info(self): | |
271 | """True if the file needed to merge or apply any input from p2 |
|
271 | """True if the file needed to merge or apply any input from p2 | |
272 |
|
272 | |||
273 | See the class documentation for details. |
|
273 | See the class documentation for details. | |
274 | """ |
|
274 | """ | |
275 | return self._wc_tracked and self._p2_info |
|
275 | return self._wc_tracked and self._p2_info | |
276 |
|
276 | |||
277 | @property |
|
277 | @property | |
278 | def removed(self): |
|
278 | def removed(self): | |
279 | """True if the file has been removed""" |
|
279 | """True if the file has been removed""" | |
280 | return not self._wc_tracked and (self._p1_tracked or self._p2_info) |
|
280 | return not self._wc_tracked and (self._p1_tracked or self._p2_info) | |
281 |
|
281 | |||
282 | def v1_state(self): |
|
282 | def v1_state(self): | |
283 | """return a "state" suitable for v1 serialization""" |
|
283 | """return a "state" suitable for v1 serialization""" | |
284 | if not self.any_tracked: |
|
284 | if not self.any_tracked: | |
285 | # the object has no state to record, this is -currently- |
|
285 | # the object has no state to record, this is -currently- | |
286 | # unsupported |
|
286 | # unsupported | |
287 | raise RuntimeError('untracked item') |
|
287 | raise RuntimeError('untracked item') | |
288 | elif self.removed: |
|
288 | elif self.removed: | |
289 | return b'r' |
|
289 | return b'r' | |
290 | elif self._p1_tracked and self._p2_info: |
|
290 | elif self._p1_tracked and self._p2_info: | |
291 | return b'm' |
|
291 | return b'm' | |
292 | elif self.added: |
|
292 | elif self.added: | |
293 | return b'a' |
|
293 | return b'a' | |
294 | else: |
|
294 | else: | |
295 | return b'n' |
|
295 | return b'n' | |
296 |
|
296 | |||
297 | def v1_mode(self): |
|
297 | def v1_mode(self): | |
298 | """return a "mode" suitable for v1 serialization""" |
|
298 | """return a "mode" suitable for v1 serialization""" | |
299 | return self._mode if self._mode is not None else 0 |
|
299 | return self._mode if self._mode is not None else 0 | |
300 |
|
300 | |||
301 | def v1_size(self): |
|
301 | def v1_size(self): | |
302 | """return a "size" suitable for v1 serialization""" |
|
302 | """return a "size" suitable for v1 serialization""" | |
303 | if not self.any_tracked: |
|
303 | if not self.any_tracked: | |
304 | # the object has no state to record, this is -currently- |
|
304 | # the object has no state to record, this is -currently- | |
305 | # unsupported |
|
305 | # unsupported | |
306 | raise RuntimeError('untracked item') |
|
306 | raise RuntimeError('untracked item') | |
307 | elif self.removed and self._p1_tracked and self._p2_info: |
|
307 | elif self.removed and self._p1_tracked and self._p2_info: | |
308 | return NONNORMAL |
|
308 | return NONNORMAL | |
309 | elif self._p2_info: |
|
309 | elif self._p2_info: | |
310 | return FROM_P2 |
|
310 | return FROM_P2 | |
311 | elif self.removed: |
|
311 | elif self.removed: | |
312 | return 0 |
|
312 | return 0 | |
313 | elif self.added: |
|
313 | elif self.added: | |
314 | return NONNORMAL |
|
314 | return NONNORMAL | |
315 | elif self._size is None: |
|
315 | elif self._size is None: | |
316 | return NONNORMAL |
|
316 | return NONNORMAL | |
317 | else: |
|
317 | else: | |
318 | return self._size |
|
318 | return self._size | |
319 |
|
319 | |||
320 | def v1_mtime(self): |
|
320 | def v1_mtime(self): | |
321 | """return a "mtime" suitable for v1 serialization""" |
|
321 | """return a "mtime" suitable for v1 serialization""" | |
322 | if not self.any_tracked: |
|
322 | if not self.any_tracked: | |
323 | # the object has no state to record, this is -currently- |
|
323 | # the object has no state to record, this is -currently- | |
324 | # unsupported |
|
324 | # unsupported | |
325 | raise RuntimeError('untracked item') |
|
325 | raise RuntimeError('untracked item') | |
326 | elif self.removed: |
|
326 | elif self.removed: | |
327 | return 0 |
|
327 | return 0 | |
328 | elif self._mtime is None: |
|
328 | elif self._mtime is None: | |
329 | return AMBIGUOUS_TIME |
|
329 | return AMBIGUOUS_TIME | |
330 | elif self._p2_info: |
|
330 | elif self._p2_info: | |
331 | return AMBIGUOUS_TIME |
|
331 | return AMBIGUOUS_TIME | |
332 | elif not self._p1_tracked: |
|
332 | elif not self._p1_tracked: | |
333 | return AMBIGUOUS_TIME |
|
333 | return AMBIGUOUS_TIME | |
334 | else: |
|
334 | else: | |
335 | return self._mtime |
|
335 | return self._mtime | |
336 |
|
336 | |||
337 | def need_delay(self, now): |
|
337 | def need_delay(self, now): | |
338 | """True if the stored mtime would be ambiguous with the current time""" |
|
338 | """True if the stored mtime would be ambiguous with the current time""" | |
339 | return self.v1_state() == b'n' and self.v1_mtime() == now |
|
339 | return self.v1_state() == b'n' and self.v1_mtime() == now | |
340 |
|
340 | |||
341 |
|
341 | |||
342 | def gettype(q): |
|
342 | def gettype(q): | |
343 | return int(q & 0xFFFF) |
|
343 | return int(q & 0xFFFF) | |
344 |
|
344 | |||
345 |
|
345 | |||
346 | class BaseIndexObject(object): |
|
346 | class BaseIndexObject(object): | |
347 | # Can I be passed to an algorithme implemented in Rust ? |
|
347 | # Can I be passed to an algorithme implemented in Rust ? | |
348 | rust_ext_compat = 0 |
|
348 | rust_ext_compat = 0 | |
349 | # Format of an index entry according to Python's `struct` language |
|
349 | # Format of an index entry according to Python's `struct` language | |
350 | index_format = revlog_constants.INDEX_ENTRY_V1 |
|
350 | index_format = revlog_constants.INDEX_ENTRY_V1 | |
351 | # Size of a C unsigned long long int, platform independent |
|
351 | # Size of a C unsigned long long int, platform independent | |
352 | big_int_size = struct.calcsize(b'>Q') |
|
352 | big_int_size = struct.calcsize(b'>Q') | |
353 | # Size of a C long int, platform independent |
|
353 | # Size of a C long int, platform independent | |
354 | int_size = struct.calcsize(b'>i') |
|
354 | int_size = struct.calcsize(b'>i') | |
355 | # An empty index entry, used as a default value to be overridden, or nullrev |
|
355 | # An empty index entry, used as a default value to be overridden, or nullrev | |
356 | null_item = ( |
|
356 | null_item = ( | |
357 | 0, |
|
357 | 0, | |
358 | 0, |
|
358 | 0, | |
359 | 0, |
|
359 | 0, | |
360 | -1, |
|
360 | -1, | |
361 | -1, |
|
361 | -1, | |
362 | -1, |
|
362 | -1, | |
363 | -1, |
|
363 | -1, | |
364 | sha1nodeconstants.nullid, |
|
364 | sha1nodeconstants.nullid, | |
365 | 0, |
|
365 | 0, | |
366 | 0, |
|
366 | 0, | |
367 | revlog_constants.COMP_MODE_INLINE, |
|
367 | revlog_constants.COMP_MODE_INLINE, | |
368 | revlog_constants.COMP_MODE_INLINE, |
|
368 | revlog_constants.COMP_MODE_INLINE, | |
369 | ) |
|
369 | ) | |
370 |
|
370 | |||
371 | @util.propertycache |
|
371 | @util.propertycache | |
372 | def entry_size(self): |
|
372 | def entry_size(self): | |
373 | return self.index_format.size |
|
373 | return self.index_format.size | |
374 |
|
374 | |||
375 | @property |
|
375 | @property | |
376 | def nodemap(self): |
|
376 | def nodemap(self): | |
377 | msg = b"index.nodemap is deprecated, use index.[has_node|rev|get_rev]" |
|
377 | msg = b"index.nodemap is deprecated, use index.[has_node|rev|get_rev]" | |
378 | util.nouideprecwarn(msg, b'5.3', stacklevel=2) |
|
378 | util.nouideprecwarn(msg, b'5.3', stacklevel=2) | |
379 | return self._nodemap |
|
379 | return self._nodemap | |
380 |
|
380 | |||
381 | @util.propertycache |
|
381 | @util.propertycache | |
382 | def _nodemap(self): |
|
382 | def _nodemap(self): | |
383 | nodemap = nodemaputil.NodeMap({sha1nodeconstants.nullid: nullrev}) |
|
383 | nodemap = nodemaputil.NodeMap({sha1nodeconstants.nullid: nullrev}) | |
384 | for r in range(0, len(self)): |
|
384 | for r in range(0, len(self)): | |
385 | n = self[r][7] |
|
385 | n = self[r][7] | |
386 | nodemap[n] = r |
|
386 | nodemap[n] = r | |
387 | return nodemap |
|
387 | return nodemap | |
388 |
|
388 | |||
389 | def has_node(self, node): |
|
389 | def has_node(self, node): | |
390 | """return True if the node exist in the index""" |
|
390 | """return True if the node exist in the index""" | |
391 | return node in self._nodemap |
|
391 | return node in self._nodemap | |
392 |
|
392 | |||
393 | def rev(self, node): |
|
393 | def rev(self, node): | |
394 | """return a revision for a node |
|
394 | """return a revision for a node | |
395 |
|
395 | |||
396 | If the node is unknown, raise a RevlogError""" |
|
396 | If the node is unknown, raise a RevlogError""" | |
397 | return self._nodemap[node] |
|
397 | return self._nodemap[node] | |
398 |
|
398 | |||
399 | def get_rev(self, node): |
|
399 | def get_rev(self, node): | |
400 | """return a revision for a node |
|
400 | """return a revision for a node | |
401 |
|
401 | |||
402 | If the node is unknown, return None""" |
|
402 | If the node is unknown, return None""" | |
403 | return self._nodemap.get(node) |
|
403 | return self._nodemap.get(node) | |
404 |
|
404 | |||
405 | def _stripnodes(self, start): |
|
405 | def _stripnodes(self, start): | |
406 | if '_nodemap' in vars(self): |
|
406 | if '_nodemap' in vars(self): | |
407 | for r in range(start, len(self)): |
|
407 | for r in range(start, len(self)): | |
408 | n = self[r][7] |
|
408 | n = self[r][7] | |
409 | del self._nodemap[n] |
|
409 | del self._nodemap[n] | |
410 |
|
410 | |||
411 | def clearcaches(self): |
|
411 | def clearcaches(self): | |
412 | self.__dict__.pop('_nodemap', None) |
|
412 | self.__dict__.pop('_nodemap', None) | |
413 |
|
413 | |||
414 | def __len__(self): |
|
414 | def __len__(self): | |
415 | return self._lgt + len(self._extra) |
|
415 | return self._lgt + len(self._extra) | |
416 |
|
416 | |||
417 | def append(self, tup): |
|
417 | def append(self, tup): | |
418 | if '_nodemap' in vars(self): |
|
418 | if '_nodemap' in vars(self): | |
419 | self._nodemap[tup[7]] = len(self) |
|
419 | self._nodemap[tup[7]] = len(self) | |
420 | data = self._pack_entry(len(self), tup) |
|
420 | data = self._pack_entry(len(self), tup) | |
421 | self._extra.append(data) |
|
421 | self._extra.append(data) | |
422 |
|
422 | |||
423 | def _pack_entry(self, rev, entry): |
|
423 | def _pack_entry(self, rev, entry): | |
424 | assert entry[8] == 0 |
|
424 | assert entry[8] == 0 | |
425 | assert entry[9] == 0 |
|
425 | assert entry[9] == 0 | |
426 | return self.index_format.pack(*entry[:8]) |
|
426 | return self.index_format.pack(*entry[:8]) | |
427 |
|
427 | |||
428 | def _check_index(self, i): |
|
428 | def _check_index(self, i): | |
429 | if not isinstance(i, int): |
|
429 | if not isinstance(i, int): | |
430 | raise TypeError(b"expecting int indexes") |
|
430 | raise TypeError(b"expecting int indexes") | |
431 | if i < 0 or i >= len(self): |
|
431 | if i < 0 or i >= len(self): | |
432 | raise IndexError |
|
432 | raise IndexError | |
433 |
|
433 | |||
434 | def __getitem__(self, i): |
|
434 | def __getitem__(self, i): | |
435 | if i == -1: |
|
435 | if i == -1: | |
436 | return self.null_item |
|
436 | return self.null_item | |
437 | self._check_index(i) |
|
437 | self._check_index(i) | |
438 | if i >= self._lgt: |
|
438 | if i >= self._lgt: | |
439 | data = self._extra[i - self._lgt] |
|
439 | data = self._extra[i - self._lgt] | |
440 | else: |
|
440 | else: | |
441 | index = self._calculate_index(i) |
|
441 | index = self._calculate_index(i) | |
442 | data = self._data[index : index + self.entry_size] |
|
442 | data = self._data[index : index + self.entry_size] | |
443 | r = self._unpack_entry(i, data) |
|
443 | r = self._unpack_entry(i, data) | |
444 | if self._lgt and i == 0: |
|
444 | if self._lgt and i == 0: | |
445 | offset = revlogutils.offset_type(0, gettype(r[0])) |
|
445 | offset = revlogutils.offset_type(0, gettype(r[0])) | |
446 | r = (offset,) + r[1:] |
|
446 | r = (offset,) + r[1:] | |
447 | return r |
|
447 | return r | |
448 |
|
448 | |||
449 | def _unpack_entry(self, rev, data): |
|
449 | def _unpack_entry(self, rev, data): | |
450 | r = self.index_format.unpack(data) |
|
450 | r = self.index_format.unpack(data) | |
451 | r = r + ( |
|
451 | r = r + ( | |
452 | 0, |
|
452 | 0, | |
453 | 0, |
|
453 | 0, | |
454 | revlog_constants.COMP_MODE_INLINE, |
|
454 | revlog_constants.COMP_MODE_INLINE, | |
455 | revlog_constants.COMP_MODE_INLINE, |
|
455 | revlog_constants.COMP_MODE_INLINE, | |
456 | ) |
|
456 | ) | |
457 | return r |
|
457 | return r | |
458 |
|
458 | |||
459 | def pack_header(self, header): |
|
459 | def pack_header(self, header): | |
460 | """pack header information as binary""" |
|
460 | """pack header information as binary""" | |
461 | v_fmt = revlog_constants.INDEX_HEADER |
|
461 | v_fmt = revlog_constants.INDEX_HEADER | |
462 | return v_fmt.pack(header) |
|
462 | return v_fmt.pack(header) | |
463 |
|
463 | |||
464 | def entry_binary(self, rev): |
|
464 | def entry_binary(self, rev): | |
465 | """return the raw binary string representing a revision""" |
|
465 | """return the raw binary string representing a revision""" | |
466 | entry = self[rev] |
|
466 | entry = self[rev] | |
467 | p = revlog_constants.INDEX_ENTRY_V1.pack(*entry[:8]) |
|
467 | p = revlog_constants.INDEX_ENTRY_V1.pack(*entry[:8]) | |
468 | if rev == 0: |
|
468 | if rev == 0: | |
469 | p = p[revlog_constants.INDEX_HEADER.size :] |
|
469 | p = p[revlog_constants.INDEX_HEADER.size :] | |
470 | return p |
|
470 | return p | |
471 |
|
471 | |||
472 |
|
472 | |||
473 | class IndexObject(BaseIndexObject): |
|
473 | class IndexObject(BaseIndexObject): | |
474 | def __init__(self, data): |
|
474 | def __init__(self, data): | |
475 | assert len(data) % self.entry_size == 0, ( |
|
475 | assert len(data) % self.entry_size == 0, ( | |
476 | len(data), |
|
476 | len(data), | |
477 | self.entry_size, |
|
477 | self.entry_size, | |
478 | len(data) % self.entry_size, |
|
478 | len(data) % self.entry_size, | |
479 | ) |
|
479 | ) | |
480 | self._data = data |
|
480 | self._data = data | |
481 | self._lgt = len(data) // self.entry_size |
|
481 | self._lgt = len(data) // self.entry_size | |
482 | self._extra = [] |
|
482 | self._extra = [] | |
483 |
|
483 | |||
484 | def _calculate_index(self, i): |
|
484 | def _calculate_index(self, i): | |
485 | return i * self.entry_size |
|
485 | return i * self.entry_size | |
486 |
|
486 | |||
487 | def __delitem__(self, i): |
|
487 | def __delitem__(self, i): | |
488 | if not isinstance(i, slice) or not i.stop == -1 or i.step is not None: |
|
488 | if not isinstance(i, slice) or not i.stop == -1 or i.step is not None: | |
489 | raise ValueError(b"deleting slices only supports a:-1 with step 1") |
|
489 | raise ValueError(b"deleting slices only supports a:-1 with step 1") | |
490 | i = i.start |
|
490 | i = i.start | |
491 | self._check_index(i) |
|
491 | self._check_index(i) | |
492 | self._stripnodes(i) |
|
492 | self._stripnodes(i) | |
493 | if i < self._lgt: |
|
493 | if i < self._lgt: | |
494 | self._data = self._data[: i * self.entry_size] |
|
494 | self._data = self._data[: i * self.entry_size] | |
495 | self._lgt = i |
|
495 | self._lgt = i | |
496 | self._extra = [] |
|
496 | self._extra = [] | |
497 | else: |
|
497 | else: | |
498 | self._extra = self._extra[: i - self._lgt] |
|
498 | self._extra = self._extra[: i - self._lgt] | |
499 |
|
499 | |||
500 |
|
500 | |||
501 | class PersistentNodeMapIndexObject(IndexObject): |
|
501 | class PersistentNodeMapIndexObject(IndexObject): | |
502 | """a Debug oriented class to test persistent nodemap |
|
502 | """a Debug oriented class to test persistent nodemap | |
503 |
|
503 | |||
504 | We need a simple python object to test API and higher level behavior. See |
|
504 | We need a simple python object to test API and higher level behavior. See | |
505 | the Rust implementation for more serious usage. This should be used only |
|
505 | the Rust implementation for more serious usage. This should be used only | |
506 | through the dedicated `devel.persistent-nodemap` config. |
|
506 | through the dedicated `devel.persistent-nodemap` config. | |
507 | """ |
|
507 | """ | |
508 |
|
508 | |||
509 | def nodemap_data_all(self): |
|
509 | def nodemap_data_all(self): | |
510 | """Return bytes containing a full serialization of a nodemap |
|
510 | """Return bytes containing a full serialization of a nodemap | |
511 |
|
511 | |||
512 | The nodemap should be valid for the full set of revisions in the |
|
512 | The nodemap should be valid for the full set of revisions in the | |
513 | index.""" |
|
513 | index.""" | |
514 | return nodemaputil.persistent_data(self) |
|
514 | return nodemaputil.persistent_data(self) | |
515 |
|
515 | |||
516 | def nodemap_data_incremental(self): |
|
516 | def nodemap_data_incremental(self): | |
517 | """Return bytes containing a incremental update to persistent nodemap |
|
517 | """Return bytes containing a incremental update to persistent nodemap | |
518 |
|
518 | |||
519 | This containst the data for an append-only update of the data provided |
|
519 | This containst the data for an append-only update of the data provided | |
520 | in the last call to `update_nodemap_data`. |
|
520 | in the last call to `update_nodemap_data`. | |
521 | """ |
|
521 | """ | |
522 | if self._nm_root is None: |
|
522 | if self._nm_root is None: | |
523 | return None |
|
523 | return None | |
524 | docket = self._nm_docket |
|
524 | docket = self._nm_docket | |
525 | changed, data = nodemaputil.update_persistent_data( |
|
525 | changed, data = nodemaputil.update_persistent_data( | |
526 | self, self._nm_root, self._nm_max_idx, self._nm_docket.tip_rev |
|
526 | self, self._nm_root, self._nm_max_idx, self._nm_docket.tip_rev | |
527 | ) |
|
527 | ) | |
528 |
|
528 | |||
529 | self._nm_root = self._nm_max_idx = self._nm_docket = None |
|
529 | self._nm_root = self._nm_max_idx = self._nm_docket = None | |
530 | return docket, changed, data |
|
530 | return docket, changed, data | |
531 |
|
531 | |||
532 | def update_nodemap_data(self, docket, nm_data): |
|
532 | def update_nodemap_data(self, docket, nm_data): | |
533 | """provide full block of persisted binary data for a nodemap |
|
533 | """provide full block of persisted binary data for a nodemap | |
534 |
|
534 | |||
535 | The data are expected to come from disk. See `nodemap_data_all` for a |
|
535 | The data are expected to come from disk. See `nodemap_data_all` for a | |
536 | produceur of such data.""" |
|
536 | produceur of such data.""" | |
537 | if nm_data is not None: |
|
537 | if nm_data is not None: | |
538 | self._nm_root, self._nm_max_idx = nodemaputil.parse_data(nm_data) |
|
538 | self._nm_root, self._nm_max_idx = nodemaputil.parse_data(nm_data) | |
539 | if self._nm_root: |
|
539 | if self._nm_root: | |
540 | self._nm_docket = docket |
|
540 | self._nm_docket = docket | |
541 | else: |
|
541 | else: | |
542 | self._nm_root = self._nm_max_idx = self._nm_docket = None |
|
542 | self._nm_root = self._nm_max_idx = self._nm_docket = None | |
543 |
|
543 | |||
544 |
|
544 | |||
545 | class InlinedIndexObject(BaseIndexObject): |
|
545 | class InlinedIndexObject(BaseIndexObject): | |
546 | def __init__(self, data, inline=0): |
|
546 | def __init__(self, data, inline=0): | |
547 | self._data = data |
|
547 | self._data = data | |
548 | self._lgt = self._inline_scan(None) |
|
548 | self._lgt = self._inline_scan(None) | |
549 | self._inline_scan(self._lgt) |
|
549 | self._inline_scan(self._lgt) | |
550 | self._extra = [] |
|
550 | self._extra = [] | |
551 |
|
551 | |||
552 | def _inline_scan(self, lgt): |
|
552 | def _inline_scan(self, lgt): | |
553 | off = 0 |
|
553 | off = 0 | |
554 | if lgt is not None: |
|
554 | if lgt is not None: | |
555 | self._offsets = [0] * lgt |
|
555 | self._offsets = [0] * lgt | |
556 | count = 0 |
|
556 | count = 0 | |
557 | while off <= len(self._data) - self.entry_size: |
|
557 | while off <= len(self._data) - self.entry_size: | |
558 | start = off + self.big_int_size |
|
558 | start = off + self.big_int_size | |
559 | (s,) = struct.unpack( |
|
559 | (s,) = struct.unpack( | |
560 | b'>i', |
|
560 | b'>i', | |
561 | self._data[start : start + self.int_size], |
|
561 | self._data[start : start + self.int_size], | |
562 | ) |
|
562 | ) | |
563 | if lgt is not None: |
|
563 | if lgt is not None: | |
564 | self._offsets[count] = off |
|
564 | self._offsets[count] = off | |
565 | count += 1 |
|
565 | count += 1 | |
566 | off += self.entry_size + s |
|
566 | off += self.entry_size + s | |
567 | if off != len(self._data): |
|
567 | if off != len(self._data): | |
568 | raise ValueError(b"corrupted data") |
|
568 | raise ValueError(b"corrupted data") | |
569 | return count |
|
569 | return count | |
570 |
|
570 | |||
571 | def __delitem__(self, i): |
|
571 | def __delitem__(self, i): | |
572 | if not isinstance(i, slice) or not i.stop == -1 or i.step is not None: |
|
572 | if not isinstance(i, slice) or not i.stop == -1 or i.step is not None: | |
573 | raise ValueError(b"deleting slices only supports a:-1 with step 1") |
|
573 | raise ValueError(b"deleting slices only supports a:-1 with step 1") | |
574 | i = i.start |
|
574 | i = i.start | |
575 | self._check_index(i) |
|
575 | self._check_index(i) | |
576 | self._stripnodes(i) |
|
576 | self._stripnodes(i) | |
577 | if i < self._lgt: |
|
577 | if i < self._lgt: | |
578 | self._offsets = self._offsets[:i] |
|
578 | self._offsets = self._offsets[:i] | |
579 | self._lgt = i |
|
579 | self._lgt = i | |
580 | self._extra = [] |
|
580 | self._extra = [] | |
581 | else: |
|
581 | else: | |
582 | self._extra = self._extra[: i - self._lgt] |
|
582 | self._extra = self._extra[: i - self._lgt] | |
583 |
|
583 | |||
584 | def _calculate_index(self, i): |
|
584 | def _calculate_index(self, i): | |
585 | return self._offsets[i] |
|
585 | return self._offsets[i] | |
586 |
|
586 | |||
587 |
|
587 | |||
588 | def parse_index2(data, inline, revlogv2=False): |
|
588 | def parse_index2(data, inline, revlogv2=False): | |
589 | if not inline: |
|
589 | if not inline: | |
590 | cls = IndexObject2 if revlogv2 else IndexObject |
|
590 | cls = IndexObject2 if revlogv2 else IndexObject | |
591 | return cls(data), None |
|
591 | return cls(data), None | |
592 | cls = InlinedIndexObject |
|
592 | cls = InlinedIndexObject | |
593 | return cls(data, inline), (0, data) |
|
593 | return cls(data, inline), (0, data) | |
594 |
|
594 | |||
595 |
|
595 | |||
596 | def parse_index_cl_v2(data): |
|
596 | def parse_index_cl_v2(data): | |
597 | return IndexChangelogV2(data), None |
|
597 | return IndexChangelogV2(data), None | |
598 |
|
598 | |||
599 |
|
599 | |||
600 | class IndexObject2(IndexObject): |
|
600 | class IndexObject2(IndexObject): | |
601 | index_format = revlog_constants.INDEX_ENTRY_V2 |
|
601 | index_format = revlog_constants.INDEX_ENTRY_V2 | |
602 |
|
602 | |||
603 | def replace_sidedata_info( |
|
603 | def replace_sidedata_info( | |
604 | self, |
|
604 | self, | |
605 | rev, |
|
605 | rev, | |
606 | sidedata_offset, |
|
606 | sidedata_offset, | |
607 | sidedata_length, |
|
607 | sidedata_length, | |
608 | offset_flags, |
|
608 | offset_flags, | |
609 | compression_mode, |
|
609 | compression_mode, | |
610 | ): |
|
610 | ): | |
611 | """ |
|
611 | """ | |
612 | Replace an existing index entry's sidedata offset and length with new |
|
612 | Replace an existing index entry's sidedata offset and length with new | |
613 | ones. |
|
613 | ones. | |
614 | This cannot be used outside of the context of sidedata rewriting, |
|
614 | This cannot be used outside of the context of sidedata rewriting, | |
615 | inside the transaction that creates the revision `rev`. |
|
615 | inside the transaction that creates the revision `rev`. | |
616 | """ |
|
616 | """ | |
617 | if rev < 0: |
|
617 | if rev < 0: | |
618 | raise KeyError |
|
618 | raise KeyError | |
619 | self._check_index(rev) |
|
619 | self._check_index(rev) | |
620 | if rev < self._lgt: |
|
620 | if rev < self._lgt: | |
621 | msg = b"cannot rewrite entries outside of this transaction" |
|
621 | msg = b"cannot rewrite entries outside of this transaction" | |
622 | raise KeyError(msg) |
|
622 | raise KeyError(msg) | |
623 | else: |
|
623 | else: | |
624 | entry = list(self[rev]) |
|
624 | entry = list(self[rev]) | |
625 | entry[0] = offset_flags |
|
625 | entry[0] = offset_flags | |
626 | entry[8] = sidedata_offset |
|
626 | entry[8] = sidedata_offset | |
627 | entry[9] = sidedata_length |
|
627 | entry[9] = sidedata_length | |
628 | entry[11] = compression_mode |
|
628 | entry[11] = compression_mode | |
629 | entry = tuple(entry) |
|
629 | entry = tuple(entry) | |
630 | new = self._pack_entry(rev, entry) |
|
630 | new = self._pack_entry(rev, entry) | |
631 | self._extra[rev - self._lgt] = new |
|
631 | self._extra[rev - self._lgt] = new | |
632 |
|
632 | |||
633 | def _unpack_entry(self, rev, data): |
|
633 | def _unpack_entry(self, rev, data): | |
634 | data = self.index_format.unpack(data) |
|
634 | data = self.index_format.unpack(data) | |
635 | entry = data[:10] |
|
635 | entry = data[:10] | |
636 | data_comp = data[10] & 3 |
|
636 | data_comp = data[10] & 3 | |
637 | sidedata_comp = (data[10] & (3 << 2)) >> 2 |
|
637 | sidedata_comp = (data[10] & (3 << 2)) >> 2 | |
638 | return entry + (data_comp, sidedata_comp) |
|
638 | return entry + (data_comp, sidedata_comp) | |
639 |
|
639 | |||
640 | def _pack_entry(self, rev, entry): |
|
640 | def _pack_entry(self, rev, entry): | |
641 | data = entry[:10] |
|
641 | data = entry[:10] | |
642 | data_comp = entry[10] & 3 |
|
642 | data_comp = entry[10] & 3 | |
643 | sidedata_comp = (entry[11] & 3) << 2 |
|
643 | sidedata_comp = (entry[11] & 3) << 2 | |
644 | data += (data_comp | sidedata_comp,) |
|
644 | data += (data_comp | sidedata_comp,) | |
645 |
|
645 | |||
646 | return self.index_format.pack(*data) |
|
646 | return self.index_format.pack(*data) | |
647 |
|
647 | |||
648 | def entry_binary(self, rev): |
|
648 | def entry_binary(self, rev): | |
649 | """return the raw binary string representing a revision""" |
|
649 | """return the raw binary string representing a revision""" | |
650 | entry = self[rev] |
|
650 | entry = self[rev] | |
651 | return self._pack_entry(rev, entry) |
|
651 | return self._pack_entry(rev, entry) | |
652 |
|
652 | |||
653 | def pack_header(self, header): |
|
653 | def pack_header(self, header): | |
654 | """pack header information as binary""" |
|
654 | """pack header information as binary""" | |
655 | msg = 'version header should go in the docket, not the index: %d' |
|
655 | msg = 'version header should go in the docket, not the index: %d' | |
656 | msg %= header |
|
656 | msg %= header | |
657 | raise error.ProgrammingError(msg) |
|
657 | raise error.ProgrammingError(msg) | |
658 |
|
658 | |||
659 |
|
659 | |||
660 | class IndexChangelogV2(IndexObject2): |
|
660 | class IndexChangelogV2(IndexObject2): | |
661 | index_format = revlog_constants.INDEX_ENTRY_CL_V2 |
|
661 | index_format = revlog_constants.INDEX_ENTRY_CL_V2 | |
662 |
|
662 | |||
663 | def _unpack_entry(self, rev, data, r=True): |
|
663 | def _unpack_entry(self, rev, data, r=True): | |
664 | items = self.index_format.unpack(data) |
|
664 | items = self.index_format.unpack(data) | |
665 | entry = items[:3] + (rev, rev) + items[3:8] |
|
665 | entry = items[:3] + (rev, rev) + items[3:8] | |
666 | data_comp = items[8] & 3 |
|
666 | data_comp = items[8] & 3 | |
667 | sidedata_comp = (items[8] >> 2) & 3 |
|
667 | sidedata_comp = (items[8] >> 2) & 3 | |
668 | return entry + (data_comp, sidedata_comp) |
|
668 | return entry + (data_comp, sidedata_comp) | |
669 |
|
669 | |||
670 | def _pack_entry(self, rev, entry): |
|
670 | def _pack_entry(self, rev, entry): | |
671 | assert entry[3] == rev, entry[3] |
|
671 | assert entry[3] == rev, entry[3] | |
672 | assert entry[4] == rev, entry[4] |
|
672 | assert entry[4] == rev, entry[4] | |
673 | data = entry[:3] + entry[5:10] |
|
673 | data = entry[:3] + entry[5:10] | |
674 | data_comp = entry[10] & 3 |
|
674 | data_comp = entry[10] & 3 | |
675 | sidedata_comp = (entry[11] & 3) << 2 |
|
675 | sidedata_comp = (entry[11] & 3) << 2 | |
676 | data += (data_comp | sidedata_comp,) |
|
676 | data += (data_comp | sidedata_comp,) | |
677 | return self.index_format.pack(*data) |
|
677 | return self.index_format.pack(*data) | |
678 |
|
678 | |||
679 |
|
679 | |||
680 | def parse_index_devel_nodemap(data, inline): |
|
680 | def parse_index_devel_nodemap(data, inline): | |
681 | """like parse_index2, but alway return a PersistentNodeMapIndexObject""" |
|
681 | """like parse_index2, but alway return a PersistentNodeMapIndexObject""" | |
682 | return PersistentNodeMapIndexObject(data), None |
|
682 | return PersistentNodeMapIndexObject(data), None | |
683 |
|
683 | |||
684 |
|
684 | |||
685 | def parse_dirstate(dmap, copymap, st): |
|
685 | def parse_dirstate(dmap, copymap, st): | |
686 | parents = [st[:20], st[20:40]] |
|
686 | parents = [st[:20], st[20:40]] | |
687 | # dereference fields so they will be local in loop |
|
687 | # dereference fields so they will be local in loop | |
688 | format = b">cllll" |
|
688 | format = b">cllll" | |
689 | e_size = struct.calcsize(format) |
|
689 | e_size = struct.calcsize(format) | |
690 | pos1 = 40 |
|
690 | pos1 = 40 | |
691 | l = len(st) |
|
691 | l = len(st) | |
692 |
|
692 | |||
693 | # the inner loop |
|
693 | # the inner loop | |
694 | while pos1 < l: |
|
694 | while pos1 < l: | |
695 | pos2 = pos1 + e_size |
|
695 | pos2 = pos1 + e_size | |
696 | e = _unpack(b">cllll", st[pos1:pos2]) # a literal here is faster |
|
696 | e = _unpack(b">cllll", st[pos1:pos2]) # a literal here is faster | |
697 | pos1 = pos2 + e[4] |
|
697 | pos1 = pos2 + e[4] | |
698 | f = st[pos2:pos1] |
|
698 | f = st[pos2:pos1] | |
699 | if b'\0' in f: |
|
699 | if b'\0' in f: | |
700 | f, c = f.split(b'\0') |
|
700 | f, c = f.split(b'\0') | |
701 | copymap[f] = c |
|
701 | copymap[f] = c | |
702 | dmap[f] = DirstateItem.from_v1_data(*e[:4]) |
|
702 | dmap[f] = DirstateItem.from_v1_data(*e[:4]) | |
703 | return parents |
|
703 | return parents | |
704 |
|
704 | |||
705 |
|
705 | |||
706 | def pack_dirstate(dmap, copymap, pl, now): |
|
706 | def pack_dirstate(dmap, copymap, pl, now): | |
707 | now = int(now) |
|
707 | now = int(now) | |
708 | cs = stringio() |
|
708 | cs = stringio() | |
709 | write = cs.write |
|
709 | write = cs.write | |
710 | write(b"".join(pl)) |
|
710 | write(b"".join(pl)) | |
711 | for f, e in pycompat.iteritems(dmap): |
|
711 | for f, e in pycompat.iteritems(dmap): | |
712 | if e.need_delay(now): |
|
712 | if e.need_delay(now): | |
713 | # The file was last modified "simultaneously" with the current |
|
713 | # The file was last modified "simultaneously" with the current | |
714 | # write to dirstate (i.e. within the same second for file- |
|
714 | # write to dirstate (i.e. within the same second for file- | |
715 | # systems with a granularity of 1 sec). This commonly happens |
|
715 | # systems with a granularity of 1 sec). This commonly happens | |
716 | # for at least a couple of files on 'update'. |
|
716 | # for at least a couple of files on 'update'. | |
717 | # The user could change the file without changing its size |
|
717 | # The user could change the file without changing its size | |
718 | # within the same second. Invalidate the file's mtime in |
|
718 | # within the same second. Invalidate the file's mtime in | |
719 | # dirstate, forcing future 'status' calls to compare the |
|
719 | # dirstate, forcing future 'status' calls to compare the | |
720 | # contents of the file if the size is the same. This prevents |
|
720 | # contents of the file if the size is the same. This prevents | |
721 | # mistakenly treating such files as clean. |
|
721 | # mistakenly treating such files as clean. | |
722 | e.set_possibly_dirty() |
|
722 | e.set_possibly_dirty() | |
723 |
|
723 | |||
724 | if f in copymap: |
|
724 | if f in copymap: | |
725 | f = b"%s\0%s" % (f, copymap[f]) |
|
725 | f = b"%s\0%s" % (f, copymap[f]) | |
726 | e = _pack( |
|
726 | e = _pack( | |
727 | b">cllll", |
|
727 | b">cllll", | |
728 | e.v1_state(), |
|
728 | e.v1_state(), | |
729 | e.v1_mode(), |
|
729 | e.v1_mode(), | |
730 | e.v1_size(), |
|
730 | e.v1_size(), | |
731 | e.v1_mtime(), |
|
731 | e.v1_mtime(), | |
732 | len(f), |
|
732 | len(f), | |
733 | ) |
|
733 | ) | |
734 | write(e) |
|
734 | write(e) | |
735 | write(f) |
|
735 | write(f) | |
736 | return cs.getvalue() |
|
736 | return cs.getvalue() |
@@ -1,792 +1,733 b'' | |||||
1 | //! The "version 2" disk representation of the dirstate |
|
1 | //! The "version 2" disk representation of the dirstate | |
2 | //! |
|
2 | //! | |
3 | //! See `mercurial/helptext/internals/dirstate-v2.txt` |
|
3 | //! See `mercurial/helptext/internals/dirstate-v2.txt` | |
4 |
|
4 | |||
5 | use crate::dirstate_tree::dirstate_map::{self, DirstateMap, NodeRef}; |
|
5 | use crate::dirstate_tree::dirstate_map::{self, DirstateMap, NodeRef}; | |
6 | use crate::dirstate_tree::path_with_basename::WithBasename; |
|
6 | use crate::dirstate_tree::path_with_basename::WithBasename; | |
7 | use crate::errors::HgError; |
|
7 | use crate::errors::HgError; | |
8 | use crate::utils::hg_path::HgPath; |
|
8 | use crate::utils::hg_path::HgPath; | |
9 | use crate::DirstateEntry; |
|
9 | use crate::DirstateEntry; | |
10 | use crate::DirstateError; |
|
10 | use crate::DirstateError; | |
11 | use crate::DirstateParents; |
|
11 | use crate::DirstateParents; | |
12 | use bitflags::bitflags; |
|
12 | use bitflags::bitflags; | |
13 | use bytes_cast::unaligned::{I32Be, I64Be, U16Be, U32Be}; |
|
13 | use bytes_cast::unaligned::{I32Be, I64Be, U16Be, U32Be}; | |
14 | use bytes_cast::BytesCast; |
|
14 | use bytes_cast::BytesCast; | |
15 | use format_bytes::format_bytes; |
|
15 | use format_bytes::format_bytes; | |
16 | use std::borrow::Cow; |
|
16 | use std::borrow::Cow; | |
17 | use std::convert::{TryFrom, TryInto}; |
|
17 | use std::convert::{TryFrom, TryInto}; | |
18 | use std::time::{Duration, SystemTime, UNIX_EPOCH}; |
|
18 | use std::time::{Duration, SystemTime, UNIX_EPOCH}; | |
19 |
|
19 | |||
20 | /// Added at the start of `.hg/dirstate` when the "v2" format is used. |
|
20 | /// Added at the start of `.hg/dirstate` when the "v2" format is used. | |
21 | /// This a redundant sanity check more than an actual "magic number" since |
|
21 | /// This a redundant sanity check more than an actual "magic number" since | |
22 | /// `.hg/requires` already governs which format should be used. |
|
22 | /// `.hg/requires` already governs which format should be used. | |
23 | pub const V2_FORMAT_MARKER: &[u8; 12] = b"dirstate-v2\n"; |
|
23 | pub const V2_FORMAT_MARKER: &[u8; 12] = b"dirstate-v2\n"; | |
24 |
|
24 | |||
25 | /// Keep space for 256-bit hashes |
|
25 | /// Keep space for 256-bit hashes | |
26 | const STORED_NODE_ID_BYTES: usize = 32; |
|
26 | const STORED_NODE_ID_BYTES: usize = 32; | |
27 |
|
27 | |||
28 | /// β¦ even though only 160 bits are used for now, with SHA-1 |
|
28 | /// β¦ even though only 160 bits are used for now, with SHA-1 | |
29 | const USED_NODE_ID_BYTES: usize = 20; |
|
29 | const USED_NODE_ID_BYTES: usize = 20; | |
30 |
|
30 | |||
31 | pub(super) const IGNORE_PATTERNS_HASH_LEN: usize = 20; |
|
31 | pub(super) const IGNORE_PATTERNS_HASH_LEN: usize = 20; | |
32 | pub(super) type IgnorePatternsHash = [u8; IGNORE_PATTERNS_HASH_LEN]; |
|
32 | pub(super) type IgnorePatternsHash = [u8; IGNORE_PATTERNS_HASH_LEN]; | |
33 |
|
33 | |||
34 | /// Must match the constant of the same name in |
|
34 | /// Must match the constant of the same name in | |
35 | /// `mercurial/dirstateutils/docket.py` |
|
35 | /// `mercurial/dirstateutils/docket.py` | |
36 | const TREE_METADATA_SIZE: usize = 44; |
|
36 | const TREE_METADATA_SIZE: usize = 44; | |
37 |
|
37 | |||
38 | /// Make sure that size-affecting changes are made knowingly |
|
38 | /// Make sure that size-affecting changes are made knowingly | |
39 | #[allow(unused)] |
|
39 | #[allow(unused)] | |
40 | fn static_assert_size_of() { |
|
40 | fn static_assert_size_of() { | |
41 | let _ = std::mem::transmute::<TreeMetadata, [u8; TREE_METADATA_SIZE]>; |
|
41 | let _ = std::mem::transmute::<TreeMetadata, [u8; TREE_METADATA_SIZE]>; | |
42 | let _ = std::mem::transmute::<DocketHeader, [u8; TREE_METADATA_SIZE + 81]>; |
|
42 | let _ = std::mem::transmute::<DocketHeader, [u8; TREE_METADATA_SIZE + 81]>; | |
43 | let _ = std::mem::transmute::<Node, [u8; 43]>; |
|
43 | let _ = std::mem::transmute::<Node, [u8; 43]>; | |
44 | } |
|
44 | } | |
45 |
|
45 | |||
46 | // Must match `HEADER` in `mercurial/dirstateutils/docket.py` |
|
46 | // Must match `HEADER` in `mercurial/dirstateutils/docket.py` | |
47 | #[derive(BytesCast)] |
|
47 | #[derive(BytesCast)] | |
48 | #[repr(C)] |
|
48 | #[repr(C)] | |
49 | struct DocketHeader { |
|
49 | struct DocketHeader { | |
50 | marker: [u8; V2_FORMAT_MARKER.len()], |
|
50 | marker: [u8; V2_FORMAT_MARKER.len()], | |
51 | parent_1: [u8; STORED_NODE_ID_BYTES], |
|
51 | parent_1: [u8; STORED_NODE_ID_BYTES], | |
52 | parent_2: [u8; STORED_NODE_ID_BYTES], |
|
52 | parent_2: [u8; STORED_NODE_ID_BYTES], | |
53 |
|
53 | |||
54 | metadata: TreeMetadata, |
|
54 | metadata: TreeMetadata, | |
55 |
|
55 | |||
56 | /// Counted in bytes |
|
56 | /// Counted in bytes | |
57 | data_size: Size, |
|
57 | data_size: Size, | |
58 |
|
58 | |||
59 | uuid_size: u8, |
|
59 | uuid_size: u8, | |
60 | } |
|
60 | } | |
61 |
|
61 | |||
62 | pub struct Docket<'on_disk> { |
|
62 | pub struct Docket<'on_disk> { | |
63 | header: &'on_disk DocketHeader, |
|
63 | header: &'on_disk DocketHeader, | |
64 | uuid: &'on_disk [u8], |
|
64 | uuid: &'on_disk [u8], | |
65 | } |
|
65 | } | |
66 |
|
66 | |||
|
67 | /// Fields are documented in the *Tree metadata in the docket file* | |||
|
68 | /// section of `mercurial/helptext/internals/dirstate-v2.txt` | |||
67 | #[derive(BytesCast)] |
|
69 | #[derive(BytesCast)] | |
68 | #[repr(C)] |
|
70 | #[repr(C)] | |
69 | struct TreeMetadata { |
|
71 | struct TreeMetadata { | |
70 | root_nodes: ChildNodes, |
|
72 | root_nodes: ChildNodes, | |
71 | nodes_with_entry_count: Size, |
|
73 | nodes_with_entry_count: Size, | |
72 | nodes_with_copy_source_count: Size, |
|
74 | nodes_with_copy_source_count: Size, | |
73 |
|
||||
74 | /// How many bytes of this data file are not used anymore |
|
|||
75 | unreachable_bytes: Size, |
|
75 | unreachable_bytes: Size, | |
76 |
|
||||
77 | /// Current version always sets these bytes to zero when creating or |
|
|||
78 | /// updating a dirstate. Future versions could assign some bits to signal |
|
|||
79 | /// for example "the version that last wrote/updated this dirstate did so |
|
|||
80 | /// in such and such way that can be relied on by versions that know to." |
|
|||
81 | unused: [u8; 4], |
|
76 | unused: [u8; 4], | |
82 |
|
77 | |||
83 | /// If non-zero, a hash of ignore files that were used for some previous |
|
78 | /// See *Optional hash of ignore patterns* section of | |
84 | /// run of the `status` algorithm. |
|
79 | /// `mercurial/helptext/internals/dirstate-v2.txt` | |
85 | /// |
|
|||
86 | /// We define: |
|
|||
87 | /// |
|
|||
88 | /// * "Root" ignore files are `.hgignore` at the root of the repository if |
|
|||
89 | /// it exists, and files from `ui.ignore.*` config. This set of files is |
|
|||
90 | /// then sorted by the string representation of their path. |
|
|||
91 | /// * The "expanded contents" of an ignore files is the byte string made |
|
|||
92 | /// by concatenating its contents with the "expanded contents" of other |
|
|||
93 | /// files included with `include:` or `subinclude:` files, in inclusion |
|
|||
94 | /// order. This definition is recursive, as included files can |
|
|||
95 | /// themselves include more files. |
|
|||
96 | /// |
|
|||
97 | /// This hash is defined as the SHA-1 of the concatenation (in sorted |
|
|||
98 | /// order) of the "expanded contents" of each "root" ignore file. |
|
|||
99 | /// (Note that computing this does not require actually concatenating byte |
|
|||
100 | /// strings into contiguous memory, instead SHA-1 hashing can be done |
|
|||
101 | /// incrementally.) |
|
|||
102 | ignore_patterns_hash: IgnorePatternsHash, |
|
80 | ignore_patterns_hash: IgnorePatternsHash, | |
103 | } |
|
81 | } | |
104 |
|
82 | |||
|
83 | /// Fields are documented in the *The data file format* | |||
|
84 | /// section of `mercurial/helptext/internals/dirstate-v2.txt` | |||
105 | #[derive(BytesCast)] |
|
85 | #[derive(BytesCast)] | |
106 | #[repr(C)] |
|
86 | #[repr(C)] | |
107 | pub(super) struct Node { |
|
87 | pub(super) struct Node { | |
108 | full_path: PathSlice, |
|
88 | full_path: PathSlice, | |
109 |
|
89 | |||
110 | /// In bytes from `self.full_path.start` |
|
90 | /// In bytes from `self.full_path.start` | |
111 | base_name_start: PathSize, |
|
91 | base_name_start: PathSize, | |
112 |
|
92 | |||
113 | copy_source: OptPathSlice, |
|
93 | copy_source: OptPathSlice, | |
114 | children: ChildNodes, |
|
94 | children: ChildNodes, | |
115 | pub(super) descendants_with_entry_count: Size, |
|
95 | pub(super) descendants_with_entry_count: Size, | |
116 | pub(super) tracked_descendants_count: Size, |
|
96 | pub(super) tracked_descendants_count: Size, | |
117 |
|
||||
118 | /// Depending on the bits in `flags`: |
|
|||
119 | /// |
|
|||
120 | /// * If any of `WDIR_TRACKED`, `P1_TRACKED`, or `P2_INFO` are set, the |
|
|||
121 | /// node has an entry. |
|
|||
122 | /// |
|
|||
123 | /// - If `HAS_MODE_AND_SIZE` is set, `data.mode` and `data.size` are |
|
|||
124 | /// meaningful. Otherwise they are set to zero |
|
|||
125 | /// - If `HAS_MTIME` is set, `data.mtime` is meaningful. Otherwise it is |
|
|||
126 | /// set to zero. |
|
|||
127 | /// |
|
|||
128 | /// * If none of `WDIR_TRACKED`, `P1_TRACKED`, `P2_INFO`, or `HAS_MTIME` |
|
|||
129 | /// are set, the node does not have an entry and `data` is set to all |
|
|||
130 | /// zeros. |
|
|||
131 | /// |
|
|||
132 | /// * If none of `WDIR_TRACKED`, `P1_TRACKED`, `P2_INFO` are set, but |
|
|||
133 | /// `HAS_MTIME` is set, the bytes of `data` should instead be |
|
|||
134 | /// interpreted as the `Timestamp` for the mtime of a cached directory. |
|
|||
135 | /// |
|
|||
136 | /// The presence of this combination of flags means that at some point, |
|
|||
137 | /// this path in the working directory was observed: |
|
|||
138 | /// |
|
|||
139 | /// - To be a directory |
|
|||
140 | /// - With the modification time as given by `Timestamp` |
|
|||
141 | /// - That timestamp was already strictly in the past when observed, |
|
|||
142 | /// meaning that later changes cannot happen in the same clock tick |
|
|||
143 | /// and must cause a different modification time (unless the system |
|
|||
144 | /// clock jumps back and we get unlucky, which is not impossible but |
|
|||
145 | /// but deemed unlikely enough). |
|
|||
146 | /// - All direct children of this directory (as returned by |
|
|||
147 | /// `std::fs::read_dir`) either have a corresponding dirstate node, or |
|
|||
148 | /// are ignored by ignore patterns whose hash is in |
|
|||
149 | /// `TreeMetadata::ignore_patterns_hash`. |
|
|||
150 | /// |
|
|||
151 | /// This means that if `std::fs::symlink_metadata` later reports the |
|
|||
152 | /// same modification time and ignored patterns havenβt changed, a run |
|
|||
153 | /// of status that is not listing ignored files can skip calling |
|
|||
154 | /// `std::fs::read_dir` again for this directory, iterate child |
|
|||
155 | /// dirstate nodes instead. |
|
|||
156 | flags: Flags, |
|
97 | flags: Flags, | |
157 | data: Entry, |
|
98 | data: Entry, | |
158 | } |
|
99 | } | |
159 |
|
100 | |||
160 | bitflags! { |
|
101 | bitflags! { | |
161 | #[derive(BytesCast)] |
|
102 | #[derive(BytesCast)] | |
162 | #[repr(C)] |
|
103 | #[repr(C)] | |
163 | struct Flags: u8 { |
|
104 | struct Flags: u8 { | |
164 | const WDIR_TRACKED = 1 << 0; |
|
105 | const WDIR_TRACKED = 1 << 0; | |
165 | const P1_TRACKED = 1 << 1; |
|
106 | const P1_TRACKED = 1 << 1; | |
166 | const P2_INFO = 1 << 2; |
|
107 | const P2_INFO = 1 << 2; | |
167 | const HAS_MODE_AND_SIZE = 1 << 3; |
|
108 | const HAS_MODE_AND_SIZE = 1 << 3; | |
168 | const HAS_MTIME = 1 << 4; |
|
109 | const HAS_MTIME = 1 << 4; | |
169 | } |
|
110 | } | |
170 | } |
|
111 | } | |
171 |
|
112 | |||
172 | #[derive(BytesCast, Copy, Clone, Debug)] |
|
113 | #[derive(BytesCast, Copy, Clone, Debug)] | |
173 | #[repr(C)] |
|
114 | #[repr(C)] | |
174 | struct Entry { |
|
115 | struct Entry { | |
175 | mode: I32Be, |
|
116 | mode: I32Be, | |
176 | mtime: I32Be, |
|
117 | mtime: I32Be, | |
177 | size: I32Be, |
|
118 | size: I32Be, | |
178 | } |
|
119 | } | |
179 |
|
120 | |||
180 | /// Duration since the Unix epoch |
|
121 | /// Duration since the Unix epoch | |
181 | #[derive(BytesCast, Copy, Clone, PartialEq)] |
|
122 | #[derive(BytesCast, Copy, Clone, PartialEq)] | |
182 | #[repr(C)] |
|
123 | #[repr(C)] | |
183 | pub(super) struct Timestamp { |
|
124 | pub(super) struct Timestamp { | |
184 | seconds: I64Be, |
|
125 | seconds: I64Be, | |
185 |
|
126 | |||
186 | /// In `0 .. 1_000_000_000`. |
|
127 | /// In `0 .. 1_000_000_000`. | |
187 | /// |
|
128 | /// | |
188 | /// This timestamp is later or earlier than `(seconds, 0)` by this many |
|
129 | /// This timestamp is later or earlier than `(seconds, 0)` by this many | |
189 | /// nanoseconds, if `seconds` is non-negative or negative, respectively. |
|
130 | /// nanoseconds, if `seconds` is non-negative or negative, respectively. | |
190 | nanoseconds: U32Be, |
|
131 | nanoseconds: U32Be, | |
191 | } |
|
132 | } | |
192 |
|
133 | |||
193 | /// Counted in bytes from the start of the file |
|
134 | /// Counted in bytes from the start of the file | |
194 | /// |
|
135 | /// | |
195 | /// NOTE: not supporting `.hg/dirstate` files larger than 4 GiB. |
|
136 | /// NOTE: not supporting `.hg/dirstate` files larger than 4 GiB. | |
196 | type Offset = U32Be; |
|
137 | type Offset = U32Be; | |
197 |
|
138 | |||
198 | /// Counted in number of items |
|
139 | /// Counted in number of items | |
199 | /// |
|
140 | /// | |
200 | /// NOTE: we choose not to support counting more than 4 billion nodes anywhere. |
|
141 | /// NOTE: we choose not to support counting more than 4 billion nodes anywhere. | |
201 | type Size = U32Be; |
|
142 | type Size = U32Be; | |
202 |
|
143 | |||
203 | /// Counted in bytes |
|
144 | /// Counted in bytes | |
204 | /// |
|
145 | /// | |
205 | /// NOTE: we choose not to support file names/paths longer than 64 KiB. |
|
146 | /// NOTE: we choose not to support file names/paths longer than 64 KiB. | |
206 | type PathSize = U16Be; |
|
147 | type PathSize = U16Be; | |
207 |
|
148 | |||
208 | /// A contiguous sequence of `len` times `Node`, representing the child nodes |
|
149 | /// A contiguous sequence of `len` times `Node`, representing the child nodes | |
209 | /// of either some other node or of the repository root. |
|
150 | /// of either some other node or of the repository root. | |
210 | /// |
|
151 | /// | |
211 | /// Always sorted by ascending `full_path`, to allow binary search. |
|
152 | /// Always sorted by ascending `full_path`, to allow binary search. | |
212 | /// Since nodes with the same parent nodes also have the same parent path, |
|
153 | /// Since nodes with the same parent nodes also have the same parent path, | |
213 | /// only the `base_name`s need to be compared during binary search. |
|
154 | /// only the `base_name`s need to be compared during binary search. | |
214 | #[derive(BytesCast, Copy, Clone)] |
|
155 | #[derive(BytesCast, Copy, Clone)] | |
215 | #[repr(C)] |
|
156 | #[repr(C)] | |
216 | struct ChildNodes { |
|
157 | struct ChildNodes { | |
217 | start: Offset, |
|
158 | start: Offset, | |
218 | len: Size, |
|
159 | len: Size, | |
219 | } |
|
160 | } | |
220 |
|
161 | |||
221 | /// A `HgPath` of `len` bytes |
|
162 | /// A `HgPath` of `len` bytes | |
222 | #[derive(BytesCast, Copy, Clone)] |
|
163 | #[derive(BytesCast, Copy, Clone)] | |
223 | #[repr(C)] |
|
164 | #[repr(C)] | |
224 | struct PathSlice { |
|
165 | struct PathSlice { | |
225 | start: Offset, |
|
166 | start: Offset, | |
226 | len: PathSize, |
|
167 | len: PathSize, | |
227 | } |
|
168 | } | |
228 |
|
169 | |||
229 | /// Either nothing if `start == 0`, or a `HgPath` of `len` bytes |
|
170 | /// Either nothing if `start == 0`, or a `HgPath` of `len` bytes | |
230 | type OptPathSlice = PathSlice; |
|
171 | type OptPathSlice = PathSlice; | |
231 |
|
172 | |||
232 | /// Unexpected file format found in `.hg/dirstate` with the "v2" format. |
|
173 | /// Unexpected file format found in `.hg/dirstate` with the "v2" format. | |
233 | /// |
|
174 | /// | |
234 | /// This should only happen if Mercurial is buggy or a repository is corrupted. |
|
175 | /// This should only happen if Mercurial is buggy or a repository is corrupted. | |
235 | #[derive(Debug)] |
|
176 | #[derive(Debug)] | |
236 | pub struct DirstateV2ParseError; |
|
177 | pub struct DirstateV2ParseError; | |
237 |
|
178 | |||
238 | impl From<DirstateV2ParseError> for HgError { |
|
179 | impl From<DirstateV2ParseError> for HgError { | |
239 | fn from(_: DirstateV2ParseError) -> Self { |
|
180 | fn from(_: DirstateV2ParseError) -> Self { | |
240 | HgError::corrupted("dirstate-v2 parse error") |
|
181 | HgError::corrupted("dirstate-v2 parse error") | |
241 | } |
|
182 | } | |
242 | } |
|
183 | } | |
243 |
|
184 | |||
244 | impl From<DirstateV2ParseError> for crate::DirstateError { |
|
185 | impl From<DirstateV2ParseError> for crate::DirstateError { | |
245 | fn from(error: DirstateV2ParseError) -> Self { |
|
186 | fn from(error: DirstateV2ParseError) -> Self { | |
246 | HgError::from(error).into() |
|
187 | HgError::from(error).into() | |
247 | } |
|
188 | } | |
248 | } |
|
189 | } | |
249 |
|
190 | |||
250 | impl<'on_disk> Docket<'on_disk> { |
|
191 | impl<'on_disk> Docket<'on_disk> { | |
251 | pub fn parents(&self) -> DirstateParents { |
|
192 | pub fn parents(&self) -> DirstateParents { | |
252 | use crate::Node; |
|
193 | use crate::Node; | |
253 | let p1 = Node::try_from(&self.header.parent_1[..USED_NODE_ID_BYTES]) |
|
194 | let p1 = Node::try_from(&self.header.parent_1[..USED_NODE_ID_BYTES]) | |
254 | .unwrap() |
|
195 | .unwrap() | |
255 | .clone(); |
|
196 | .clone(); | |
256 | let p2 = Node::try_from(&self.header.parent_2[..USED_NODE_ID_BYTES]) |
|
197 | let p2 = Node::try_from(&self.header.parent_2[..USED_NODE_ID_BYTES]) | |
257 | .unwrap() |
|
198 | .unwrap() | |
258 | .clone(); |
|
199 | .clone(); | |
259 | DirstateParents { p1, p2 } |
|
200 | DirstateParents { p1, p2 } | |
260 | } |
|
201 | } | |
261 |
|
202 | |||
262 | pub fn tree_metadata(&self) -> &[u8] { |
|
203 | pub fn tree_metadata(&self) -> &[u8] { | |
263 | self.header.metadata.as_bytes() |
|
204 | self.header.metadata.as_bytes() | |
264 | } |
|
205 | } | |
265 |
|
206 | |||
266 | pub fn data_size(&self) -> usize { |
|
207 | pub fn data_size(&self) -> usize { | |
267 | // This `unwrap` could only panic on a 16-bit CPU |
|
208 | // This `unwrap` could only panic on a 16-bit CPU | |
268 | self.header.data_size.get().try_into().unwrap() |
|
209 | self.header.data_size.get().try_into().unwrap() | |
269 | } |
|
210 | } | |
270 |
|
211 | |||
271 | pub fn data_filename(&self) -> String { |
|
212 | pub fn data_filename(&self) -> String { | |
272 | String::from_utf8(format_bytes!(b"dirstate.{}", self.uuid)).unwrap() |
|
213 | String::from_utf8(format_bytes!(b"dirstate.{}", self.uuid)).unwrap() | |
273 | } |
|
214 | } | |
274 | } |
|
215 | } | |
275 |
|
216 | |||
276 | pub fn read_docket( |
|
217 | pub fn read_docket( | |
277 | on_disk: &[u8], |
|
218 | on_disk: &[u8], | |
278 | ) -> Result<Docket<'_>, DirstateV2ParseError> { |
|
219 | ) -> Result<Docket<'_>, DirstateV2ParseError> { | |
279 | let (header, uuid) = |
|
220 | let (header, uuid) = | |
280 | DocketHeader::from_bytes(on_disk).map_err(|_| DirstateV2ParseError)?; |
|
221 | DocketHeader::from_bytes(on_disk).map_err(|_| DirstateV2ParseError)?; | |
281 | let uuid_size = header.uuid_size as usize; |
|
222 | let uuid_size = header.uuid_size as usize; | |
282 | if header.marker == *V2_FORMAT_MARKER && uuid.len() == uuid_size { |
|
223 | if header.marker == *V2_FORMAT_MARKER && uuid.len() == uuid_size { | |
283 | Ok(Docket { header, uuid }) |
|
224 | Ok(Docket { header, uuid }) | |
284 | } else { |
|
225 | } else { | |
285 | Err(DirstateV2ParseError) |
|
226 | Err(DirstateV2ParseError) | |
286 | } |
|
227 | } | |
287 | } |
|
228 | } | |
288 |
|
229 | |||
289 | pub(super) fn read<'on_disk>( |
|
230 | pub(super) fn read<'on_disk>( | |
290 | on_disk: &'on_disk [u8], |
|
231 | on_disk: &'on_disk [u8], | |
291 | metadata: &[u8], |
|
232 | metadata: &[u8], | |
292 | ) -> Result<DirstateMap<'on_disk>, DirstateV2ParseError> { |
|
233 | ) -> Result<DirstateMap<'on_disk>, DirstateV2ParseError> { | |
293 | if on_disk.is_empty() { |
|
234 | if on_disk.is_empty() { | |
294 | return Ok(DirstateMap::empty(on_disk)); |
|
235 | return Ok(DirstateMap::empty(on_disk)); | |
295 | } |
|
236 | } | |
296 | let (meta, _) = TreeMetadata::from_bytes(metadata) |
|
237 | let (meta, _) = TreeMetadata::from_bytes(metadata) | |
297 | .map_err(|_| DirstateV2ParseError)?; |
|
238 | .map_err(|_| DirstateV2ParseError)?; | |
298 | let dirstate_map = DirstateMap { |
|
239 | let dirstate_map = DirstateMap { | |
299 | on_disk, |
|
240 | on_disk, | |
300 | root: dirstate_map::ChildNodes::OnDisk(read_nodes( |
|
241 | root: dirstate_map::ChildNodes::OnDisk(read_nodes( | |
301 | on_disk, |
|
242 | on_disk, | |
302 | meta.root_nodes, |
|
243 | meta.root_nodes, | |
303 | )?), |
|
244 | )?), | |
304 | nodes_with_entry_count: meta.nodes_with_entry_count.get(), |
|
245 | nodes_with_entry_count: meta.nodes_with_entry_count.get(), | |
305 | nodes_with_copy_source_count: meta.nodes_with_copy_source_count.get(), |
|
246 | nodes_with_copy_source_count: meta.nodes_with_copy_source_count.get(), | |
306 | ignore_patterns_hash: meta.ignore_patterns_hash, |
|
247 | ignore_patterns_hash: meta.ignore_patterns_hash, | |
307 | unreachable_bytes: meta.unreachable_bytes.get(), |
|
248 | unreachable_bytes: meta.unreachable_bytes.get(), | |
308 | }; |
|
249 | }; | |
309 | Ok(dirstate_map) |
|
250 | Ok(dirstate_map) | |
310 | } |
|
251 | } | |
311 |
|
252 | |||
312 | impl Node { |
|
253 | impl Node { | |
313 | pub(super) fn full_path<'on_disk>( |
|
254 | pub(super) fn full_path<'on_disk>( | |
314 | &self, |
|
255 | &self, | |
315 | on_disk: &'on_disk [u8], |
|
256 | on_disk: &'on_disk [u8], | |
316 | ) -> Result<&'on_disk HgPath, DirstateV2ParseError> { |
|
257 | ) -> Result<&'on_disk HgPath, DirstateV2ParseError> { | |
317 | read_hg_path(on_disk, self.full_path) |
|
258 | read_hg_path(on_disk, self.full_path) | |
318 | } |
|
259 | } | |
319 |
|
260 | |||
320 | pub(super) fn base_name_start<'on_disk>( |
|
261 | pub(super) fn base_name_start<'on_disk>( | |
321 | &self, |
|
262 | &self, | |
322 | ) -> Result<usize, DirstateV2ParseError> { |
|
263 | ) -> Result<usize, DirstateV2ParseError> { | |
323 | let start = self.base_name_start.get(); |
|
264 | let start = self.base_name_start.get(); | |
324 | if start < self.full_path.len.get() { |
|
265 | if start < self.full_path.len.get() { | |
325 | let start = usize::try_from(start) |
|
266 | let start = usize::try_from(start) | |
326 | // u32 -> usize, could only panic on a 16-bit CPU |
|
267 | // u32 -> usize, could only panic on a 16-bit CPU | |
327 | .expect("dirstate-v2 base_name_start out of bounds"); |
|
268 | .expect("dirstate-v2 base_name_start out of bounds"); | |
328 | Ok(start) |
|
269 | Ok(start) | |
329 | } else { |
|
270 | } else { | |
330 | Err(DirstateV2ParseError) |
|
271 | Err(DirstateV2ParseError) | |
331 | } |
|
272 | } | |
332 | } |
|
273 | } | |
333 |
|
274 | |||
334 | pub(super) fn base_name<'on_disk>( |
|
275 | pub(super) fn base_name<'on_disk>( | |
335 | &self, |
|
276 | &self, | |
336 | on_disk: &'on_disk [u8], |
|
277 | on_disk: &'on_disk [u8], | |
337 | ) -> Result<&'on_disk HgPath, DirstateV2ParseError> { |
|
278 | ) -> Result<&'on_disk HgPath, DirstateV2ParseError> { | |
338 | let full_path = self.full_path(on_disk)?; |
|
279 | let full_path = self.full_path(on_disk)?; | |
339 | let base_name_start = self.base_name_start()?; |
|
280 | let base_name_start = self.base_name_start()?; | |
340 | Ok(HgPath::new(&full_path.as_bytes()[base_name_start..])) |
|
281 | Ok(HgPath::new(&full_path.as_bytes()[base_name_start..])) | |
341 | } |
|
282 | } | |
342 |
|
283 | |||
343 | pub(super) fn path<'on_disk>( |
|
284 | pub(super) fn path<'on_disk>( | |
344 | &self, |
|
285 | &self, | |
345 | on_disk: &'on_disk [u8], |
|
286 | on_disk: &'on_disk [u8], | |
346 | ) -> Result<dirstate_map::NodeKey<'on_disk>, DirstateV2ParseError> { |
|
287 | ) -> Result<dirstate_map::NodeKey<'on_disk>, DirstateV2ParseError> { | |
347 | Ok(WithBasename::from_raw_parts( |
|
288 | Ok(WithBasename::from_raw_parts( | |
348 | Cow::Borrowed(self.full_path(on_disk)?), |
|
289 | Cow::Borrowed(self.full_path(on_disk)?), | |
349 | self.base_name_start()?, |
|
290 | self.base_name_start()?, | |
350 | )) |
|
291 | )) | |
351 | } |
|
292 | } | |
352 |
|
293 | |||
353 | pub(super) fn has_copy_source<'on_disk>(&self) -> bool { |
|
294 | pub(super) fn has_copy_source<'on_disk>(&self) -> bool { | |
354 | self.copy_source.start.get() != 0 |
|
295 | self.copy_source.start.get() != 0 | |
355 | } |
|
296 | } | |
356 |
|
297 | |||
357 | pub(super) fn copy_source<'on_disk>( |
|
298 | pub(super) fn copy_source<'on_disk>( | |
358 | &self, |
|
299 | &self, | |
359 | on_disk: &'on_disk [u8], |
|
300 | on_disk: &'on_disk [u8], | |
360 | ) -> Result<Option<&'on_disk HgPath>, DirstateV2ParseError> { |
|
301 | ) -> Result<Option<&'on_disk HgPath>, DirstateV2ParseError> { | |
361 | Ok(if self.has_copy_source() { |
|
302 | Ok(if self.has_copy_source() { | |
362 | Some(read_hg_path(on_disk, self.copy_source)?) |
|
303 | Some(read_hg_path(on_disk, self.copy_source)?) | |
363 | } else { |
|
304 | } else { | |
364 | None |
|
305 | None | |
365 | }) |
|
306 | }) | |
366 | } |
|
307 | } | |
367 |
|
308 | |||
368 | fn has_entry(&self) -> bool { |
|
309 | fn has_entry(&self) -> bool { | |
369 | self.flags.intersects( |
|
310 | self.flags.intersects( | |
370 | Flags::WDIR_TRACKED | Flags::P1_TRACKED | Flags::P2_INFO, |
|
311 | Flags::WDIR_TRACKED | Flags::P1_TRACKED | Flags::P2_INFO, | |
371 | ) |
|
312 | ) | |
372 | } |
|
313 | } | |
373 |
|
314 | |||
374 | pub(super) fn node_data( |
|
315 | pub(super) fn node_data( | |
375 | &self, |
|
316 | &self, | |
376 | ) -> Result<dirstate_map::NodeData, DirstateV2ParseError> { |
|
317 | ) -> Result<dirstate_map::NodeData, DirstateV2ParseError> { | |
377 | if self.has_entry() { |
|
318 | if self.has_entry() { | |
378 | Ok(dirstate_map::NodeData::Entry(self.assume_entry())) |
|
319 | Ok(dirstate_map::NodeData::Entry(self.assume_entry())) | |
379 | } else if let Some(&mtime) = self.cached_directory_mtime() { |
|
320 | } else if let Some(&mtime) = self.cached_directory_mtime() { | |
380 | Ok(dirstate_map::NodeData::CachedDirectory { mtime }) |
|
321 | Ok(dirstate_map::NodeData::CachedDirectory { mtime }) | |
381 | } else { |
|
322 | } else { | |
382 | Ok(dirstate_map::NodeData::None) |
|
323 | Ok(dirstate_map::NodeData::None) | |
383 | } |
|
324 | } | |
384 | } |
|
325 | } | |
385 |
|
326 | |||
386 | pub(super) fn cached_directory_mtime(&self) -> Option<&Timestamp> { |
|
327 | pub(super) fn cached_directory_mtime(&self) -> Option<&Timestamp> { | |
387 | if self.flags.contains(Flags::HAS_MTIME) && !self.has_entry() { |
|
328 | if self.flags.contains(Flags::HAS_MTIME) && !self.has_entry() { | |
388 | Some(self.data.as_timestamp()) |
|
329 | Some(self.data.as_timestamp()) | |
389 | } else { |
|
330 | } else { | |
390 | None |
|
331 | None | |
391 | } |
|
332 | } | |
392 | } |
|
333 | } | |
393 |
|
334 | |||
394 | fn assume_entry(&self) -> DirstateEntry { |
|
335 | fn assume_entry(&self) -> DirstateEntry { | |
395 | // TODO: convert through raw bits instead? |
|
336 | // TODO: convert through raw bits instead? | |
396 | let wdir_tracked = self.flags.contains(Flags::WDIR_TRACKED); |
|
337 | let wdir_tracked = self.flags.contains(Flags::WDIR_TRACKED); | |
397 | let p1_tracked = self.flags.contains(Flags::P1_TRACKED); |
|
338 | let p1_tracked = self.flags.contains(Flags::P1_TRACKED); | |
398 | let p2_info = self.flags.contains(Flags::P2_INFO); |
|
339 | let p2_info = self.flags.contains(Flags::P2_INFO); | |
399 | let mode_size = if self.flags.contains(Flags::HAS_MODE_AND_SIZE) { |
|
340 | let mode_size = if self.flags.contains(Flags::HAS_MODE_AND_SIZE) { | |
400 | Some((self.data.mode.into(), self.data.size.into())) |
|
341 | Some((self.data.mode.into(), self.data.size.into())) | |
401 | } else { |
|
342 | } else { | |
402 | None |
|
343 | None | |
403 | }; |
|
344 | }; | |
404 | let mtime = if self.flags.contains(Flags::HAS_MTIME) { |
|
345 | let mtime = if self.flags.contains(Flags::HAS_MTIME) { | |
405 | Some(self.data.mtime.into()) |
|
346 | Some(self.data.mtime.into()) | |
406 | } else { |
|
347 | } else { | |
407 | None |
|
348 | None | |
408 | }; |
|
349 | }; | |
409 | DirstateEntry::from_v2_data( |
|
350 | DirstateEntry::from_v2_data( | |
410 | wdir_tracked, |
|
351 | wdir_tracked, | |
411 | p1_tracked, |
|
352 | p1_tracked, | |
412 | p2_info, |
|
353 | p2_info, | |
413 | mode_size, |
|
354 | mode_size, | |
414 | mtime, |
|
355 | mtime, | |
415 | ) |
|
356 | ) | |
416 | } |
|
357 | } | |
417 |
|
358 | |||
418 | pub(super) fn entry( |
|
359 | pub(super) fn entry( | |
419 | &self, |
|
360 | &self, | |
420 | ) -> Result<Option<DirstateEntry>, DirstateV2ParseError> { |
|
361 | ) -> Result<Option<DirstateEntry>, DirstateV2ParseError> { | |
421 | if self.has_entry() { |
|
362 | if self.has_entry() { | |
422 | Ok(Some(self.assume_entry())) |
|
363 | Ok(Some(self.assume_entry())) | |
423 | } else { |
|
364 | } else { | |
424 | Ok(None) |
|
365 | Ok(None) | |
425 | } |
|
366 | } | |
426 | } |
|
367 | } | |
427 |
|
368 | |||
428 | pub(super) fn children<'on_disk>( |
|
369 | pub(super) fn children<'on_disk>( | |
429 | &self, |
|
370 | &self, | |
430 | on_disk: &'on_disk [u8], |
|
371 | on_disk: &'on_disk [u8], | |
431 | ) -> Result<&'on_disk [Node], DirstateV2ParseError> { |
|
372 | ) -> Result<&'on_disk [Node], DirstateV2ParseError> { | |
432 | read_nodes(on_disk, self.children) |
|
373 | read_nodes(on_disk, self.children) | |
433 | } |
|
374 | } | |
434 |
|
375 | |||
435 | pub(super) fn to_in_memory_node<'on_disk>( |
|
376 | pub(super) fn to_in_memory_node<'on_disk>( | |
436 | &self, |
|
377 | &self, | |
437 | on_disk: &'on_disk [u8], |
|
378 | on_disk: &'on_disk [u8], | |
438 | ) -> Result<dirstate_map::Node<'on_disk>, DirstateV2ParseError> { |
|
379 | ) -> Result<dirstate_map::Node<'on_disk>, DirstateV2ParseError> { | |
439 | Ok(dirstate_map::Node { |
|
380 | Ok(dirstate_map::Node { | |
440 | children: dirstate_map::ChildNodes::OnDisk( |
|
381 | children: dirstate_map::ChildNodes::OnDisk( | |
441 | self.children(on_disk)?, |
|
382 | self.children(on_disk)?, | |
442 | ), |
|
383 | ), | |
443 | copy_source: self.copy_source(on_disk)?.map(Cow::Borrowed), |
|
384 | copy_source: self.copy_source(on_disk)?.map(Cow::Borrowed), | |
444 | data: self.node_data()?, |
|
385 | data: self.node_data()?, | |
445 | descendants_with_entry_count: self |
|
386 | descendants_with_entry_count: self | |
446 | .descendants_with_entry_count |
|
387 | .descendants_with_entry_count | |
447 | .get(), |
|
388 | .get(), | |
448 | tracked_descendants_count: self.tracked_descendants_count.get(), |
|
389 | tracked_descendants_count: self.tracked_descendants_count.get(), | |
449 | }) |
|
390 | }) | |
450 | } |
|
391 | } | |
451 | } |
|
392 | } | |
452 |
|
393 | |||
453 | impl Entry { |
|
394 | impl Entry { | |
454 | fn from_dirstate_entry(entry: &DirstateEntry) -> (Flags, Self) { |
|
395 | fn from_dirstate_entry(entry: &DirstateEntry) -> (Flags, Self) { | |
455 | let (wdir_tracked, p1_tracked, p2_info, mode_size_opt, mtime_opt) = |
|
396 | let (wdir_tracked, p1_tracked, p2_info, mode_size_opt, mtime_opt) = | |
456 | entry.v2_data(); |
|
397 | entry.v2_data(); | |
457 | // TODO: convert throug raw flag bits instead? |
|
398 | // TODO: convert throug raw flag bits instead? | |
458 | let mut flags = Flags::empty(); |
|
399 | let mut flags = Flags::empty(); | |
459 | flags.set(Flags::WDIR_TRACKED, wdir_tracked); |
|
400 | flags.set(Flags::WDIR_TRACKED, wdir_tracked); | |
460 | flags.set(Flags::P1_TRACKED, p1_tracked); |
|
401 | flags.set(Flags::P1_TRACKED, p1_tracked); | |
461 | flags.set(Flags::P2_INFO, p2_info); |
|
402 | flags.set(Flags::P2_INFO, p2_info); | |
462 | let (mode, size, mtime); |
|
403 | let (mode, size, mtime); | |
463 | if let Some((m, s)) = mode_size_opt { |
|
404 | if let Some((m, s)) = mode_size_opt { | |
464 | mode = m; |
|
405 | mode = m; | |
465 | size = s; |
|
406 | size = s; | |
466 | flags.insert(Flags::HAS_MODE_AND_SIZE) |
|
407 | flags.insert(Flags::HAS_MODE_AND_SIZE) | |
467 | } else { |
|
408 | } else { | |
468 | mode = 0; |
|
409 | mode = 0; | |
469 | size = 0; |
|
410 | size = 0; | |
470 | } |
|
411 | } | |
471 | if let Some(m) = mtime_opt { |
|
412 | if let Some(m) = mtime_opt { | |
472 | mtime = m; |
|
413 | mtime = m; | |
473 | flags.insert(Flags::HAS_MTIME); |
|
414 | flags.insert(Flags::HAS_MTIME); | |
474 | } else { |
|
415 | } else { | |
475 | mtime = 0; |
|
416 | mtime = 0; | |
476 | } |
|
417 | } | |
477 | let raw_entry = Entry { |
|
418 | let raw_entry = Entry { | |
478 | mode: mode.into(), |
|
419 | mode: mode.into(), | |
479 | size: size.into(), |
|
420 | size: size.into(), | |
480 | mtime: mtime.into(), |
|
421 | mtime: mtime.into(), | |
481 | }; |
|
422 | }; | |
482 | (flags, raw_entry) |
|
423 | (flags, raw_entry) | |
483 | } |
|
424 | } | |
484 |
|
425 | |||
485 | fn from_timestamp(timestamp: Timestamp) -> Self { |
|
426 | fn from_timestamp(timestamp: Timestamp) -> Self { | |
486 | // Safety: both types implement the `ByteCast` trait, so we could |
|
427 | // Safety: both types implement the `ByteCast` trait, so we could | |
487 | // safely use `as_bytes` and `from_bytes` to do this conversion. Using |
|
428 | // safely use `as_bytes` and `from_bytes` to do this conversion. Using | |
488 | // `transmute` instead makes the compiler check that the two types |
|
429 | // `transmute` instead makes the compiler check that the two types | |
489 | // have the same size, which eliminates the error case of |
|
430 | // have the same size, which eliminates the error case of | |
490 | // `from_bytes`. |
|
431 | // `from_bytes`. | |
491 | unsafe { std::mem::transmute::<Timestamp, Entry>(timestamp) } |
|
432 | unsafe { std::mem::transmute::<Timestamp, Entry>(timestamp) } | |
492 | } |
|
433 | } | |
493 |
|
434 | |||
494 | fn as_timestamp(&self) -> &Timestamp { |
|
435 | fn as_timestamp(&self) -> &Timestamp { | |
495 | // Safety: same as above in `from_timestamp` |
|
436 | // Safety: same as above in `from_timestamp` | |
496 | unsafe { &*(self as *const Entry as *const Timestamp) } |
|
437 | unsafe { &*(self as *const Entry as *const Timestamp) } | |
497 | } |
|
438 | } | |
498 | } |
|
439 | } | |
499 |
|
440 | |||
500 | impl Timestamp { |
|
441 | impl Timestamp { | |
501 | pub fn seconds(&self) -> i64 { |
|
442 | pub fn seconds(&self) -> i64 { | |
502 | self.seconds.get() |
|
443 | self.seconds.get() | |
503 | } |
|
444 | } | |
504 | } |
|
445 | } | |
505 |
|
446 | |||
506 | impl From<SystemTime> for Timestamp { |
|
447 | impl From<SystemTime> for Timestamp { | |
507 | fn from(system_time: SystemTime) -> Self { |
|
448 | fn from(system_time: SystemTime) -> Self { | |
508 | let (secs, nanos) = match system_time.duration_since(UNIX_EPOCH) { |
|
449 | let (secs, nanos) = match system_time.duration_since(UNIX_EPOCH) { | |
509 | Ok(duration) => { |
|
450 | Ok(duration) => { | |
510 | (duration.as_secs() as i64, duration.subsec_nanos()) |
|
451 | (duration.as_secs() as i64, duration.subsec_nanos()) | |
511 | } |
|
452 | } | |
512 | Err(error) => { |
|
453 | Err(error) => { | |
513 | let negative = error.duration(); |
|
454 | let negative = error.duration(); | |
514 | (-(negative.as_secs() as i64), negative.subsec_nanos()) |
|
455 | (-(negative.as_secs() as i64), negative.subsec_nanos()) | |
515 | } |
|
456 | } | |
516 | }; |
|
457 | }; | |
517 | Timestamp { |
|
458 | Timestamp { | |
518 | seconds: secs.into(), |
|
459 | seconds: secs.into(), | |
519 | nanoseconds: nanos.into(), |
|
460 | nanoseconds: nanos.into(), | |
520 | } |
|
461 | } | |
521 | } |
|
462 | } | |
522 | } |
|
463 | } | |
523 |
|
464 | |||
524 | impl From<&'_ Timestamp> for SystemTime { |
|
465 | impl From<&'_ Timestamp> for SystemTime { | |
525 | fn from(timestamp: &'_ Timestamp) -> Self { |
|
466 | fn from(timestamp: &'_ Timestamp) -> Self { | |
526 | let secs = timestamp.seconds.get(); |
|
467 | let secs = timestamp.seconds.get(); | |
527 | let nanos = timestamp.nanoseconds.get(); |
|
468 | let nanos = timestamp.nanoseconds.get(); | |
528 | if secs >= 0 { |
|
469 | if secs >= 0 { | |
529 | UNIX_EPOCH + Duration::new(secs as u64, nanos) |
|
470 | UNIX_EPOCH + Duration::new(secs as u64, nanos) | |
530 | } else { |
|
471 | } else { | |
531 | UNIX_EPOCH - Duration::new((-secs) as u64, nanos) |
|
472 | UNIX_EPOCH - Duration::new((-secs) as u64, nanos) | |
532 | } |
|
473 | } | |
533 | } |
|
474 | } | |
534 | } |
|
475 | } | |
535 |
|
476 | |||
536 | fn read_hg_path( |
|
477 | fn read_hg_path( | |
537 | on_disk: &[u8], |
|
478 | on_disk: &[u8], | |
538 | slice: PathSlice, |
|
479 | slice: PathSlice, | |
539 | ) -> Result<&HgPath, DirstateV2ParseError> { |
|
480 | ) -> Result<&HgPath, DirstateV2ParseError> { | |
540 | read_slice(on_disk, slice.start, slice.len.get()).map(HgPath::new) |
|
481 | read_slice(on_disk, slice.start, slice.len.get()).map(HgPath::new) | |
541 | } |
|
482 | } | |
542 |
|
483 | |||
543 | fn read_nodes( |
|
484 | fn read_nodes( | |
544 | on_disk: &[u8], |
|
485 | on_disk: &[u8], | |
545 | slice: ChildNodes, |
|
486 | slice: ChildNodes, | |
546 | ) -> Result<&[Node], DirstateV2ParseError> { |
|
487 | ) -> Result<&[Node], DirstateV2ParseError> { | |
547 | read_slice(on_disk, slice.start, slice.len.get()) |
|
488 | read_slice(on_disk, slice.start, slice.len.get()) | |
548 | } |
|
489 | } | |
549 |
|
490 | |||
550 | fn read_slice<T, Len>( |
|
491 | fn read_slice<T, Len>( | |
551 | on_disk: &[u8], |
|
492 | on_disk: &[u8], | |
552 | start: Offset, |
|
493 | start: Offset, | |
553 | len: Len, |
|
494 | len: Len, | |
554 | ) -> Result<&[T], DirstateV2ParseError> |
|
495 | ) -> Result<&[T], DirstateV2ParseError> | |
555 | where |
|
496 | where | |
556 | T: BytesCast, |
|
497 | T: BytesCast, | |
557 | Len: TryInto<usize>, |
|
498 | Len: TryInto<usize>, | |
558 | { |
|
499 | { | |
559 | // Either `usize::MAX` would result in "out of bounds" error since a single |
|
500 | // Either `usize::MAX` would result in "out of bounds" error since a single | |
560 | // `&[u8]` cannot occupy the entire addess space. |
|
501 | // `&[u8]` cannot occupy the entire addess space. | |
561 | let start = start.get().try_into().unwrap_or(std::usize::MAX); |
|
502 | let start = start.get().try_into().unwrap_or(std::usize::MAX); | |
562 | let len = len.try_into().unwrap_or(std::usize::MAX); |
|
503 | let len = len.try_into().unwrap_or(std::usize::MAX); | |
563 | on_disk |
|
504 | on_disk | |
564 | .get(start..) |
|
505 | .get(start..) | |
565 | .and_then(|bytes| T::slice_from_bytes(bytes, len).ok()) |
|
506 | .and_then(|bytes| T::slice_from_bytes(bytes, len).ok()) | |
566 | .map(|(slice, _rest)| slice) |
|
507 | .map(|(slice, _rest)| slice) | |
567 | .ok_or_else(|| DirstateV2ParseError) |
|
508 | .ok_or_else(|| DirstateV2ParseError) | |
568 | } |
|
509 | } | |
569 |
|
510 | |||
570 | pub(crate) fn for_each_tracked_path<'on_disk>( |
|
511 | pub(crate) fn for_each_tracked_path<'on_disk>( | |
571 | on_disk: &'on_disk [u8], |
|
512 | on_disk: &'on_disk [u8], | |
572 | metadata: &[u8], |
|
513 | metadata: &[u8], | |
573 | mut f: impl FnMut(&'on_disk HgPath), |
|
514 | mut f: impl FnMut(&'on_disk HgPath), | |
574 | ) -> Result<(), DirstateV2ParseError> { |
|
515 | ) -> Result<(), DirstateV2ParseError> { | |
575 | let (meta, _) = TreeMetadata::from_bytes(metadata) |
|
516 | let (meta, _) = TreeMetadata::from_bytes(metadata) | |
576 | .map_err(|_| DirstateV2ParseError)?; |
|
517 | .map_err(|_| DirstateV2ParseError)?; | |
577 | fn recur<'on_disk>( |
|
518 | fn recur<'on_disk>( | |
578 | on_disk: &'on_disk [u8], |
|
519 | on_disk: &'on_disk [u8], | |
579 | nodes: ChildNodes, |
|
520 | nodes: ChildNodes, | |
580 | f: &mut impl FnMut(&'on_disk HgPath), |
|
521 | f: &mut impl FnMut(&'on_disk HgPath), | |
581 | ) -> Result<(), DirstateV2ParseError> { |
|
522 | ) -> Result<(), DirstateV2ParseError> { | |
582 | for node in read_nodes(on_disk, nodes)? { |
|
523 | for node in read_nodes(on_disk, nodes)? { | |
583 | if let Some(entry) = node.entry()? { |
|
524 | if let Some(entry) = node.entry()? { | |
584 | if entry.state().is_tracked() { |
|
525 | if entry.state().is_tracked() { | |
585 | f(node.full_path(on_disk)?) |
|
526 | f(node.full_path(on_disk)?) | |
586 | } |
|
527 | } | |
587 | } |
|
528 | } | |
588 | recur(on_disk, node.children, f)? |
|
529 | recur(on_disk, node.children, f)? | |
589 | } |
|
530 | } | |
590 | Ok(()) |
|
531 | Ok(()) | |
591 | } |
|
532 | } | |
592 | recur(on_disk, meta.root_nodes, &mut f) |
|
533 | recur(on_disk, meta.root_nodes, &mut f) | |
593 | } |
|
534 | } | |
594 |
|
535 | |||
595 | /// Returns new data and metadata, together with whether that data should be |
|
536 | /// Returns new data and metadata, together with whether that data should be | |
596 | /// appended to the existing data file whose content is at |
|
537 | /// appended to the existing data file whose content is at | |
597 | /// `dirstate_map.on_disk` (true), instead of written to a new data file |
|
538 | /// `dirstate_map.on_disk` (true), instead of written to a new data file | |
598 | /// (false). |
|
539 | /// (false). | |
599 | pub(super) fn write( |
|
540 | pub(super) fn write( | |
600 | dirstate_map: &mut DirstateMap, |
|
541 | dirstate_map: &mut DirstateMap, | |
601 | can_append: bool, |
|
542 | can_append: bool, | |
602 | ) -> Result<(Vec<u8>, Vec<u8>, bool), DirstateError> { |
|
543 | ) -> Result<(Vec<u8>, Vec<u8>, bool), DirstateError> { | |
603 | let append = can_append && dirstate_map.write_should_append(); |
|
544 | let append = can_append && dirstate_map.write_should_append(); | |
604 |
|
545 | |||
605 | // This ignores the space for paths, and for nodes without an entry. |
|
546 | // This ignores the space for paths, and for nodes without an entry. | |
606 | // TODO: better estimate? Skip the `Vec` and write to a file directly? |
|
547 | // TODO: better estimate? Skip the `Vec` and write to a file directly? | |
607 | let size_guess = std::mem::size_of::<Node>() |
|
548 | let size_guess = std::mem::size_of::<Node>() | |
608 | * dirstate_map.nodes_with_entry_count as usize; |
|
549 | * dirstate_map.nodes_with_entry_count as usize; | |
609 |
|
550 | |||
610 | let mut writer = Writer { |
|
551 | let mut writer = Writer { | |
611 | dirstate_map, |
|
552 | dirstate_map, | |
612 | append, |
|
553 | append, | |
613 | out: Vec::with_capacity(size_guess), |
|
554 | out: Vec::with_capacity(size_guess), | |
614 | }; |
|
555 | }; | |
615 |
|
556 | |||
616 | let root_nodes = writer.write_nodes(dirstate_map.root.as_ref())?; |
|
557 | let root_nodes = writer.write_nodes(dirstate_map.root.as_ref())?; | |
617 |
|
558 | |||
618 | let meta = TreeMetadata { |
|
559 | let meta = TreeMetadata { | |
619 | root_nodes, |
|
560 | root_nodes, | |
620 | nodes_with_entry_count: dirstate_map.nodes_with_entry_count.into(), |
|
561 | nodes_with_entry_count: dirstate_map.nodes_with_entry_count.into(), | |
621 | nodes_with_copy_source_count: dirstate_map |
|
562 | nodes_with_copy_source_count: dirstate_map | |
622 | .nodes_with_copy_source_count |
|
563 | .nodes_with_copy_source_count | |
623 | .into(), |
|
564 | .into(), | |
624 | unreachable_bytes: dirstate_map.unreachable_bytes.into(), |
|
565 | unreachable_bytes: dirstate_map.unreachable_bytes.into(), | |
625 | unused: [0; 4], |
|
566 | unused: [0; 4], | |
626 | ignore_patterns_hash: dirstate_map.ignore_patterns_hash, |
|
567 | ignore_patterns_hash: dirstate_map.ignore_patterns_hash, | |
627 | }; |
|
568 | }; | |
628 | Ok((writer.out, meta.as_bytes().to_vec(), append)) |
|
569 | Ok((writer.out, meta.as_bytes().to_vec(), append)) | |
629 | } |
|
570 | } | |
630 |
|
571 | |||
631 | struct Writer<'dmap, 'on_disk> { |
|
572 | struct Writer<'dmap, 'on_disk> { | |
632 | dirstate_map: &'dmap DirstateMap<'on_disk>, |
|
573 | dirstate_map: &'dmap DirstateMap<'on_disk>, | |
633 | append: bool, |
|
574 | append: bool, | |
634 | out: Vec<u8>, |
|
575 | out: Vec<u8>, | |
635 | } |
|
576 | } | |
636 |
|
577 | |||
637 | impl Writer<'_, '_> { |
|
578 | impl Writer<'_, '_> { | |
638 | fn write_nodes( |
|
579 | fn write_nodes( | |
639 | &mut self, |
|
580 | &mut self, | |
640 | nodes: dirstate_map::ChildNodesRef, |
|
581 | nodes: dirstate_map::ChildNodesRef, | |
641 | ) -> Result<ChildNodes, DirstateError> { |
|
582 | ) -> Result<ChildNodes, DirstateError> { | |
642 | // Reuse already-written nodes if possible |
|
583 | // Reuse already-written nodes if possible | |
643 | if self.append { |
|
584 | if self.append { | |
644 | if let dirstate_map::ChildNodesRef::OnDisk(nodes_slice) = nodes { |
|
585 | if let dirstate_map::ChildNodesRef::OnDisk(nodes_slice) = nodes { | |
645 | let start = self.on_disk_offset_of(nodes_slice).expect( |
|
586 | let start = self.on_disk_offset_of(nodes_slice).expect( | |
646 | "dirstate-v2 OnDisk nodes not found within on_disk", |
|
587 | "dirstate-v2 OnDisk nodes not found within on_disk", | |
647 | ); |
|
588 | ); | |
648 | let len = child_nodes_len_from_usize(nodes_slice.len()); |
|
589 | let len = child_nodes_len_from_usize(nodes_slice.len()); | |
649 | return Ok(ChildNodes { start, len }); |
|
590 | return Ok(ChildNodes { start, len }); | |
650 | } |
|
591 | } | |
651 | } |
|
592 | } | |
652 |
|
593 | |||
653 | // `dirstate_map::ChildNodes::InMemory` contains a `HashMap` which has |
|
594 | // `dirstate_map::ChildNodes::InMemory` contains a `HashMap` which has | |
654 | // undefined iteration order. Sort to enable binary search in the |
|
595 | // undefined iteration order. Sort to enable binary search in the | |
655 | // written file. |
|
596 | // written file. | |
656 | let nodes = nodes.sorted(); |
|
597 | let nodes = nodes.sorted(); | |
657 | let nodes_len = nodes.len(); |
|
598 | let nodes_len = nodes.len(); | |
658 |
|
599 | |||
659 | // First accumulate serialized nodes in a `Vec` |
|
600 | // First accumulate serialized nodes in a `Vec` | |
660 | let mut on_disk_nodes = Vec::with_capacity(nodes_len); |
|
601 | let mut on_disk_nodes = Vec::with_capacity(nodes_len); | |
661 | for node in nodes { |
|
602 | for node in nodes { | |
662 | let children = |
|
603 | let children = | |
663 | self.write_nodes(node.children(self.dirstate_map.on_disk)?)?; |
|
604 | self.write_nodes(node.children(self.dirstate_map.on_disk)?)?; | |
664 | let full_path = node.full_path(self.dirstate_map.on_disk)?; |
|
605 | let full_path = node.full_path(self.dirstate_map.on_disk)?; | |
665 | let full_path = self.write_path(full_path.as_bytes()); |
|
606 | let full_path = self.write_path(full_path.as_bytes()); | |
666 | let copy_source = if let Some(source) = |
|
607 | let copy_source = if let Some(source) = | |
667 | node.copy_source(self.dirstate_map.on_disk)? |
|
608 | node.copy_source(self.dirstate_map.on_disk)? | |
668 | { |
|
609 | { | |
669 | self.write_path(source.as_bytes()) |
|
610 | self.write_path(source.as_bytes()) | |
670 | } else { |
|
611 | } else { | |
671 | PathSlice { |
|
612 | PathSlice { | |
672 | start: 0.into(), |
|
613 | start: 0.into(), | |
673 | len: 0.into(), |
|
614 | len: 0.into(), | |
674 | } |
|
615 | } | |
675 | }; |
|
616 | }; | |
676 | on_disk_nodes.push(match node { |
|
617 | on_disk_nodes.push(match node { | |
677 | NodeRef::InMemory(path, node) => { |
|
618 | NodeRef::InMemory(path, node) => { | |
678 | let (flags, data) = match &node.data { |
|
619 | let (flags, data) = match &node.data { | |
679 | dirstate_map::NodeData::Entry(entry) => { |
|
620 | dirstate_map::NodeData::Entry(entry) => { | |
680 | Entry::from_dirstate_entry(entry) |
|
621 | Entry::from_dirstate_entry(entry) | |
681 | } |
|
622 | } | |
682 | dirstate_map::NodeData::CachedDirectory { mtime } => { |
|
623 | dirstate_map::NodeData::CachedDirectory { mtime } => { | |
683 | (Flags::HAS_MTIME, Entry::from_timestamp(*mtime)) |
|
624 | (Flags::HAS_MTIME, Entry::from_timestamp(*mtime)) | |
684 | } |
|
625 | } | |
685 | dirstate_map::NodeData::None => ( |
|
626 | dirstate_map::NodeData::None => ( | |
686 | Flags::empty(), |
|
627 | Flags::empty(), | |
687 | Entry { |
|
628 | Entry { | |
688 | mode: 0.into(), |
|
629 | mode: 0.into(), | |
689 | size: 0.into(), |
|
630 | size: 0.into(), | |
690 | mtime: 0.into(), |
|
631 | mtime: 0.into(), | |
691 | }, |
|
632 | }, | |
692 | ), |
|
633 | ), | |
693 | }; |
|
634 | }; | |
694 | Node { |
|
635 | Node { | |
695 | children, |
|
636 | children, | |
696 | copy_source, |
|
637 | copy_source, | |
697 | full_path, |
|
638 | full_path, | |
698 | base_name_start: u16::try_from(path.base_name_start()) |
|
639 | base_name_start: u16::try_from(path.base_name_start()) | |
699 | // Could only panic for paths over 64 KiB |
|
640 | // Could only panic for paths over 64 KiB | |
700 | .expect("dirstate-v2 path length overflow") |
|
641 | .expect("dirstate-v2 path length overflow") | |
701 | .into(), |
|
642 | .into(), | |
702 | descendants_with_entry_count: node |
|
643 | descendants_with_entry_count: node | |
703 | .descendants_with_entry_count |
|
644 | .descendants_with_entry_count | |
704 | .into(), |
|
645 | .into(), | |
705 | tracked_descendants_count: node |
|
646 | tracked_descendants_count: node | |
706 | .tracked_descendants_count |
|
647 | .tracked_descendants_count | |
707 | .into(), |
|
648 | .into(), | |
708 | flags, |
|
649 | flags, | |
709 | data, |
|
650 | data, | |
710 | } |
|
651 | } | |
711 | } |
|
652 | } | |
712 | NodeRef::OnDisk(node) => Node { |
|
653 | NodeRef::OnDisk(node) => Node { | |
713 | children, |
|
654 | children, | |
714 | copy_source, |
|
655 | copy_source, | |
715 | full_path, |
|
656 | full_path, | |
716 | ..*node |
|
657 | ..*node | |
717 | }, |
|
658 | }, | |
718 | }) |
|
659 | }) | |
719 | } |
|
660 | } | |
720 | // β¦ so we can write them contiguously, after writing everything else |
|
661 | // β¦ so we can write them contiguously, after writing everything else | |
721 | // they refer to. |
|
662 | // they refer to. | |
722 | let start = self.current_offset(); |
|
663 | let start = self.current_offset(); | |
723 | let len = child_nodes_len_from_usize(nodes_len); |
|
664 | let len = child_nodes_len_from_usize(nodes_len); | |
724 | self.out.extend(on_disk_nodes.as_bytes()); |
|
665 | self.out.extend(on_disk_nodes.as_bytes()); | |
725 | Ok(ChildNodes { start, len }) |
|
666 | Ok(ChildNodes { start, len }) | |
726 | } |
|
667 | } | |
727 |
|
668 | |||
728 | /// If the given slice of items is within `on_disk`, returns its offset |
|
669 | /// If the given slice of items is within `on_disk`, returns its offset | |
729 | /// from the start of `on_disk`. |
|
670 | /// from the start of `on_disk`. | |
730 | fn on_disk_offset_of<T>(&self, slice: &[T]) -> Option<Offset> |
|
671 | fn on_disk_offset_of<T>(&self, slice: &[T]) -> Option<Offset> | |
731 | where |
|
672 | where | |
732 | T: BytesCast, |
|
673 | T: BytesCast, | |
733 | { |
|
674 | { | |
734 | fn address_range(slice: &[u8]) -> std::ops::RangeInclusive<usize> { |
|
675 | fn address_range(slice: &[u8]) -> std::ops::RangeInclusive<usize> { | |
735 | let start = slice.as_ptr() as usize; |
|
676 | let start = slice.as_ptr() as usize; | |
736 | let end = start + slice.len(); |
|
677 | let end = start + slice.len(); | |
737 | start..=end |
|
678 | start..=end | |
738 | } |
|
679 | } | |
739 | let slice_addresses = address_range(slice.as_bytes()); |
|
680 | let slice_addresses = address_range(slice.as_bytes()); | |
740 | let on_disk_addresses = address_range(self.dirstate_map.on_disk); |
|
681 | let on_disk_addresses = address_range(self.dirstate_map.on_disk); | |
741 | if on_disk_addresses.contains(slice_addresses.start()) |
|
682 | if on_disk_addresses.contains(slice_addresses.start()) | |
742 | && on_disk_addresses.contains(slice_addresses.end()) |
|
683 | && on_disk_addresses.contains(slice_addresses.end()) | |
743 | { |
|
684 | { | |
744 | let offset = slice_addresses.start() - on_disk_addresses.start(); |
|
685 | let offset = slice_addresses.start() - on_disk_addresses.start(); | |
745 | Some(offset_from_usize(offset)) |
|
686 | Some(offset_from_usize(offset)) | |
746 | } else { |
|
687 | } else { | |
747 | None |
|
688 | None | |
748 | } |
|
689 | } | |
749 | } |
|
690 | } | |
750 |
|
691 | |||
751 | fn current_offset(&mut self) -> Offset { |
|
692 | fn current_offset(&mut self) -> Offset { | |
752 | let mut offset = self.out.len(); |
|
693 | let mut offset = self.out.len(); | |
753 | if self.append { |
|
694 | if self.append { | |
754 | offset += self.dirstate_map.on_disk.len() |
|
695 | offset += self.dirstate_map.on_disk.len() | |
755 | } |
|
696 | } | |
756 | offset_from_usize(offset) |
|
697 | offset_from_usize(offset) | |
757 | } |
|
698 | } | |
758 |
|
699 | |||
759 | fn write_path(&mut self, slice: &[u8]) -> PathSlice { |
|
700 | fn write_path(&mut self, slice: &[u8]) -> PathSlice { | |
760 | let len = path_len_from_usize(slice.len()); |
|
701 | let len = path_len_from_usize(slice.len()); | |
761 | // Reuse an already-written path if possible |
|
702 | // Reuse an already-written path if possible | |
762 | if self.append { |
|
703 | if self.append { | |
763 | if let Some(start) = self.on_disk_offset_of(slice) { |
|
704 | if let Some(start) = self.on_disk_offset_of(slice) { | |
764 | return PathSlice { start, len }; |
|
705 | return PathSlice { start, len }; | |
765 | } |
|
706 | } | |
766 | } |
|
707 | } | |
767 | let start = self.current_offset(); |
|
708 | let start = self.current_offset(); | |
768 | self.out.extend(slice.as_bytes()); |
|
709 | self.out.extend(slice.as_bytes()); | |
769 | PathSlice { start, len } |
|
710 | PathSlice { start, len } | |
770 | } |
|
711 | } | |
771 | } |
|
712 | } | |
772 |
|
713 | |||
773 | fn offset_from_usize(x: usize) -> Offset { |
|
714 | fn offset_from_usize(x: usize) -> Offset { | |
774 | u32::try_from(x) |
|
715 | u32::try_from(x) | |
775 | // Could only panic for a dirstate file larger than 4 GiB |
|
716 | // Could only panic for a dirstate file larger than 4 GiB | |
776 | .expect("dirstate-v2 offset overflow") |
|
717 | .expect("dirstate-v2 offset overflow") | |
777 | .into() |
|
718 | .into() | |
778 | } |
|
719 | } | |
779 |
|
720 | |||
780 | fn child_nodes_len_from_usize(x: usize) -> Size { |
|
721 | fn child_nodes_len_from_usize(x: usize) -> Size { | |
781 | u32::try_from(x) |
|
722 | u32::try_from(x) | |
782 | // Could only panic with over 4 billion nodes |
|
723 | // Could only panic with over 4 billion nodes | |
783 | .expect("dirstate-v2 slice length overflow") |
|
724 | .expect("dirstate-v2 slice length overflow") | |
784 | .into() |
|
725 | .into() | |
785 | } |
|
726 | } | |
786 |
|
727 | |||
787 | fn path_len_from_usize(x: usize) -> PathSize { |
|
728 | fn path_len_from_usize(x: usize) -> PathSize { | |
788 | u16::try_from(x) |
|
729 | u16::try_from(x) | |
789 | // Could only panic for paths over 64 KiB |
|
730 | // Could only panic for paths over 64 KiB | |
790 | .expect("dirstate-v2 path length overflow") |
|
731 | .expect("dirstate-v2 path length overflow") | |
791 | .into() |
|
732 | .into() | |
792 | } |
|
733 | } |
General Comments 0
You need to be logged in to leave comments.
Login now