1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
//! lower-level access to filters which are applied to create working tree checkouts or to 'clean' working tree contents for storage in git.
use std::borrow::Cow;
pub use gix_filter as plumbing;
use gix_object::Find;
use crate::{
bstr::BStr,
config::{
cache::util::{ApplyLeniency, ApplyLeniencyDefaultValue},
tree::Core,
},
prelude::ObjectIdExt,
Repository,
};
///
pub mod pipeline {
///
pub mod options {
use crate::{bstr::BString, config};
/// The error returned by [Pipeline::options()](crate::filter::Pipeline::options()).
#[derive(Debug, thiserror::Error)]
#[allow(missing_docs)]
pub enum Error {
#[error(transparent)]
CheckRoundTripEncodings(#[from] config::encoding::Error),
#[error(transparent)]
SafeCrlf(#[from] config::key::GenericErrorWithValue),
#[error("Could not interpret 'filter.{name}.required' configuration")]
Driver {
name: BString,
source: gix_config::value::Error,
},
#[error(transparent)]
CommandContext(#[from] config::command_context::Error),
}
}
///
pub mod convert_to_git {
/// The error returned by [Pipeline::convert_to_git()](crate::filter::Pipeline::convert_to_git()).
#[derive(Debug, thiserror::Error)]
#[allow(missing_docs)]
pub enum Error {
#[error("Failed to prime attributes to the path at which the data resides")]
WorktreeCacheAtPath(#[from] std::io::Error),
#[error(transparent)]
Convert(#[from] gix_filter::pipeline::convert::to_git::Error),
}
}
///
pub mod convert_to_worktree {
/// The error returned by [Pipeline::convert_to_worktree()](crate::filter::Pipeline::convert_to_worktree()).
#[derive(Debug, thiserror::Error)]
#[allow(missing_docs)]
pub enum Error {
#[error("Failed to prime attributes to the path at which the data resides")]
WorktreeCacheAtPath(#[from] std::io::Error),
#[error(transparent)]
Convert(#[from] gix_filter::pipeline::convert::to_worktree::Error),
}
}
///
pub mod worktree_file_to_object {
use std::path::PathBuf;
/// The error returned by [Pipeline::worktree_file_to_object()](crate::filter::Pipeline::worktree_file_to_object()).
#[derive(Debug, thiserror::Error)]
#[allow(missing_docs)]
pub enum Error {
#[error("Cannot add worktree files in bare repositories")]
MissingWorktree,
#[error("Failed to perform IO for object creation for '{}'", path.display())]
IO { source: std::io::Error, path: PathBuf },
#[error(transparent)]
WriteBlob(#[from] crate::object::write::Error),
#[error(transparent)]
ConvertToGit(#[from] crate::filter::pipeline::convert_to_git::Error),
}
}
}
/// A git pipeline for transforming data *to-git* and *to-worktree*, based
/// [on git configuration and attributes](https://git-scm.com/docs/gitattributes).
#[derive(Clone)]
pub struct Pipeline<'repo> {
inner: gix_filter::Pipeline,
cache: gix_worktree::Stack,
/// The repository this pipeline is associated with.
pub repo: &'repo Repository,
}
/// Lifecycle
impl<'repo> Pipeline<'repo> {
/// Extract options from `repo` that are needed to properly drive a standard git filter pipeline.
pub fn options(repo: &'repo Repository) -> Result<gix_filter::pipeline::Options, pipeline::options::Error> {
let config = &repo.config.resolved;
let encodings =
Core::CHECK_ROUND_TRIP_ENCODING.try_into_encodings(config.string("core.checkRoundtripEncoding"))?;
let safe_crlf = config
.string("core.safecrlf")
.map(|value| Core::SAFE_CRLF.try_into_safecrlf(value))
.transpose()
.map(Option::unwrap_or_default)
.with_lenient_default_value(
repo.config.lenient_config,
// in lenient mode, we prefer the safe option, instead of just (trying) to output warnings.
gix_filter::pipeline::CrlfRoundTripCheck::Fail,
)?;
let auto_crlf = config
.string("core.autocrlf")
.map(|value| Core::AUTO_CRLF.try_into_autocrlf(value))
.transpose()
.with_leniency(repo.config.lenient_config)?
.unwrap_or_default();
let eol = config
.string("core.eol")
.map(|value| Core::EOL.try_into_eol(value))
.transpose()?;
let drivers = extract_drivers(repo)?;
Ok(gix_filter::pipeline::Options {
drivers,
eol_config: gix_filter::eol::Configuration { auto_crlf, eol },
encodings_with_roundtrip_check: encodings,
crlf_roundtrip_check: safe_crlf,
object_hash: repo.object_hash(),
})
}
/// Create a new instance by extracting all necessary information and configuration from a `repo` along with `cache` for accessing
/// attributes. The `index` is used for some filters which may access it under very specific circumstances.
pub fn new(repo: &'repo Repository, cache: gix_worktree::Stack) -> Result<Self, pipeline::options::Error> {
let pipeline = gix_filter::Pipeline::new(repo.command_context()?, Self::options(repo)?);
Ok(Pipeline {
inner: pipeline,
cache,
repo,
})
}
/// Detach the repository and obtain the individual functional parts.
pub fn into_parts(self) -> (gix_filter::Pipeline, gix_worktree::Stack) {
(self.inner, self.cache)
}
}
/// Conversions
impl Pipeline<'_> {
/// Convert a `src` stream (to be found at `rela_path`, a repo-relative path) to a representation suitable for storage in `git`
/// by using all attributes at `rela_path` and configuration of the repository to know exactly which filters apply.
/// `index` is used in particularly rare cases where the CRLF filter in auto-mode tries to determine whether to apply itself,
/// and it should match the state used when [instantiating this instance](Self::new()).
/// Note that the return-type implements [`std::io::Read`].
pub fn convert_to_git<R>(
&mut self,
src: R,
rela_path: &std::path::Path,
index: &gix_index::State,
) -> Result<gix_filter::pipeline::convert::ToGitOutcome<'_, R>, pipeline::convert_to_git::Error>
where
R: std::io::Read,
{
let entry = self.cache.at_path(rela_path, None, &self.repo.objects)?;
Ok(self.inner.convert_to_git(
src,
rela_path,
&mut |_, attrs| {
entry.matching_attributes(attrs);
},
&mut |buf| -> Result<_, gix_object::find::Error> {
let entry = match index
.entry_by_path(gix_path::to_unix_separators_on_windows(gix_path::into_bstr(rela_path)).as_ref())
{
None => return Ok(None),
Some(entry) => entry,
};
let obj = self.repo.objects.try_find(&entry.id, buf)?;
Ok(obj.filter(|obj| obj.kind == gix_object::Kind::Blob).map(|_| ()))
},
)?)
}
/// Convert a `src` buffer located at `rela_path` (in the index) from what's in `git` to the worktree representation.
/// This method will obtain all attributes and configuration necessary to know exactly which filters to apply.
/// Note that the return-type implements [`std::io::Read`].
///
/// Use `can_delay` to tell driver processes that they may delay the return of data. Doing this will require the caller to specifically
/// handle delayed files by keeping state and using [`Self::into_parts()`] to get access to the driver state to follow the delayed-files
/// protocol. For simplicity, most will want to disallow delayed processing.
pub fn convert_to_worktree<'input>(
&mut self,
src: &'input [u8],
rela_path: &BStr,
can_delay: gix_filter::driver::apply::Delay,
) -> Result<gix_filter::pipeline::convert::ToWorktreeOutcome<'input, '_>, pipeline::convert_to_worktree::Error>
{
let entry = self.cache.at_entry(rela_path, None, &self.repo.objects)?;
Ok(self.inner.convert_to_worktree(
src,
rela_path,
&mut |_, attrs| {
entry.matching_attributes(attrs);
},
can_delay,
)?)
}
/// Add the worktree file at `rela_path` to the object database and return its `(id, entry, symlink_metadata)` for use in a tree or in the index, for instance.
/// If `rela_path` is a directory *and* is a repository with its `HEAD` pointing to a commit, it will also be provided with the appropriate kind.
/// Note that this can easily lead to embedded repositories as no submodule-crosscheck is performed. Otherwise, unreadable repositories or directories
/// are ignored with `None` as return value.
///
/// `index` is used in particularly rare cases where the CRLF filter in auto-mode tries to determine whether to apply itself,
/// and it should match the state used when [instantiating this instance](Self::new()).
///
/// Return `Ok(None)` the file didn't exist in the worktree, or if it was of an untrackable type.
pub fn worktree_file_to_object(
&mut self,
rela_path: &BStr,
index: &gix_index::State,
) -> Result<
Option<(gix_hash::ObjectId, gix_object::tree::EntryKind, std::fs::Metadata)>,
pipeline::worktree_file_to_object::Error,
> {
use pipeline::worktree_file_to_object::Error;
let rela_path_as_path = gix_path::from_bstr(rela_path);
let repo = self.repo;
let worktree_dir = repo.workdir().ok_or(Error::MissingWorktree)?;
let path = worktree_dir.join(&rela_path_as_path);
let md = match std::fs::symlink_metadata(&path) {
Ok(md) => md,
Err(err) => {
if gix_fs::io_err::is_not_found(err.kind(), err.raw_os_error()) {
return Ok(None);
} else {
return Err(Error::IO { source: err, path });
}
}
};
let (id, kind) = if md.is_symlink() {
let target = std::fs::read_link(&path).map_err(|source| Error::IO { source, path })?;
let id = repo.write_blob(gix_path::into_bstr(target).as_ref())?;
(id, gix_object::tree::EntryKind::Link)
} else if md.is_file() {
use gix_filter::pipeline::convert::ToGitOutcome;
let file = std::fs::File::open(&path).map_err(|source| Error::IO { source, path })?;
let file_for_git = self.convert_to_git(file, rela_path_as_path.as_ref(), index)?;
let id = match file_for_git {
ToGitOutcome::Unchanged(mut file) => repo.write_blob_stream(&mut file)?,
ToGitOutcome::Buffer(buf) => repo.write_blob(buf)?,
ToGitOutcome::Process(mut read) => repo.write_blob_stream(&mut read)?,
};
let kind = if gix_fs::is_executable(&md) {
gix_object::tree::EntryKind::BlobExecutable
} else {
gix_object::tree::EntryKind::Blob
};
(id, kind)
} else if md.is_dir() {
let Some(submodule_repo) = crate::open_opts(&path, repo.open_options().clone()).ok() else {
return Ok(None);
};
let Some(id) = submodule_repo.head_id().ok() else {
return Ok(None);
};
(id.detach().attach(repo), gix_object::tree::EntryKind::Commit)
} else {
// This is probably a type-change to something we can't track.
return Ok(None);
};
Ok(Some((id.detach(), kind, md)))
}
/// Retrieve the static context that is made available to the process filters.
///
/// The context set here is relevant for the [`convert_to_git()`][Self::convert_to_git()] and
/// [`convert_to_worktree()`][Self::convert_to_worktree()] methods.
pub fn driver_context_mut(&mut self) -> &mut gix_filter::pipeline::Context {
self.inner.driver_context_mut()
}
}
/// Obtain a list of all configured driver, but ignore those in sections that we don't trust enough.
fn extract_drivers(repo: &Repository) -> Result<Vec<gix_filter::Driver>, pipeline::options::Error> {
repo.config
.resolved
.sections_by_name("filter")
.into_iter()
.flatten()
.filter(|s| repo.filter_config_section()(s.meta()))
.filter_map(|s| {
s.header().subsection_name().map(|name| {
Ok(gix_filter::Driver {
name: name.to_owned(),
clean: s.value("clean").map(Cow::into_owned),
smudge: s.value("smudge").map(Cow::into_owned),
process: s.value("process").map(Cow::into_owned),
required: s
.value("required")
.map(|value| gix_config::Boolean::try_from(value.as_ref()))
.transpose()
.map_err(|err| pipeline::options::Error::Driver {
name: name.to_owned(),
source: err,
})?
.unwrap_or_default()
.into(),
})
})
})
.collect::<Result<Vec<_>, pipeline::options::Error>>()
}