Compare commits
3
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
 Diomidis Spinellis
|
cb4ea5ef15 | ||
 mukunda katta
|
04023934fd | ||
|
mukunda kattaandDiomidis Spinellis
|
addc2db04a |
No files matched your search
@@ -49,19 +49,6 @@ jobs:
|
||||
|
||||
# Omit installing part of https://github.com/actions/runner-images/tree/main/images/ubuntu
|
||||
|
||||
### Locales
|
||||
# Several testsuite scripts guard themselves with require_*_locale_ and
|
||||
# silently skip when the locale is missing, which hides both regressions
|
||||
# and improvements in the multibyte code paths.
|
||||
- name: Generate the locales required by the testsuite
|
||||
shell: bash
|
||||
run: |
|
||||
## Generate the locales required by the testsuite
|
||||
sudo localedef -i en_US -f UTF-8 en_US.UTF-8
|
||||
sudo localedef -i ru_RU -f UTF-8 ru_RU.UTF-8
|
||||
sudo localedef -i el_GR -f ISO-8859-7 el_GR.iso88597
|
||||
locale -a
|
||||
|
||||
### Build
|
||||
- name: Build Rust sed binary
|
||||
shell: bash
|
||||
@@ -147,7 +134,7 @@ jobs:
|
||||
persist-credentials: false
|
||||
|
||||
- name: Retrieve reference artifacts
|
||||
uses: dawidd6/action-download-artifact@v25
|
||||
uses: dawidd6/action-download-artifact@v21
|
||||
continue-on-error: true
|
||||
with:
|
||||
workflow: GnuTests.yml
|
||||
|
||||
@@ -1,41 +0,0 @@
|
||||
name: Documentation
|
||||
|
||||
on:
|
||||
pull_request:
|
||||
paths:
|
||||
- 'docs/**'
|
||||
- '.github/workflows/documentation.yml'
|
||||
push:
|
||||
branches:
|
||||
- main
|
||||
paths:
|
||||
- 'docs/**'
|
||||
- '.github/workflows/documentation.yml'
|
||||
|
||||
permissions:
|
||||
contents: read
|
||||
|
||||
jobs:
|
||||
build-doc:
|
||||
name: Build the mdBook documentation
|
||||
runs-on: ubuntu-latest
|
||||
|
||||
steps:
|
||||
- name: Checkout repository
|
||||
uses: actions/checkout@v7
|
||||
with:
|
||||
persist-credentials: false
|
||||
|
||||
- name: Install mdbook
|
||||
uses: taiki-e/install-action@v2
|
||||
with:
|
||||
tool: mdbook@0.5.0
|
||||
|
||||
- name: Build documentation
|
||||
run: mdbook build docs
|
||||
|
||||
- name: Upload documentation
|
||||
uses: actions/upload-artifact@v7
|
||||
with:
|
||||
name: documentation
|
||||
path: docs/book
|
||||
@@ -64,7 +64,7 @@ jobs:
|
||||
# we specify bash to get pipefail; it guards against the `curl` command
|
||||
# failing. otherwise `sh` won't catch that `curl` returned non-0
|
||||
shell: bash
|
||||
run: "curl --proto '=https' --tlsv1.2 -LsSf https://github.com/axodotdev/cargo-dist/releases/download/v0.33.0/cargo-dist-installer.sh | sh"
|
||||
run: "curl --proto '=https' --tlsv1.2 -LsSf https://github.com/axodotdev/cargo-dist/releases/download/v0.30.3/cargo-dist-installer.sh | sh"
|
||||
- name: Cache dist
|
||||
uses: actions/upload-artifact@v7
|
||||
with:
|
||||
|
||||
Generated
+625
-77
File diff suppressed because it is too large.
Load diff
+10
-26
@@ -1,12 +1,12 @@
|
||||
# sed (uutils)
|
||||
# * see the repository LICENSE, README, and CONTRIBUTING files for more information
|
||||
|
||||
# spell-checker:ignore (libs) mangen sed binstall chrono codspeed libc memchr memmap rlimit rustix textwrap uucore uutests xattr
|
||||
# spell-checker:ignore (misc) clippy codegen msvc startswith uninlined uutils
|
||||
# spell-checker:ignore (libs) mangen sed
|
||||
|
||||
[package]
|
||||
name = "sed"
|
||||
version = "0.3.0"
|
||||
version = "0.1.1"
|
||||
authors = ["uutils developers"]
|
||||
license = "MIT"
|
||||
description = "sed ~ implemented as universal (cross-platform) utils, written in Rust"
|
||||
default-run = "sed"
|
||||
@@ -21,19 +21,6 @@ rust-version = "1.88"
|
||||
|
||||
build = "build.rs"
|
||||
|
||||
# cargo-binstall: fetch the archives published by `dist` in the GitHub releases
|
||||
# instead of building from source. The layout below has to stay in sync with
|
||||
# `dist-workspace.toml` (targets) and the archives it produces:
|
||||
# sed-<target>.tar.xz (zip on Windows), containing a sed-<target>/ directory.
|
||||
# Note that the release tags carry no `v` prefix, hence `{ version }`.
|
||||
[package.metadata.binstall]
|
||||
pkg-url = "{ repo }/releases/download/{ version }/{ name }-{ target }.{ archive-suffix }"
|
||||
bin-dir = "{ name }-{ target }/{ bin }{ binary-ext }"
|
||||
pkg-fmt = "txz"
|
||||
|
||||
[package.metadata.binstall.overrides.x86_64-pc-windows-msvc]
|
||||
pkg-fmt = "zip"
|
||||
|
||||
[features]
|
||||
default = ["feat_common_core"]
|
||||
|
||||
@@ -42,6 +29,7 @@ sed = []
|
||||
|
||||
[workspace.dependencies]
|
||||
assert_fs = "1.1.3"
|
||||
bytesize = "2.0.0"
|
||||
chrono = { version = "0.4.37", default-features = false, features = [
|
||||
"clock",
|
||||
] }
|
||||
@@ -59,12 +47,12 @@ phf_codegen = "0.14.0"
|
||||
predicates = "3.1.3"
|
||||
rand = { version = "0.10.0" }
|
||||
regex = "1.10.4"
|
||||
sha2 = { version = "0.11.0", default-features = false, features = ["alloc"] }
|
||||
sha2 = "0.11"
|
||||
sysinfo = "0.38"
|
||||
tempfile = "3.10.1"
|
||||
terminal_size = "0.4.2"
|
||||
textwrap = { version = "0.16.1", features = ["terminal_size"] }
|
||||
uucore = { version = "0.12.0", features = ["libc"] }
|
||||
rustix = "1.1.4"
|
||||
uucore = { version = "0.9.0", features = ["libc"] }
|
||||
xattr = "1.3.1"
|
||||
|
||||
|
||||
@@ -78,11 +66,11 @@ memmap2.workspace = true
|
||||
phf = { workspace = true }
|
||||
predicates = { workspace = true }
|
||||
regex = { workspace = true }
|
||||
sysinfo = { workspace = true }
|
||||
tempfile = { workspace = true }
|
||||
terminal_size = { workspace = true }
|
||||
textwrap = { workspace = true }
|
||||
uucore = { workspace = true }
|
||||
rustix = { workspace = true }
|
||||
|
||||
[dev-dependencies]
|
||||
assert_fs = { workspace = true }
|
||||
@@ -97,7 +85,7 @@ regex = { workspace = true }
|
||||
sha2 = { workspace = true }
|
||||
tempfile = { workspace = true }
|
||||
uucore = { workspace = true, features = ["entries", "process", "signals", "benchmark"] }
|
||||
uutests = "0.12.0"
|
||||
uutests = "0.9.0"
|
||||
|
||||
[target.'cfg(unix)'.dev-dependencies]
|
||||
xattr = { workspace = true }
|
||||
@@ -131,10 +119,6 @@ inherits = "release"
|
||||
opt-level = "z"
|
||||
strip = true
|
||||
|
||||
# The profile that 'dist' will build with
|
||||
[profile.dist]
|
||||
inherits = "release"
|
||||
|
||||
[lints.clippy]
|
||||
all = { level = "warn", priority = -1 }
|
||||
|
||||
@@ -151,7 +135,7 @@ must_use_candidate = "allow" # 56
|
||||
needless_raw_string_hashes = "allow" # 20
|
||||
needless_pass_by_value = "allow" # 15
|
||||
missing_panics_doc = "allow" # 12
|
||||
# cast_possible_truncation = "allow" # 7
|
||||
cast_possible_truncation = "allow" # 7
|
||||
unnecessary_wraps = "allow" # 6
|
||||
match_wildcard_for_single_variants = "allow" # 4
|
||||
cast_sign_loss = "allow" # 4
|
||||
|
||||
@@ -15,30 +15,22 @@ and other extensions.
|
||||
## Status
|
||||
|
||||
At this state _sed_ implements all [POSIX features](https://pubs.opengroup.org/onlinepubs/9799919799/)
|
||||
and can run correctly the three complex scripts of its integration tests:
|
||||
[hanoi.sed](https://github.com/uutils/sed/blob/main/tests/fixtures/sed/script/hanoi.sed) (solves the Towers of Hanoi puzzle),
|
||||
[mandelbrot.sed](https://github.com/uutils/sed/blob/main/tests/fixtures/sed/script/mandelbrot.sed), (draws the Mandelbrot set) and
|
||||
and can run correctly the two complex scripts of its integration tests:
|
||||
[hanoi.sed](https://github.com/uutils/sed/blob/main/tests/fixtures/sed/script/hanoi.sed) (solves the Towers of Hanoi puzzle) and
|
||||
[math.sed](https://github.com/uutils/sed/blob/main/tests/fixtures/sed/script/math.sed) (implements an arbitrary precision integer math calculator).
|
||||
|
||||
The performance of this Rust implementation is now better than the GNU and FreeBSD implementations for most benchmarked cases.
|
||||
|
||||
Further work aims to:
|
||||
* improve GNU _sed_ compatibility, especially on the regular expression front,
|
||||
* implement more GNU extensions, and
|
||||
* improve performance where possible.
|
||||
* Adjust buffering on terminal output to match current implementations,
|
||||
* Implement more GNU extensions,
|
||||
* Improve performance where possible.
|
||||
|
||||
## Installation and Use
|
||||
|
||||
We provide a Linux x86_64 binary archive from the main branch at
|
||||
https://github.com/uutils/sed/releases/tag/latest-commit .
|
||||
|
||||
If you have [cargo-binstall](https://github.com/cargo-bins/cargo-binstall),
|
||||
the released binaries can be installed directly with:
|
||||
|
||||
```bash
|
||||
cargo binstall sed
|
||||
```
|
||||
|
||||
For other platforms, ensure you have Rust installed on your system. You can install Rust through [rustup](https://rustup.rs/).
|
||||
|
||||
Clone the repository and build the project using Cargo:
|
||||
@@ -88,8 +80,65 @@ cargo test
|
||||
```
|
||||
|
||||
## Extensions and incompatibilities
|
||||
The GNU, BSD and new extensions _sed_ supports, and where it differs from GNU
|
||||
_sed_, are listed in [docs/src/extensions.md](docs/src/extensions.md).
|
||||
### Supported GNU extensions
|
||||
* Command-line arguments can be specified in long (`--`) form.
|
||||
* Spaces can precede a regular expression modifier.
|
||||
* `I` can be used in as a synonym for the `i` (case insensitive) substitution
|
||||
flag.
|
||||
* `M` and `m` substitution flags allow multi-line matching.
|
||||
* In addition to `\n`, other escape sequences (octal, hex, C) are supported
|
||||
in the strings of the `y` command.
|
||||
Under POSIX these yield undefined behavior.
|
||||
* The `a`, `c`, and `i` commands do not require an initial backslash,
|
||||
allow text to appear on the same line, and support escape sequences
|
||||
in the specified text.
|
||||
* The `a`, `i`, `=`, `l`, `q` and `r` commands support address range as an extension to POSIX.
|
||||
* The substitution command replacement group `\0` is a synonym for &.
|
||||
* A `Q` command (optionally followed by an exit code) quits immediately.
|
||||
* The `q` command can be optionally followed by an exit code.
|
||||
* The `l` command can be optionally followed by the output width.
|
||||
* The `--follow-symlinks` option for in-place editing.
|
||||
* The `--sandbox` option that limits potentially destructive commands.
|
||||
* Address 0 can be used to specify an address range that is already
|
||||
active on line 1 and can finish with the specified regular expression.
|
||||
* Address steps can be specified in the form of start~step and start,~step
|
||||
ranges.
|
||||
* Address 0 can be used in the `r` command to prepend a file.
|
||||
|
||||
### Supported BSD and GNU extensions
|
||||
* The second address in a range can be specified as a relative address with +N.
|
||||
* In-place editing of file with the `-i` flag.
|
||||
|
||||
### New extensions
|
||||
* Unicode characters can be specified in regular expression pattern, replacement
|
||||
and transliteration sequences using `\uXXXX` or `\UXXXXXXXX` sequences.
|
||||
* The `l` command lists Unicode characters using the `\uXXXX` and `\UXXXXXXXX`
|
||||
sequences.
|
||||
|
||||
### Incompatibilities
|
||||
* Similarly to GNU _sed_, input is processed as raw bytes or as valid UTF-8
|
||||
(this includes 7-bit ASCII) based on the locale as specified by the
|
||||
`LC_ALL`, `LC_CTYPE`, and `LANG` environment variables,
|
||||
with the default being byte processing.
|
||||
However, in contrast with GNU _sed_, other locales (e.g. ISO-8859-1)
|
||||
are not supported. If the input is in another code page or encoding
|
||||
and requires locale-specific processing (e.g. ignore/map case,
|
||||
character classes), consider converting it through UTF-8 to ensure
|
||||
the correct handling of locale-specific regular expressions.
|
||||
This _sed_ program can also handle arbitrary byte sequences
|
||||
if no part of the input requires treating it as a Rust String.
|
||||
* Back-references aren't supported when input is processed as bytes
|
||||
(`LC_ALL=C`).
|
||||
* The command will report an error and fail if duplicate labels are found
|
||||
in the script.
|
||||
This matches the BSD behavior. The GNU version accepts duplicate labels.
|
||||
* The last line (`$`) address is interpreted as the last non-empty line of
|
||||
the last file. If files specified in subsequent arguments until the last
|
||||
one are empty, then the last line condition will never be triggered.
|
||||
This behavior is consistent with the
|
||||
[original implementation](https://github.com/dspinellis/unix-history-repo/blob/Research-V7/usr/src/cmd/sed/sed1.c#L665).
|
||||
* Labels are parsed for alphanumeric characters. The BSD version parses them
|
||||
until the end of the line, preventing ; to be used as a separator.
|
||||
|
||||
## GNU test suite compatibility
|
||||
|
||||
|
||||
+2
-2
@@ -4,7 +4,7 @@ members = ["cargo:."]
|
||||
# Config for 'dist'
|
||||
[dist]
|
||||
# The preferred dist version to use in CI (Cargo.toml SemVer syntax)
|
||||
cargo-dist-version = "0.33.0"
|
||||
cargo-dist-version = "0.30.3"
|
||||
# CI backends to support
|
||||
ci = "github"
|
||||
# The installers to generate for each app
|
||||
@@ -13,5 +13,5 @@ installers = []
|
||||
targets = ["aarch64-apple-darwin", "aarch64-unknown-linux-gnu", "x86_64-apple-darwin", "x86_64-unknown-linux-gnu", "x86_64-unknown-linux-musl", "x86_64-pc-windows-msvc"]
|
||||
# Which actions to run on pull requests
|
||||
pr-run-mode = "plan"
|
||||
# Skip checking whether the specified configuration files are up to date
|
||||
# Allow dirty workflows to ignore out-of-date contents
|
||||
allow-dirty = ["ci"]
|
||||
@@ -1 +0,0 @@
|
||||
book
|
||||
@@ -1,7 +0,0 @@
|
||||
[book]
|
||||
language = "en"
|
||||
src = "src"
|
||||
title = "uutils sed Documentation"
|
||||
|
||||
[output.html]
|
||||
git-repository-url = "https://github.com/uutils/sed/tree/main/docs/src"
|
||||
@@ -1,5 +0,0 @@
|
||||
# Summary
|
||||
|
||||
[Introduction](index.md)
|
||||
|
||||
* [Extensions and incompatibilities](extensions.md)
|
||||
@@ -1,73 +0,0 @@
|
||||
# Extensions and incompatibilities
|
||||
|
||||
The main goal of the project is compatibility with GNU _sed_, but _sed_ also
|
||||
supports features that GNU _sed_ does not, and differs from it in a few places.
|
||||
Below is a list of these extensions and incompatibilities.
|
||||
|
||||
## Supported GNU extensions
|
||||
* Command-line arguments can be specified in long (`--`) form.
|
||||
* Spaces can precede a regular expression modifier.
|
||||
* `I` can be used in as a synonym for the `i` (case insensitive) substitution
|
||||
flag.
|
||||
* `M` and `m` substitution flags allow multi-line matching.
|
||||
* In addition to `\n`, other escape sequences (octal, hex, C) are supported
|
||||
in the strings of the `y` command.
|
||||
Under POSIX these yield undefined behavior.
|
||||
* The `a`, `c`, and `i` commands do not require an initial backslash,
|
||||
allow text to appear on the same line, and support escape sequences
|
||||
in the specified text.
|
||||
* The `a`, `i`, `=`, `l`, `q` and `r` commands support address range as an extension to POSIX.
|
||||
* The substitution command replacement group `\0` is a synonym for &.
|
||||
* An `F` command outputs the name of the file currently being processed.
|
||||
* A `Q` command (optionally followed by an exit code) quits immediately.
|
||||
* The `q` command can be optionally followed by an exit code.
|
||||
* A `W` command writes to a file the pattern's first line.
|
||||
* An `R` command reads one line at a time from a file.
|
||||
* The `l` command can be optionally followed by the output width.
|
||||
* The `--follow-symlinks` option for in-place editing.
|
||||
* The `--sandbox` option that limits potentially destructive commands.
|
||||
* Address 0 can be used to specify an address range that is already
|
||||
active on line 1 and can finish with the specified regular expression.
|
||||
* Address steps can be specified in the form of start~step and start,~step
|
||||
ranges.
|
||||
* Address 0 can be used in the `r` command to prepend a file.
|
||||
|
||||
## Supported BSD and GNU extensions
|
||||
* The second address in a range can be specified as a relative address with +N.
|
||||
* In-place editing of file with the `-i` flag.
|
||||
|
||||
## New extensions
|
||||
* Unicode characters can be specified in regular expression pattern, replacement
|
||||
and transliteration sequences using `\uXXXX` or `\UXXXXXXXX` sequences.
|
||||
|
||||
## Incompatible extensions
|
||||
The `-U` or `--uutil-extensions` option enables useful extensions or bug fixes
|
||||
that aren't compatible with GNU sed or POSIX.
|
||||
|
||||
* The `l` command lists Unicode characters using the `\uXXXX` and `\UXXXXXXXX`
|
||||
escapes rather than as octal UTF-8 byte sequences.
|
||||
|
||||
## Incompatibilities
|
||||
* Similarly to GNU _sed_, input is processed as raw bytes or as valid UTF-8
|
||||
(this includes 7-bit ASCII) based on the locale as specified by the
|
||||
`LC_ALL`, `LC_CTYPE`, and `LANG` environment variables,
|
||||
with the default being byte processing.
|
||||
However, in contrast with GNU _sed_, other locales (e.g. ISO-8859-1)
|
||||
are not supported. If the input is in another code page or encoding
|
||||
and requires locale-specific processing (e.g. ignore/map case,
|
||||
character classes), consider converting it through UTF-8 to ensure
|
||||
the correct handling of locale-specific regular expressions.
|
||||
This _sed_ program can also handle arbitrary byte sequences
|
||||
if no part of the input requires treating it as a Rust String.
|
||||
* Back-references aren't supported when input is processed as bytes
|
||||
(`LC_ALL=C`).
|
||||
* The command will report an error and fail if duplicate labels are found
|
||||
in the script.
|
||||
This matches the BSD behavior. The GNU version accepts duplicate labels.
|
||||
* The last line (`$`) address is interpreted as the last non-empty line of
|
||||
the last file. If files specified in subsequent arguments until the last
|
||||
one are empty, then the last line condition will never be triggered.
|
||||
This behavior is consistent with the
|
||||
[original implementation](https://github.com/dspinellis/unix-history-repo/blob/Research-V7/usr/src/cmd/sed/sed1.c#L665).
|
||||
* Labels are parsed for alphanumeric characters. The BSD version parses them
|
||||
until the end of the line, preventing ; to be used as a separator.
|
||||
@@ -1,10 +0,0 @@
|
||||
# uutils sed
|
||||
|
||||
_sed_ is a Rust reimplementation of the
|
||||
[sed utility](https://pubs.opengroup.org/onlinepubs/9799919799/utilities/sed.html)
|
||||
with some [GNU sed](https://www.gnu.org/software/sed/manual/sed.html),
|
||||
[FreeBSD sed](https://man.freebsd.org/cgi/man.cgi?sed(1)),
|
||||
and other extensions.
|
||||
|
||||
It is part of the [uutils](https://uutils.github.io/) project. The source code
|
||||
is on [GitHub](https://github.com/uutils/sed).
|
||||
Generated
+130
-15
@@ -245,9 +245,9 @@ checksum = "1d07550c9036bf2ae0c684c4297d503f838287c83c53686d05370d0e139ae570"
|
||||
|
||||
[[package]]
|
||||
name = "console"
|
||||
version = "0.16.6"
|
||||
version = "0.16.4"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "e96a4956774c13c126a8b5af4daa79384f4d826534c95a02d76afb39e2ab64e3"
|
||||
checksum = "4fe5f465a4f6fee88fad41b85d990f84c835335e85b5d9e6e63e0d06d28cba7c"
|
||||
dependencies = [
|
||||
"encode_unicode",
|
||||
"libc",
|
||||
@@ -646,6 +646,15 @@ version = "0.3.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "61807f77802ff30975e01f4f071c8ba10c022052f98b3294119f3e615d13e5be"
|
||||
|
||||
[[package]]
|
||||
name = "ntapi"
|
||||
version = "0.4.3"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "c3b335231dfd352ffb0f8017f3b6027a4917f7df785ea2143d8af2adc66980ae"
|
||||
dependencies = [
|
||||
"winapi",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "num-bigint"
|
||||
version = "0.4.6"
|
||||
@@ -674,6 +683,25 @@ dependencies = [
|
||||
"autocfg",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "objc2-core-foundation"
|
||||
version = "0.3.2"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "2a180dd8642fa45cdb7dd721cd4c11b1cadd4929ce112ebd8b9f5803cc79d536"
|
||||
dependencies = [
|
||||
"bitflags",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "objc2-io-kit"
|
||||
version = "0.3.2"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "33fafba39597d6dc1fb709123dfa8289d39406734be322956a69f0931c73bb15"
|
||||
dependencies = [
|
||||
"libc",
|
||||
"objc2-core-foundation",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "once_cell"
|
||||
version = "1.21.4"
|
||||
@@ -820,9 +848,9 @@ checksum = "f8dcc9c7d52a811697d2151c701e0d08956f92b0e24136cf4cf27b57a6a0d9bf"
|
||||
|
||||
[[package]]
|
||||
name = "rand"
|
||||
version = "0.10.3"
|
||||
version = "0.10.2"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "65c9fb96cbc91e3478eaae79a69fcd3f1ae4ad052e471fe6732fff548984b4af"
|
||||
checksum = "c7f5fa3a058cd35567ef9bfa5e75732bee0f9e4c55fa90477bef2dfcdbc4be80"
|
||||
dependencies = [
|
||||
"chacha20",
|
||||
"getrandom 0.4.2",
|
||||
@@ -897,7 +925,7 @@ checksum = "b39cdef0fa800fc44525c84ccb54a029961a8215f9619753635a9c0d2538d46d"
|
||||
|
||||
[[package]]
|
||||
name = "sed"
|
||||
version = "0.3.0"
|
||||
version = "0.1.1"
|
||||
dependencies = [
|
||||
"clap",
|
||||
"clap_complete",
|
||||
@@ -909,7 +937,7 @@ dependencies = [
|
||||
"phf_codegen",
|
||||
"predicates",
|
||||
"regex",
|
||||
"rustix",
|
||||
"sysinfo",
|
||||
"tempfile",
|
||||
"terminal_size",
|
||||
"textwrap",
|
||||
@@ -985,9 +1013,9 @@ checksum = "703d5c7ef118737c72f1af64ad2f6f8c5e1921f818cdcb97b8fe6fc69bf66214"
|
||||
|
||||
[[package]]
|
||||
name = "similar"
|
||||
version = "3.2.0"
|
||||
version = "3.1.1"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "4f66ca1f7aca2474dc10c942eb22feffc897735f54cd1db90138c2fddb490987"
|
||||
checksum = "e6505efef05804732ed8a3f2d4f279429eb485bd69d5b0cc6b19cc02005cda16"
|
||||
dependencies = [
|
||||
"bstr",
|
||||
]
|
||||
@@ -1027,6 +1055,20 @@ dependencies = [
|
||||
"unicode-ident",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "sysinfo"
|
||||
version = "0.38.4"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "92ab6a2f8bfe508deb3c6406578252e491d299cbbf3bc0529ecc3313aee4a52f"
|
||||
dependencies = [
|
||||
"libc",
|
||||
"memchr",
|
||||
"ntapi",
|
||||
"objc2-core-foundation",
|
||||
"objc2-io-kit",
|
||||
"windows",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "tempfile"
|
||||
version = "3.27.0"
|
||||
@@ -1152,9 +1194,9 @@ checksum = "06abde3611657adf66d383f00b093d7faecc7fa57071cce2578660c9f1010821"
|
||||
|
||||
[[package]]
|
||||
name = "uucore"
|
||||
version = "0.12.0"
|
||||
version = "0.9.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "f24a5910ebbf2c0baf486a9125705f8e026a8f7741a3ea8cf5eca3c9257c9bdf"
|
||||
checksum = "069b34217c27f611e1589f540f58118dbf226a9e407d38ab472052ff075a1dc2"
|
||||
dependencies = [
|
||||
"bigdecimal",
|
||||
"clap",
|
||||
@@ -1167,9 +1209,9 @@ dependencies = [
|
||||
"os_display",
|
||||
"procfs",
|
||||
"rustc-hash",
|
||||
"rustix",
|
||||
"thiserror",
|
||||
"unic-langid",
|
||||
"unicode-width",
|
||||
"uucore_procs",
|
||||
"wild",
|
||||
]
|
||||
@@ -1191,9 +1233,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "uucore_procs"
|
||||
version = "0.12.0"
|
||||
version = "0.9.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "50aa1f3bba55168888ae6f1f2714ce58897d8020d742d21f474f01e86468fedd"
|
||||
checksum = "34337da211e7abfff7189b794afb3b5018fe356fb36474b73645458fc1201350"
|
||||
dependencies = [
|
||||
"proc-macro2",
|
||||
"quote",
|
||||
@@ -1201,9 +1243,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "uufuzz"
|
||||
version = "0.12.0"
|
||||
version = "0.9.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "7147da7bd2de08e6647116e005b0992cc602e34c34972793844a45c4e8648dd9"
|
||||
checksum = "8e4192b67d5e7f0a3300db353ec3217beee89442f02c11d8ae108867eee392d0"
|
||||
dependencies = [
|
||||
"console",
|
||||
"rand",
|
||||
@@ -1319,6 +1361,49 @@ dependencies = [
|
||||
"glob",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "winapi"
|
||||
version = "0.3.9"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
|
||||
dependencies = [
|
||||
"winapi-i686-pc-windows-gnu",
|
||||
"winapi-x86_64-pc-windows-gnu",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "winapi-i686-pc-windows-gnu"
|
||||
version = "0.4.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
|
||||
|
||||
[[package]]
|
||||
name = "winapi-x86_64-pc-windows-gnu"
|
||||
version = "0.4.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
|
||||
|
||||
[[package]]
|
||||
name = "windows"
|
||||
version = "0.62.2"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "527fadee13e0c05939a6a05d5bd6eec6cd2e3dbd648b9f8e447c6518133d8580"
|
||||
dependencies = [
|
||||
"windows-collections",
|
||||
"windows-core",
|
||||
"windows-future",
|
||||
"windows-numerics",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "windows-collections"
|
||||
version = "0.3.2"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "23b2d95af1a8a14a3c7367e1ed4fc9c20e0a26e79551b1454d72583c97cc6610"
|
||||
dependencies = [
|
||||
"windows-core",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "windows-core"
|
||||
version = "0.62.2"
|
||||
@@ -1332,6 +1417,17 @@ dependencies = [
|
||||
"windows-strings",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "windows-future"
|
||||
version = "0.3.2"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "e1d6f90251fe18a279739e78025bd6ddc52a7e22f921070ccdc67dde84c605cb"
|
||||
dependencies = [
|
||||
"windows-core",
|
||||
"windows-link",
|
||||
"windows-threading",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "windows-implement"
|
||||
version = "0.60.2"
|
||||
@@ -1360,6 +1456,16 @@ version = "0.2.1"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "f0805222e57f7521d6a62e36fa9163bc891acd422f971defe97d64e70d0a4fe5"
|
||||
|
||||
[[package]]
|
||||
name = "windows-numerics"
|
||||
version = "0.3.1"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "6e2e40844ac143cdb44aead537bbf727de9b044e107a0f1220392177d15b0f26"
|
||||
dependencies = [
|
||||
"windows-core",
|
||||
"windows-link",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "windows-result"
|
||||
version = "0.4.1"
|
||||
@@ -1387,6 +1493,15 @@ dependencies = [
|
||||
"windows-link",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "windows-threading"
|
||||
version = "0.2.1"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "3949bd5b99cafdf1c7ca86b43ca564028dfe27d66958f2470940f73d86d75b37"
|
||||
dependencies = [
|
||||
"windows-link",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "wit-bindgen"
|
||||
version = "0.51.0"
|
||||
|
||||
+2
-2
@@ -14,8 +14,8 @@ libc = "0.2.153"
|
||||
tempfile = "3.15.0"
|
||||
rand = { version = "0.10.0" }
|
||||
similar = "3.0.0"
|
||||
uucore = { version = "0.12.0", features = ["libc"] }
|
||||
uufuzz = "0.12.0"
|
||||
uucore = { version = "0.9.0", features = ["libc"] }
|
||||
uufuzz = "0.9.0"
|
||||
|
||||
sed = { path = ".." }
|
||||
|
||||
|
||||
+7
-13
@@ -10,7 +10,6 @@
|
||||
|
||||
use crate::sed::error_handling::{ScriptLocation, runtime_error};
|
||||
use crate::sed::fast_regex::{Captures, Match, Regex};
|
||||
use crate::sed::named_reader::NamedReader;
|
||||
use crate::sed::named_writer::NamedWriter;
|
||||
use crate::sed::script_char_provider::ScriptCharProvider;
|
||||
use crate::sed::script_line_provider::ScriptLineProvider;
|
||||
@@ -39,11 +38,10 @@ pub struct ProcessingContext {
|
||||
pub sandbox: bool,
|
||||
pub unbuffered: bool,
|
||||
pub null_data: bool,
|
||||
pub uutil_extensions: bool,
|
||||
|
||||
// Other context
|
||||
/// Currently processed input file name (not script)
|
||||
pub input_name: PathBuf,
|
||||
/// Currently processed input file name (not script) in quoted form
|
||||
pub input_name: String,
|
||||
/// Current input line number
|
||||
pub line_number: usize,
|
||||
/// True if this is the last address of a range
|
||||
@@ -68,9 +66,6 @@ pub struct ProcessingContext {
|
||||
pub parsed_block_nesting: usize,
|
||||
/// Command associated with each label
|
||||
pub label_to_command_map: HashMap<String, Rc<RefCell<Command>>>,
|
||||
/// Shared `R` readers keyed by file path, so that all `R` commands naming
|
||||
/// the same file advance a single cursor, matching GNU sed.
|
||||
pub named_readers: HashMap<PathBuf, Rc<RefCell<NamedReader>>>,
|
||||
/// Commands with a (latchable and resetable) address range
|
||||
pub range_commands: Vec<Rc<RefCell<Command>>>,
|
||||
/// True if a substitution was made as specified in the t command
|
||||
@@ -260,12 +255,12 @@ impl Default for Transliteration {
|
||||
/// Create a new Transliteration with identity mapping for the fast-path.
|
||||
fn default() -> Self {
|
||||
let mut fast = [0u8; 256];
|
||||
for (slot, value) in fast.iter_mut().zip(0..=u8::MAX) {
|
||||
*slot = value;
|
||||
for (i, slot) in fast.iter_mut().enumerate() {
|
||||
*slot = i as u8;
|
||||
}
|
||||
let mut unicode_fast = ['\0'; COMMON_UNICODE];
|
||||
for (slot, cp) in unicode_fast.iter_mut().zip(0u32..) {
|
||||
*slot = char::from_u32(cp).unwrap_or('\0');
|
||||
for (i, slot) in unicode_fast.iter_mut().enumerate() {
|
||||
*slot = char::from_u32(i as u32).unwrap_or('\0');
|
||||
}
|
||||
Self {
|
||||
byte_fast: fast,
|
||||
@@ -372,7 +367,6 @@ pub enum CommandData {
|
||||
BranchTarget(Option<Rc<RefCell<Command>>>), // Commands for 'b', 't', 'T', '{'
|
||||
Label(Option<String>), // Label name for 'b', 't', 'T', ':'
|
||||
Path(PathBuf), // File path for 'r'
|
||||
NamedReader(Rc<RefCell<NamedReader>>), // Successive file lines for 'R'
|
||||
NamedWriter(Rc<RefCell<NamedWriter>>), // File output for 'w'
|
||||
Number(usize), // Number for 'l', 'q', 'Q' (GNU)
|
||||
Substitution(Box<Substitution>), // Substitute command 's'
|
||||
@@ -599,7 +593,7 @@ mod tests {
|
||||
fn test_all_fast_path_mapped_to_space() {
|
||||
let mut t = Transliteration::default();
|
||||
for cp in 0..COMMON_UNICODE {
|
||||
if let Some(ch) = u32::try_from(cp).ok().and_then(char::from_u32) {
|
||||
if let Some(ch) = char::from_u32(cp as u32) {
|
||||
t.insert(ch, ' ');
|
||||
}
|
||||
}
|
||||
|
||||
+31
-168
@@ -18,7 +18,6 @@ use crate::sed::delimited_parser::{
|
||||
};
|
||||
use crate::sed::error_handling::{ScriptLocation, compilation_error, semantic_error};
|
||||
use crate::sed::fast_regex::Regex;
|
||||
use crate::sed::named_reader::NamedReader;
|
||||
use crate::sed::named_writer::NamedWriter;
|
||||
use crate::sed::script_char_provider::ScriptCharProvider;
|
||||
use crate::sed::script_line_provider::{ScriptLineProvider, ScriptValue};
|
||||
@@ -359,7 +358,7 @@ fn compile_address_range(
|
||||
let is_step_match = line.current() == '~'; // E.g. 0~2: Pick even-numbered lines
|
||||
line.advance();
|
||||
line.eat_spaces();
|
||||
let is_step_end = if !line.eol() && line.current() == '~' {
|
||||
let is_step_end = if line.current() == '~' {
|
||||
// E.g. /foo/,~10: Start at foo, include all lines until multiple of 10 is reached.
|
||||
line.advance();
|
||||
line.eat_spaces();
|
||||
@@ -373,46 +372,39 @@ fn compile_address_range(
|
||||
return compilation_error(lines, line, "~step is invalid in POSIX mode");
|
||||
}
|
||||
|
||||
// Look for second address; a missing ~step counts as 0.
|
||||
let has_addr2 = !line.eol() && (is_address_char(line.current()) || line.current() == '+');
|
||||
if !(has_addr2 || is_step_match || is_step_end) {
|
||||
return compilation_error(lines, line, "unexpected `,'");
|
||||
}
|
||||
let addr2 = if has_addr2 {
|
||||
compile_address(lines, line, context)?
|
||||
} else {
|
||||
Address::Line(0)
|
||||
};
|
||||
// Set step_n to the number specified in the (required numeric) address.
|
||||
let step_n = if is_step_match || is_step_end {
|
||||
match addr2 {
|
||||
Address::Line(n) => n,
|
||||
_ => {
|
||||
return compilation_error(
|
||||
lines,
|
||||
line,
|
||||
"~step can only be specified through numeric values",
|
||||
);
|
||||
// Look for second address.
|
||||
if !line.eol() {
|
||||
let addr2 = compile_address(lines, line, context)?;
|
||||
// Set step_n to the number specified in the (required numeric) address.
|
||||
let step_n = if is_step_match || is_step_end {
|
||||
match addr2 {
|
||||
Address::Line(n) => n,
|
||||
_ => {
|
||||
return compilation_error(
|
||||
lines,
|
||||
line,
|
||||
"~step can only be specified through numeric values",
|
||||
);
|
||||
}
|
||||
}
|
||||
} else {
|
||||
0 // dummy, not used
|
||||
};
|
||||
|
||||
if is_line0 && !matches!(addr2, Address::Re(_)) && !is_step_match {
|
||||
return compilation_error(lines, line, ERR_ADDRESS_0_USAGE);
|
||||
}
|
||||
} else {
|
||||
0 // dummy, not used
|
||||
};
|
||||
|
||||
// Address 0 needs a regex end or a positive ~step (0~0 selects no line).
|
||||
if is_line0 && !matches!(addr2, Address::Re(_)) && !(is_step_match && step_n > 0) {
|
||||
return compilation_error(lines, line, ERR_ADDRESS_0_USAGE);
|
||||
// If needed, transform Address::Line into Address::Step*.
|
||||
cmd.addr2 = if is_step_match {
|
||||
Some(Address::StepMatch(step_n))
|
||||
} else if is_step_end {
|
||||
Some(Address::StepEnd(step_n))
|
||||
} else {
|
||||
Some(addr2)
|
||||
};
|
||||
n_addr += 1;
|
||||
}
|
||||
|
||||
// If needed, transform Address::Line into Address::Step*.
|
||||
cmd.addr2 = if is_step_match {
|
||||
Some(Address::StepMatch(step_n))
|
||||
} else if is_step_end {
|
||||
Some(Address::StepEnd(step_n))
|
||||
} else {
|
||||
Some(addr2)
|
||||
};
|
||||
n_addr += 1;
|
||||
}
|
||||
|
||||
// Zero-address read command check
|
||||
@@ -503,7 +495,7 @@ fn compile_address(
|
||||
let number = parse_number(lines, line, true)?.unwrap();
|
||||
Ok(Address::Line(number))
|
||||
}
|
||||
_ => compilation_error(lines, line, "expected context address"),
|
||||
_ => panic!("invalid context address"),
|
||||
}
|
||||
}
|
||||
|
||||
@@ -1130,28 +1122,6 @@ fn compile_read_file_command(
|
||||
Ok(CommandHandling::Continue)
|
||||
}
|
||||
|
||||
// Handles R
|
||||
fn compile_read_line_command(
|
||||
lines: &mut ScriptLineProvider,
|
||||
line: &mut ScriptCharProvider,
|
||||
cmd: &mut Command,
|
||||
context: &mut ProcessingContext,
|
||||
) -> UResult<CommandHandling> {
|
||||
if context.sandbox {
|
||||
return compilation_error(lines, line, ERR_SANDBOX);
|
||||
}
|
||||
let path = read_file_path(lines, line)?;
|
||||
// Reuse a reader keyed by path so all `R` commands on the same file share
|
||||
// one cursor (see `ProcessingContext::named_readers`).
|
||||
let reader = context
|
||||
.named_readers
|
||||
.entry(path.clone())
|
||||
.or_insert_with(|| NamedReader::new(path))
|
||||
.clone();
|
||||
cmd.data = CommandData::NamedReader(reader);
|
||||
Ok(CommandHandling::Continue)
|
||||
}
|
||||
|
||||
// Handles w
|
||||
fn compile_write_file_command(
|
||||
lines: &mut ScriptLineProvider,
|
||||
@@ -1403,72 +1373,6 @@ fn compile_text_command_posix(
|
||||
Ok(CommandHandling::Continue)
|
||||
}
|
||||
|
||||
// Handle v
|
||||
fn compile_version_command(
|
||||
lines: &mut ScriptLineProvider,
|
||||
line: &mut ScriptCharProvider,
|
||||
_cmd: &mut Command,
|
||||
_context: &mut ProcessingContext,
|
||||
) -> UResult<CommandHandling> {
|
||||
// Claim version partify with GNU sed 4.9
|
||||
const GNU_MAJOR: u8 = 4;
|
||||
const GNU_MINOR: u8 = 9;
|
||||
const GNU_PATCH: u8 = 0;
|
||||
|
||||
line.advance();
|
||||
line.eat_spaces(); // Skip any leading whitespace.
|
||||
|
||||
let mut major = String::new();
|
||||
let mut minor = String::new();
|
||||
let mut patch = String::new();
|
||||
|
||||
let mut ver_semantic = 0;
|
||||
|
||||
while !line.eol() {
|
||||
if line.current() == '.' {
|
||||
ver_semantic += 1;
|
||||
line.advance();
|
||||
} else {
|
||||
match ver_semantic {
|
||||
0 => major.push(line.current()),
|
||||
1 => minor.push(line.current()),
|
||||
2 => patch.push(line.current()),
|
||||
_ => return compilation_error(lines, line, "invalid version of sed"),
|
||||
}
|
||||
line.advance();
|
||||
}
|
||||
}
|
||||
|
||||
if major.is_empty() {
|
||||
major = GNU_MAJOR.to_string();
|
||||
minor = GNU_MINOR.to_string();
|
||||
patch = GNU_PATCH.to_string();
|
||||
}
|
||||
|
||||
if minor.is_empty() {
|
||||
minor.push('0');
|
||||
}
|
||||
if patch.is_empty() {
|
||||
patch.push('0');
|
||||
}
|
||||
|
||||
match major.parse::<u8>() {
|
||||
Ok(major_int) => match minor.parse::<u8>() {
|
||||
Ok(minor_int) => match patch.parse::<u8>() {
|
||||
Ok(patch_int) => {
|
||||
if minor_int <= GNU_MINOR && major_int <= GNU_MAJOR && patch_int == GNU_PATCH {
|
||||
return Ok(CommandHandling::Continue);
|
||||
}
|
||||
compilation_error(lines, line, "expected newer version of sed")
|
||||
}
|
||||
Err(_) => compilation_error(lines, line, "invalid version of sed"),
|
||||
},
|
||||
Err(_) => compilation_error(lines, line, "invalid version of sed"),
|
||||
},
|
||||
Err(_) => compilation_error(lines, line, "invalid version of sed"),
|
||||
}
|
||||
}
|
||||
|
||||
// Handles e
|
||||
// With no argument, the command executes the pattern space as a shell
|
||||
// command at runtime. With an argument, the rest of the line is the
|
||||
@@ -1662,14 +1566,6 @@ fn get_cmd_spec(
|
||||
n_addr: 2,
|
||||
handler: compile_execute_command,
|
||||
}),
|
||||
'F' if !posix => Ok(CommandSpec {
|
||||
n_addr: 2,
|
||||
handler: compile_empty_command,
|
||||
}),
|
||||
'R' if !posix => Ok(CommandSpec {
|
||||
n_addr: 2,
|
||||
handler: compile_read_line_command,
|
||||
}),
|
||||
'r' => Ok(CommandSpec {
|
||||
n_addr: if posix { 1 } else { 2 },
|
||||
handler: compile_read_file_command,
|
||||
@@ -1682,10 +1578,6 @@ fn get_cmd_spec(
|
||||
n_addr: 2,
|
||||
handler: compile_label_command,
|
||||
}),
|
||||
'W' if !posix => Ok(CommandSpec {
|
||||
n_addr: 2,
|
||||
handler: compile_write_file_command,
|
||||
}),
|
||||
'w' => Ok(CommandSpec {
|
||||
n_addr: 2,
|
||||
handler: compile_write_file_command,
|
||||
@@ -1694,10 +1586,6 @@ fn get_cmd_spec(
|
||||
n_addr: 2,
|
||||
handler: compile_trans_command,
|
||||
}),
|
||||
'v' if !posix => Ok(CommandSpec {
|
||||
n_addr: 0,
|
||||
handler: compile_version_command,
|
||||
}),
|
||||
_ => compilation_error(lines, line, format!("invalid command code `{cmd_code}'")),
|
||||
}
|
||||
}
|
||||
@@ -3055,31 +2943,6 @@ mod tests {
|
||||
assert!(err.to_string().contains(ERR_SANDBOX));
|
||||
}
|
||||
|
||||
// compile_read_line_command (R)
|
||||
#[test]
|
||||
fn test_compile_read_line_command_rejected_under_sandbox() {
|
||||
let (mut lines, mut chars) = make_providers("R input.txt");
|
||||
let mut cmd = Command::default();
|
||||
let mut context = ctx();
|
||||
context.sandbox = true;
|
||||
|
||||
let err =
|
||||
compile_read_line_command(&mut lines, &mut chars, &mut cmd, &mut context).unwrap_err();
|
||||
assert!(err.to_string().contains(ERR_SANDBOX));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_compile_read_line_command_sets_named_reader() {
|
||||
let (mut lines, mut chars) = make_providers("R input.txt");
|
||||
let mut cmd = Command::default();
|
||||
let mut context = ctx();
|
||||
|
||||
let handling =
|
||||
compile_read_line_command(&mut lines, &mut chars, &mut cmd, &mut context).unwrap();
|
||||
assert!(matches!(handling, CommandHandling::Continue));
|
||||
assert!(matches!(cmd.data, CommandData::NamedReader(_)));
|
||||
}
|
||||
|
||||
// compile_write_file_command
|
||||
#[test]
|
||||
fn test_compile_write_file_command_rejected_under_sandbox() {
|
||||
|
||||
@@ -14,7 +14,6 @@ use crate::sed::script_line_provider::ScriptLineProvider;
|
||||
|
||||
use std::rc::Rc;
|
||||
|
||||
use uucore::display::Quotable;
|
||||
use uucore::error::{UResult, USimpleError};
|
||||
|
||||
#[derive(Clone, Debug)]
|
||||
@@ -109,7 +108,7 @@ pub fn input_runtime_error<T>(
|
||||
location.input_name,
|
||||
location.line_number,
|
||||
location.column_number,
|
||||
context.input_name.quote(),
|
||||
context.input_name,
|
||||
context.line_number,
|
||||
msg.to_string()
|
||||
),
|
||||
|
||||
+46
-156
@@ -29,11 +29,6 @@ use std::marker::PhantomData;
|
||||
#[cfg(unix)]
|
||||
use std::os::fd::RawFd;
|
||||
|
||||
#[cfg(unix)]
|
||||
use rustix::fd::BorrowedFd;
|
||||
#[cfg(target_os = "linux")]
|
||||
use rustix::fs::copy_file_range as rustix_copy_file_range;
|
||||
|
||||
#[cfg(unix)]
|
||||
use std::os::unix::io::AsRawFd;
|
||||
|
||||
@@ -407,24 +402,9 @@ impl FastCopy {
|
||||
Self {
|
||||
fd,
|
||||
is_regular: ftype as libc::mode_t == libc::S_IFREG,
|
||||
// st_blksize is always positive and small; fall back to a common
|
||||
// block size if it ever doesn't fit in usize.
|
||||
block_size: usize::try_from(st.st_blksize).unwrap_or(4096),
|
||||
block_size: st.st_blksize as usize,
|
||||
}
|
||||
}
|
||||
|
||||
/// Return a borrowed file descriptor.
|
||||
///
|
||||
/// # Safety invariant
|
||||
///
|
||||
/// `FastCopy` only stores file descriptors obtained from open
|
||||
/// [`File`](std::fs::File) objects. Those `File`s are kept alive
|
||||
/// by the owning [`MmapLineCursor`] / output-buffer, guaranteeing
|
||||
/// the fd remains valid for the duration of any borrow.
|
||||
pub fn as_fd(&self) -> BorrowedFd<'_> {
|
||||
// SAFETY: self.fd is a valid file descriptor owned by a live File.
|
||||
unsafe { BorrowedFd::borrow_raw(self.fd) }
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
@@ -818,10 +798,10 @@ impl OutputBuffer {
|
||||
if chunk.in_fast_copy.is_regular && self.fast_copy.is_regular {
|
||||
portable_copy_file_range(
|
||||
chunk.out_ptr,
|
||||
chunk.in_fast_copy.as_fd(),
|
||||
chunk.in_fast_copy.fd,
|
||||
// Input file offset
|
||||
unsafe { chunk.out_ptr.offset_from(chunk.base_ptr) } as libc::off_t,
|
||||
self.fast_copy.as_fd(),
|
||||
self.fast_copy.fd,
|
||||
chunk.len,
|
||||
// Alignment block size: the largest of the two
|
||||
chunk.in_fast_copy.block_size.max(self.fast_copy.block_size),
|
||||
@@ -936,9 +916,9 @@ fn reliable_write(fd: i32, ptr: *const u8, len: usize) -> std::io::Result<usize>
|
||||
#[allow(unused_variables)]
|
||||
fn portable_copy_file_range(
|
||||
in_ptr: *const u8,
|
||||
in_fd: BorrowedFd<'_>,
|
||||
in_fd: i32,
|
||||
in_off: libc::off_t,
|
||||
out_fd: BorrowedFd<'_>,
|
||||
out_fd: i32,
|
||||
len: usize,
|
||||
block_size: usize,
|
||||
cover: WriteRange,
|
||||
@@ -953,7 +933,7 @@ fn portable_copy_file_range(
|
||||
}
|
||||
#[cfg(not(all(target_os = "linux", target_env = "gnu")))]
|
||||
{
|
||||
reliable_write(out_fd.as_raw_fd(), in_ptr, len)
|
||||
reliable_write(out_fd, in_ptr, len)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -964,34 +944,37 @@ fn portable_copy_file_range(
|
||||
#[cfg(all(target_os = "linux", target_env = "gnu"))]
|
||||
fn reliable_copy_file_range(
|
||||
in_ptr: *const u8,
|
||||
in_fd: BorrowedFd<'_>,
|
||||
in_fd: i32,
|
||||
mut in_off: libc::off_t,
|
||||
out_fd: BorrowedFd<'_>,
|
||||
out_fd: i32,
|
||||
len: usize,
|
||||
) -> std::io::Result<usize> {
|
||||
let mut pending = len;
|
||||
while pending > 0 {
|
||||
let mut in_off_u64 = in_off as u64;
|
||||
let result: std::io::Result<usize> =
|
||||
rustix_copy_file_range(in_fd, Some(&mut in_off_u64), out_fd, None, pending)
|
||||
.map_err(std::io::Error::from);
|
||||
|
||||
match result {
|
||||
Ok(0) => break,
|
||||
Ok(ret) => {
|
||||
pending -= ret;
|
||||
in_off = in_off_u64 as libc::off_t;
|
||||
}
|
||||
Err(err) => {
|
||||
return match err.raw_os_error() {
|
||||
Some(libc::ENOSYS) | Some(libc::EOPNOTSUPP) | Some(libc::EXDEV) => {
|
||||
// Fallback to write(2).
|
||||
reliable_write(out_fd.as_raw_fd(), in_ptr, pending)
|
||||
}
|
||||
_ => Err(err),
|
||||
};
|
||||
}
|
||||
let ret = unsafe {
|
||||
libc::copy_file_range(
|
||||
in_fd,
|
||||
&raw mut in_off,
|
||||
out_fd,
|
||||
std::ptr::null_mut(), // Use and update output offset
|
||||
pending,
|
||||
0,
|
||||
)
|
||||
};
|
||||
if ret < 0 {
|
||||
let err = io::Error::last_os_error();
|
||||
return match err.raw_os_error() {
|
||||
Some(libc::ENOSYS) | Some(libc::EOPNOTSUPP) | Some(libc::EXDEV) => {
|
||||
// Fallback to write(2).
|
||||
reliable_write(out_fd, in_ptr, pending)
|
||||
}
|
||||
_ => Err(err),
|
||||
};
|
||||
} else if ret == 0 {
|
||||
// EOF reached
|
||||
break;
|
||||
}
|
||||
pending -= ret as usize;
|
||||
}
|
||||
Ok(len)
|
||||
}
|
||||
@@ -1003,27 +986,24 @@ fn reliable_copy_file_range(
|
||||
#[cfg(all(target_os = "linux", target_env = "gnu"))]
|
||||
fn aligned_copy_file_range(
|
||||
mut in_ptr: *const u8,
|
||||
in_fd: BorrowedFd<'_>,
|
||||
in_fd: i32,
|
||||
mut in_off: libc::off_t,
|
||||
out_fd: BorrowedFd<'_>,
|
||||
out_fd: i32,
|
||||
len: usize,
|
||||
block_size: usize,
|
||||
cover: WriteRange,
|
||||
) -> std::io::Result<usize> {
|
||||
// Get current output offset.
|
||||
// If either offset doesn't fit in usize (only possible on 32-bit
|
||||
// targets), give up on alignment and just copy everything.
|
||||
let (Ok(out_off), Ok(in_off_bytes)) = (
|
||||
usize::try_from(rustix::fs::tell(out_fd)?),
|
||||
usize::try_from(in_off),
|
||||
) else {
|
||||
return reliable_copy_file_range(in_ptr, in_fd, in_off, out_fd, len);
|
||||
};
|
||||
// 1. Get current output offset.
|
||||
let res = unsafe { libc::lseek(out_fd, 0, libc::SEEK_CUR) as i64 };
|
||||
if res < 0 {
|
||||
return Err(std::io::Error::last_os_error());
|
||||
}
|
||||
let out_off = res as usize;
|
||||
let mut pending = len;
|
||||
|
||||
// Obtain head alignment.
|
||||
// Bytes to end of block:
|
||||
let remainder = in_off_bytes % block_size;
|
||||
let remainder = in_off as usize % block_size;
|
||||
// Bytes to write (block_size becomes 0):
|
||||
let head_align = (block_size - remainder) % block_size;
|
||||
|
||||
@@ -1035,7 +1015,7 @@ fn aligned_copy_file_range(
|
||||
if head_align > 0 {
|
||||
// Align the two files on a block boundary.
|
||||
let head_len = head_align.min(pending);
|
||||
reliable_write(out_fd.as_raw_fd(), in_ptr, head_len)?;
|
||||
reliable_write(out_fd, in_ptr, head_len)?;
|
||||
in_ptr = unsafe { in_ptr.add(head_len) };
|
||||
in_off += head_len as i64;
|
||||
pending -= head_len;
|
||||
@@ -1049,7 +1029,7 @@ fn aligned_copy_file_range(
|
||||
// Copy tail if needed.
|
||||
if pending > 0 && cover == WriteRange::Complete {
|
||||
in_ptr = unsafe { in_ptr.add(aligned_len) };
|
||||
pending -= reliable_write(out_fd.as_raw_fd(), in_ptr, pending)?;
|
||||
pending -= reliable_write(out_fd, in_ptr, pending)?;
|
||||
}
|
||||
|
||||
Ok(len - pending)
|
||||
@@ -1082,8 +1062,6 @@ mod tests {
|
||||
#[cfg(all(target_os = "linux", target_env = "gnu"))]
|
||||
use std::io::{self, Write};
|
||||
use std::io::{Seek, SeekFrom};
|
||||
#[cfg(target_os = "linux")]
|
||||
use std::os::unix::io::AsFd;
|
||||
use tempfile::NamedTempFile;
|
||||
use tempfile::tempfile;
|
||||
|
||||
@@ -1815,8 +1793,8 @@ mod tests {
|
||||
infile.write_all(data).unwrap();
|
||||
infile.rewind().unwrap();
|
||||
|
||||
let in_fd = infile.as_fd();
|
||||
let out_fd = outfile.as_fd();
|
||||
let in_fd = infile.as_raw_fd();
|
||||
let out_fd = outfile.as_raw_fd();
|
||||
let in_ptr = data.as_ptr();
|
||||
|
||||
// Copy with block size 4, cover = Complete
|
||||
@@ -1851,8 +1829,8 @@ mod tests {
|
||||
infile.write_all(data).unwrap();
|
||||
infile.rewind().unwrap();
|
||||
|
||||
let in_fd = infile.as_fd();
|
||||
let out_fd = outfile.as_fd();
|
||||
let in_fd = infile.as_raw_fd();
|
||||
let out_fd = outfile.as_raw_fd();
|
||||
let in_ptr = data.as_ptr();
|
||||
|
||||
// Copy with block size 4, cover = Blocks
|
||||
@@ -1870,94 +1848,6 @@ mod tests {
|
||||
assert_eq!(&buf, b"abcdefgh");
|
||||
}
|
||||
|
||||
#[test]
|
||||
#[cfg(target_os = "linux")]
|
||||
fn test_aligned_copy_respects_existing_out_offset() {
|
||||
let mut infile = tempfile().unwrap();
|
||||
let mut outfile = tempfile().unwrap();
|
||||
|
||||
// Write a prefix into the output file so tell() returns a non-zero offset.
|
||||
let prefix = b"HEADER";
|
||||
outfile.write_all(prefix).unwrap();
|
||||
outfile.flush().unwrap();
|
||||
|
||||
// Input data = 12 bytes (3 blocks of 4 when block_size=4).
|
||||
let data = b"abcdefghijkl";
|
||||
infile.write_all(data).unwrap();
|
||||
infile.rewind().unwrap();
|
||||
|
||||
let in_fd = infile.as_fd();
|
||||
let out_fd = outfile.as_fd();
|
||||
let in_ptr = data.as_ptr();
|
||||
|
||||
let copied = aligned_copy_file_range(
|
||||
in_ptr,
|
||||
in_fd,
|
||||
0,
|
||||
out_fd,
|
||||
data.len(),
|
||||
4,
|
||||
WriteRange::Complete,
|
||||
)
|
||||
.unwrap();
|
||||
|
||||
assert_eq!(copied, data.len());
|
||||
|
||||
outfile.rewind().unwrap();
|
||||
let mut buf = Vec::new();
|
||||
outfile.read_to_end(&mut buf).unwrap();
|
||||
|
||||
assert_eq!(&buf, b"HEADERabcdefghijkl");
|
||||
}
|
||||
|
||||
#[test]
|
||||
#[cfg(target_os = "linux")]
|
||||
fn test_reliable_copy_file_range_rustix_path() {
|
||||
let mut infile = tempfile().unwrap();
|
||||
let mut outfile = tempfile().unwrap();
|
||||
|
||||
let data = b"0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
|
||||
infile.write_all(data).unwrap();
|
||||
infile.rewind().unwrap();
|
||||
|
||||
let in_fd = infile.as_fd();
|
||||
let out_fd = outfile.as_fd();
|
||||
let in_ptr = data.as_ptr();
|
||||
|
||||
// Copy a subrange starting at offset 10.
|
||||
let copied = reliable_copy_file_range(in_ptr, in_fd, 10, out_fd, 16).unwrap();
|
||||
|
||||
assert_eq!(copied, 16);
|
||||
|
||||
outfile.rewind().unwrap();
|
||||
let mut buf = Vec::new();
|
||||
outfile.read_to_end(&mut buf).unwrap();
|
||||
assert_eq!(&buf, b"ABCDEFGHIJKLMNOP");
|
||||
|
||||
// Now verify that copying into a file with existing data appends correctly.
|
||||
let mut infile2 = tempfile().unwrap();
|
||||
let mut outfile2 = tempfile().unwrap();
|
||||
let input_data = b"abcdefghijklmnopqrstuvwxyz";
|
||||
infile2.write_all(input_data).unwrap();
|
||||
infile2.rewind().unwrap();
|
||||
|
||||
outfile2.write_all(b"PRE:").unwrap();
|
||||
outfile2.flush().unwrap();
|
||||
|
||||
let in_fd2 = infile2.as_fd();
|
||||
let out_fd2 = outfile2.as_fd();
|
||||
let in_ptr2 = input_data.as_ptr();
|
||||
|
||||
let copied2 = reliable_copy_file_range(in_ptr2, in_fd2, 5, out_fd2, 10).unwrap();
|
||||
|
||||
assert_eq!(copied2, 10);
|
||||
|
||||
outfile2.rewind().unwrap();
|
||||
let mut buf2 = Vec::new();
|
||||
outfile2.read_to_end(&mut buf2).unwrap();
|
||||
assert_eq!(&buf2, b"PRE:fghijklmno");
|
||||
}
|
||||
|
||||
///////////////////////////////
|
||||
// Unit tests for write_chunk()
|
||||
///////////////////////////////
|
||||
|
||||
+5
-17
@@ -15,7 +15,6 @@ pub mod error_handling;
|
||||
pub mod fast_io;
|
||||
pub mod fast_regex;
|
||||
pub mod in_place;
|
||||
pub mod named_reader;
|
||||
pub mod named_writer;
|
||||
pub mod processor;
|
||||
pub mod script_char_provider;
|
||||
@@ -25,16 +24,15 @@ use crate::sed::command::{ByteSpace, CharacterMode, ProcessingContext};
|
||||
use crate::sed::compiler::compile;
|
||||
use crate::sed::processor::process_all_files;
|
||||
use crate::sed::script_line_provider::ScriptValue;
|
||||
use clap::{Arg, ArgMatches, Command, arg};
|
||||
use clap::{Arg, ArgMatches, Command, arg, crate_version};
|
||||
use std::collections::HashMap;
|
||||
use std::env;
|
||||
use std::path::PathBuf;
|
||||
use uucore::error::{UResult, USimpleError, UUsageError};
|
||||
use uucore::format_usage;
|
||||
|
||||
const ABOUT: &str = "Stream editor for filtering and transforming text (part of uutils)";
|
||||
const ABOUT: &str = "Stream editor for filtering and transforming text";
|
||||
const USAGE: &str = "sed [OPTION]... [script] [file]...";
|
||||
const VERSION: &str = concat!(env!("CARGO_PKG_VERSION"), " (uutils)");
|
||||
|
||||
#[uucore::main]
|
||||
pub fn uumain(args: impl uucore::Args) -> UResult<()> {
|
||||
@@ -64,7 +62,7 @@ pub fn uu_app() -> Command {
|
||||
let util_name = uucore::util_name();
|
||||
|
||||
Command::new(util_name)
|
||||
.version(VERSION)
|
||||
.version(crate_version!())
|
||||
.about(ABOUT)
|
||||
.override_usage(format_usage(USAGE))
|
||||
.args_override_self(true)
|
||||
@@ -115,11 +113,6 @@ pub fn uu_app() -> Command {
|
||||
arg!(-s --separate "Consider files as separate rather than as a long stream."),
|
||||
arg!(--sandbox "Operate in a sandbox by disabling e/r/w commands."),
|
||||
arg!(-u --unbuffered "Load minimal input data and flush output buffers regularly."),
|
||||
Arg::new("uutil-extensions")
|
||||
.short('U')
|
||||
.long("uutil-extensions")
|
||||
.help("Enable incompatible extensions.")
|
||||
.action(clap::ArgAction::SetTrue),
|
||||
Arg::new("null-data")
|
||||
.short('z')
|
||||
.long("null-data")
|
||||
@@ -245,17 +238,16 @@ fn build_context(matches: &ArgMatches) -> UResult<ProcessingContext> {
|
||||
length: matches.get_one::<u32>("length").map_or(70, |v| *v as usize),
|
||||
quiet: matches.get_flag("quiet"),
|
||||
posix: matches.get_flag("posix"),
|
||||
separate: matches.get_flag("separate") || matches.contains_id("in-place"),
|
||||
separate: matches.get_flag("separate"),
|
||||
sandbox: matches.get_flag("sandbox"),
|
||||
unbuffered: matches.get_flag("unbuffered"),
|
||||
null_data: matches.get_flag("null-data"),
|
||||
uutil_extensions: matches.get_flag("uutil-extensions"),
|
||||
|
||||
// Environment
|
||||
character_mode: character_mode_for_locale(&locale)?,
|
||||
|
||||
// Other context
|
||||
input_name: PathBuf::from("-"),
|
||||
input_name: "<stdin>".to_string(),
|
||||
line_number: 0,
|
||||
last_address: false,
|
||||
last_line: false,
|
||||
@@ -269,7 +261,6 @@ fn build_context(matches: &ArgMatches) -> UResult<ProcessingContext> {
|
||||
},
|
||||
parsed_block_nesting: 0,
|
||||
label_to_command_map: HashMap::new(),
|
||||
named_readers: HashMap::new(),
|
||||
range_commands: Vec::new(),
|
||||
substitution_made: false,
|
||||
append_elements: Vec::new(),
|
||||
@@ -397,7 +388,6 @@ mod tests {
|
||||
assert!(!ctx.sandbox);
|
||||
assert!(!ctx.unbuffered);
|
||||
assert!(!ctx.null_data);
|
||||
assert!(!ctx.uutil_extensions);
|
||||
}
|
||||
|
||||
#[test]
|
||||
@@ -415,7 +405,6 @@ mod tests {
|
||||
"-s",
|
||||
"--sandbox",
|
||||
"-u",
|
||||
"-U",
|
||||
"-z",
|
||||
]);
|
||||
|
||||
@@ -434,7 +423,6 @@ mod tests {
|
||||
assert!(ctx.sandbox);
|
||||
assert!(ctx.unbuffered);
|
||||
assert!(ctx.null_data);
|
||||
assert!(ctx.uutil_extensions);
|
||||
}
|
||||
|
||||
#[test]
|
||||
|
||||
@@ -1,101 +0,0 @@
|
||||
// An abstraction for input files read one line at a time by the `R` command
|
||||
//
|
||||
// SPDX-License-Identifier: MIT
|
||||
// Copyright (c) 2025 Diomidis Spinellis
|
||||
//
|
||||
// This file is part of the uutils sed package.
|
||||
// It is licensed under the MIT License.
|
||||
// For the full copyright and license information, please view the LICENSE
|
||||
// file that was distributed with this source code.
|
||||
|
||||
use std::cell::RefCell;
|
||||
use std::fs::File;
|
||||
use std::io::{BufRead, BufReader};
|
||||
use std::path::PathBuf;
|
||||
use std::rc::Rc;
|
||||
|
||||
#[derive(Debug)]
|
||||
/// State of the file backing an `R` command, opened lazily on first use.
|
||||
enum State {
|
||||
Unopened,
|
||||
Open(BufReader<File>),
|
||||
Exhausted,
|
||||
}
|
||||
|
||||
#[derive(Debug)]
|
||||
/// Reader that yields successive lines of a file for the GNU `R` command.
|
||||
/// The file is opened on first use; a file that cannot be opened or read is
|
||||
/// treated as having no more lines, matching GNU sed (no error is raised).
|
||||
pub struct NamedReader {
|
||||
path: PathBuf,
|
||||
state: State,
|
||||
}
|
||||
|
||||
impl NamedReader {
|
||||
/// Create a reader for `path` without opening it yet.
|
||||
pub fn new(path: PathBuf) -> Rc<RefCell<Self>> {
|
||||
Rc::new(RefCell::new(NamedReader {
|
||||
path,
|
||||
state: State::Unopened,
|
||||
}))
|
||||
}
|
||||
|
||||
/// Return the next line of the file, including its trailing newline if
|
||||
/// present, or `None` once the file is exhausted or could not be read.
|
||||
pub fn next_line(&mut self) -> Option<Vec<u8>> {
|
||||
if matches!(self.state, State::Unopened) {
|
||||
self.state = match File::open(&self.path) {
|
||||
Ok(file) => State::Open(BufReader::new(file)),
|
||||
Err(_) => State::Exhausted,
|
||||
};
|
||||
}
|
||||
|
||||
let State::Open(reader) = &mut self.state else {
|
||||
return None;
|
||||
};
|
||||
|
||||
let mut line = Vec::new();
|
||||
match reader.read_until(b'\n', &mut line) {
|
||||
Ok(0) | Err(_) => {
|
||||
self.state = State::Exhausted;
|
||||
None
|
||||
}
|
||||
Ok(_) => Some(line),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::*;
|
||||
use std::io::Write;
|
||||
use tempfile::NamedTempFile;
|
||||
|
||||
#[test]
|
||||
fn yields_successive_lines_then_none() {
|
||||
let mut file = NamedTempFile::new().unwrap();
|
||||
file.write_all(b"one\ntwo\n").unwrap();
|
||||
let reader = NamedReader::new(file.path().to_path_buf());
|
||||
|
||||
assert_eq!(reader.borrow_mut().next_line(), Some(b"one\n".to_vec()));
|
||||
assert_eq!(reader.borrow_mut().next_line(), Some(b"two\n".to_vec()));
|
||||
assert_eq!(reader.borrow_mut().next_line(), None);
|
||||
assert_eq!(reader.borrow_mut().next_line(), None);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn last_line_without_newline_is_preserved() {
|
||||
let mut file = NamedTempFile::new().unwrap();
|
||||
file.write_all(b"abc").unwrap();
|
||||
let reader = NamedReader::new(file.path().to_path_buf());
|
||||
|
||||
assert_eq!(reader.borrow_mut().next_line(), Some(b"abc".to_vec()));
|
||||
assert_eq!(reader.borrow_mut().next_line(), None);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn missing_file_yields_no_lines() {
|
||||
let reader = NamedReader::new(PathBuf::from("/nonexistent/xyzzy-42-does-not-exist"));
|
||||
assert_eq!(reader.borrow_mut().next_line(), None);
|
||||
}
|
||||
}
|
||||
Loaded 20 of 29 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user