Compare commits

...

6 commits

Author SHA1 Message Date
tobi 4fcea3d6ae
[chore] Hopefully final tweak to changelog generation (#3437) 2024-10-14 16:02:16 +02:00
tobi 400fd8c62b
[docs] Document experimental, unsupported nowasm tag (#3436)
* [docs] Document experimental, unsupported `nowasm` tag

* change some awkward wording
2024-10-14 14:06:19 +02:00
kim 6a76b9d609
[feature/OFFICIALLY UNSUPPORTED] add nowasm build tag to disable building with WebAssembly (#3429)
* add experimental build-tag 'nowasm' which uses local ffmpeg / ffprobe

* updated experimental support message

* add comment to build script explaining build tag

* add nowasm build tags to moderncsqlite files
2024-10-14 11:59:12 +02:00
dependabot[bot] 157ee3193d
[chore]: Bump github.com/minio/minio-go/v7 from 7.0.77 to 7.0.78 (#3431)
Bumps [github.com/minio/minio-go/v7](https://github.com/minio/minio-go) from 7.0.77 to 7.0.78.
- [Release notes](https://github.com/minio/minio-go/releases)
- [Commits](https://github.com/minio/minio-go/compare/v7.0.77...v7.0.78)

---
updated-dependencies:
- dependency-name: github.com/minio/minio-go/v7
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-10-14 11:46:02 +02:00
dependabot[bot] be3b8076ca
[chore]: Bump github.com/yuin/goldmark from 1.7.4 to 1.7.6 (#3430)
Bumps [github.com/yuin/goldmark](https://github.com/yuin/goldmark) from 1.7.4 to 1.7.6.
- [Release notes](https://github.com/yuin/goldmark/releases)
- [Commits](https://github.com/yuin/goldmark/compare/v1.7.4...v1.7.6)

---
updated-dependencies:
- dependency-name: github.com/yuin/goldmark
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-10-14 11:43:44 +02:00
tobi 020da28daa
[chore] Clarify supported platforms, add notes + docs (#3427)
* [chore] Don't build for OpenBSD, add support level notes + tips

* tweak a bit

* add supported platforms matrix
2024-10-14 11:41:55 +02:00
56 changed files with 16024 additions and 15324 deletions

View file

@ -39,7 +39,6 @@ builds:
goos: goos:
- linux - linux
- freebsd - freebsd
- openbsd
goarch: goarch:
- 386 - 386
- amd64 - amd64
@ -49,19 +48,14 @@ builds:
- 6 - 6
- 7 - 7
ignore: ignore:
# build freebsd + openbsd only for amd64 # Build FreeBSD
# only for amd64.
- goos: freebsd - goos: freebsd
goarch: arm64 goarch: arm64
- goos: freebsd - goos: freebsd
goarch: arm goarch: arm
- goos: freebsd - goos: freebsd
goarch: 386 goarch: 386
- goos: openbsd
goarch: arm64
- goos: openbsd
goarch: arm
- goos: openbsd
goarch: 386
mod_timestamp: "{{ .CommitTimestamp }}" mod_timestamp: "{{ .CommitTimestamp }}"
# MODERNC SQLITE BINARY BUILDS # MODERNC SQLITE BINARY BUILDS
- -
@ -88,7 +82,6 @@ builds:
goos: goos:
- linux - linux
- freebsd - freebsd
- openbsd
goarch: goarch:
- 386 - 386
- amd64 - amd64
@ -98,19 +91,14 @@ builds:
- 6 - 6
- 7 - 7
ignore: ignore:
# build freebsd + openbsd only for amd64 # Build FreeBSD
# only for amd64.
- goos: freebsd - goos: freebsd
goarch: arm64 goarch: arm64
- goos: freebsd - goos: freebsd
goarch: arm goarch: arm
- goos: freebsd - goos: freebsd
goarch: 386 goarch: 386
- goos: openbsd
goarch: arm64
- goos: openbsd
goarch: arm
- goos: openbsd
goarch: 386
mod_timestamp: "{{ .CommitTimestamp }}" mod_timestamp: "{{ .CommitTimestamp }}"
# https://goreleaser.com/customization/docker/ # https://goreleaser.com/customization/docker/
@ -413,7 +401,7 @@ release:
header: | header: |
Here's version {{ .Version }} of GoToSocial. Here's version {{ .Version }} of GoToSocial.
Please read the [migration notes](#migration-notes) carefully for instructions on how to upgrade to this version. Please read the migration notes carefully for instructions on how to upgrade to this version.
## Release highlights ## Release highlights
@ -474,35 +462,58 @@ release:
Tl;dr: Regardless of whether you're using SQLite or Postgres as your DB driver, you most likely you want the regular version without `moderncsqlite` in the name. Tl;dr: Regardless of whether you're using SQLite or Postgres as your DB driver, you most likely you want the regular version without `moderncsqlite` in the name.
However, if you're on FreeBSD or OpenBSD, use the `moderncsqlite` version instead. See the table below: However, if you're on FreeBSD, 32-bit Linux or 32-bit ARM, we recommend using the `moderncsqlite` version instead.
| OS | Architecture | Binary archive | Docker | You may need to change some configuration options too. See the table below:
| ------- | ----------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- |
| Linux | x86-64/AMD64 (64-bit) | [linux_amd64.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_amd64.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` | | OS | Architecture | Support level | Binary archive | Docker |
| Linux | x86-32/i386 (32-bit) | [linux_386.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_386.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` | | ------- | ----------------------- | ---------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- |
| Linux | Armv8/ARM64 (64-bit) | [linux_arm64.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_arm64.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` | | Linux | x86-64/AMD64 (64-bit) | 🟢 Full | [linux_amd64.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_amd64.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` |
| Linux | Armv7/ARM32 (32-bit) | [linux_armv7.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_armv7.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` | | Linux | Armv8/ARM64 (64-bit) | 🟢 Full | [linux_arm64.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_arm64.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` |
| Linux | Armv6/ARM32 (32-bit) | [linux_armv6.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_armv6.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}` | | FreeBSD | x86-64/AMD64 (64-bit) | 🟢 Full<sup>[1](#freebsd)</sup> | [freebsd_amd64_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_freebsd_amd64_moderncsqlite.tar.gz) | None provided |
| OpenBSD | x86-64/AMD64 (64-bit) | [openbsd_amd64_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_openbsd_amd64_moderncsqlite.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}-moderncsqlite` | | Linux | x86-32/i386 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | [linux_386_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_386_moderncsqlite.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}-moderncsqlite` |
| FreeBSD | x86-64/AMD64 (64-bit) | [freebsd_amd64_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_freebsd_amd64_moderncsqlite.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}-moderncsqlite` | | Linux | Armv7/ARM32 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | [linux_armv7_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_armv7_moderncsqlite.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}-moderncsqlite` |
| Linux | Armv6/ARM32 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | [linux_armv6_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_armv7_moderncsqlite.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}-moderncsqlite` |
| Linux | x86-32/i386 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | [linux_386_moderncsqlite.tar.gz](https://github.com/superseriousbusiness/gotosocial/releases/download/{{ .Tag }}/gotosocial_{{ .Version }}_linux_386_moderncsqlite.tar.gz) | `superseriousbusiness/gotosocial:{{ .Version }}-moderncsqlite` |
#### FreeBSD
`moderncsqlite` version currently recommended, though you might have success with the regular WASM SQLite version.
If running with regular WASM SQLite and having instability or memory issues, the following settings *may* help:
```yaml
db-max-open-conns-multiplier: 0
db-sqlite-journal-mode: "TRUNCATE"
db-sqlite-synchronous: "FULL"
```
#### 32-bit
`moderncsqlite` version is needed, as performance with regular WASM SQLite is not guaranteed when running on 32-bit.
Remote media processing will likely not work with reasonable performance, so you should set the following config variables to prevent download of remote media onto your instance:
```yaml
media-remote-max-size: 0
media-emoji-remote-max-size: 0
```
# https://goreleaser.com/customization/changelog/ # https://goreleaser.com/customization/changelog/
changelog: changelog:
use: github use: github
abbrev: 9
format: "{{ .Message }}, by {{ with .AuthorUsername }}@{{ . }}{{ else }}{{ .AuthorName }} <{{ .AuthorEmail }}>{{ end }} ({{ .SHA }})"
groups: groups:
- title: Features and performance - title: Features and performance
regexp: '^\[(?:feature|performance).*\].*$' regexp: '^.*\[(?:feature|performance).*\].*$'
order: 0 order: 0
- title: Bug fixes - title: Bug fixes
regexp: '^\[bug.*\].*$' regexp: '^.*\[bug.*\].*$'
order: 1 order: 1
- title: Chores & version bumps - title: Chores & version bumps
regexp: '^\[chore.*\].*$' regexp: '^.*\[chore.*\].*$'
order: 2 order: 2
- title: Documentation - title: Documentation
regexp: '^\[doc.*\].*$' regexp: '^.*\[doc.*\].*$'
order: 3 order: 3
- title: Other - title: Other
order: 999 order: 999

View file

@ -41,7 +41,11 @@ Here's a screenshot of the instance landing page!
- [OIDC integration](#oidc-integration) - [OIDC integration](#oidc-integration)
- [Backend-first design](#backend-first-design) - [Backend-first design](#backend-first-design)
- [Known Issues](#known-issues) - [Known Issues](#known-issues)
- [Getting Started](#getting-started) - [Installing GoToSocial](#installing-gotosocial)
- [Supported Platforms](#supported-platforms)
- [FreeBSD](#freebsd)
- [32-bit](#32-bit)
- [OpenBSD](#openbsd)
- [Stable Releases](#stable-releases) - [Stable Releases](#stable-releases)
- [Snapshot Releases](#snapshot-releases) - [Snapshot Releases](#snapshot-releases)
- [Docker](#docker) - [Docker](#docker)
@ -258,9 +262,47 @@ Since every ActivityPub server implementation has a slightly different interpret
--- ---
## Getting Started ## Installing GoToSocial
Check our [releases page](https://github.com/superseriousbusiness/gotosocial/releases) and our [getting started](https://docs.gotosocial.org/en/latest/getting_started/) documentation. Check our [getting started](https://docs.gotosocial.org/en/latest/getting_started/) documentation! And have a peruse of our [releases page](https://github.com/superseriousbusiness/gotosocial/releases).
<!--releases-start-->
### Supported Platforms
While we try to support a reasonable number of architectures and operating systems, it's not always possible to support a given platform due to library constraints or performance issues.
Platforms that we don't officially support *may* still work, but we can't test or guarantee performance or stability.
This is the current status of support offered by GoToSocial for different platforms (if something is unlisted it means we haven't checked yet so we don't know):
| OS | Architecture | Support level | Binary archive | Docker container |
| ------- | ----------------------- | ---------------------------------- | -------------- | ---------------- |
| Linux | x86-64/AMD64 (64-bit) | 🟢 Full | Yes | Yes |
| Linux | Armv8/ARM64 (64-bit) | 🟢 Full | Yes | Yes |
| FreeBSD | x86-64/AMD64 (64-bit) | 🟢 Full<sup>[1](#freebsd)</sup> | Yes | No |
| Linux | x86-32/i386 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | Yes | Yes |
| Linux | Armv7/ARM32 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | Yes | Yes |
| Linux | Armv6/ARM32 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | Yes | Yes |
| Linux | x86-32/i386 (32-bit) | 🟡 Partial<sup>[2](#32-bit)</sup> | Yes | Yes |
| OpenBSD | Any | 🔴 None<sup>[3](#openbsd)</sup> | No | No |
#### FreeBSD
Mostly works, just a few issues with WASM SQLite; check release notes carefully when installing on FreeBSD. If running with Postgres you should have no issues.
#### 32-bit
GtS doesn't work well on 32-bit systems like i386, or Armv6/v7, mainly due to performance of media decoding.
We don't recommend running GtS on 32-bit, but you may have some success either turning off remote media processing altogether, or building a binary yourself with the totally **unsupported, experimental** [nowasm](https://docs.gotosocial.org/en/latest/advanced/builds/nowasm/) tag.
For more guidance, check release notes when trying to install on 32-bit.
#### OpenBSD
Marked as unsupported due to performance issues (high memory usage when idle, crashes while processing media).
While we don't support running GtS on OpenBSD, you may have some success building a binary yourself with the totally **unsupported, experimental** [nowasm](https://docs.gotosocial.org/en/latest/advanced/builds/nowasm/) tag.
### Stable Releases ### Stable Releases
@ -302,6 +344,7 @@ You can also deploy your own instance of GoToSocial with the help of:
- [Ansible Playbook (MASH)](https://github.com/mother-of-all-self-hosting/mash-playbook): The playbook supports a many services, including GoToSocial. [Documentation](https://github.com/mother-of-all-self-hosting/mash-playbook/blob/main/docs/services/gotosocial.md) - [Ansible Playbook (MASH)](https://github.com/mother-of-all-self-hosting/mash-playbook): The playbook supports a many services, including GoToSocial. [Documentation](https://github.com/mother-of-all-self-hosting/mash-playbook/blob/main/docs/services/gotosocial.md)
- [GoToSocial Helm Chart](https://github.com/fSocietySocial/charts/tree/main/charts/gotosocial) by [0hlov3](https://github.com/0hlov3). - [GoToSocial Helm Chart](https://github.com/fSocietySocial/charts/tree/main/charts/gotosocial) by [0hlov3](https://github.com/0hlov3).
<!--releases-end-->
--- ---
## Contributing ## Contributing

View file

@ -0,0 +1,27 @@
# Build without Wazero / WASM
!!! Danger "This is unsupported"
We do not offer any kind of support for deployments of GoToSocial built with the `nowasm` tag described in this section. Such builds should be considered strictly experimental, and any issues that come when running them are none of our business! Please don't open issues on the repo looking for help debugging deployments of `nowasm` builds.
On [supported platforms](../../getting_started/releases.md#supported-platforms), GoToSocial uses the WebAssembly runtime [Wazero](https://wazero.io/) to sandbox `ffmpeg`, `ffprobe`, and `sqlite3` WebAssembly binaries, allowing these applications to be packaged and run inside the GoToSocial binary, without requiring admins to install + manage any external dependencies.
This has the advantage of making it easier for admins to maintain their GoToSocial instance, as their GtS binary is completely isolated from any changes to their system-installed `ffmpeg`, `ffprobe`, and `sqlite`. It's also a bit safer to run `ffmpeg` in this way, as GoToSocial wraps the `ffmpeg` binary in a very constrained file system that doesn't permit the `ffmpeg` binary to access any files other than the ones it's decoding + reencoding. In other words, GoToSocial on supported platforms offers most of the functionality of `ffmpeg` and so on, without some of the headaches.
However, not all platforms are capable of running Wazero in the much-faster "compiler" mode, and have to fall back to the very slow (and resource-heavy) "interpreter" mode. See [this table](https://github.com/tetratelabs/wazero?tab=readme-ov-file#conformance) from Wazero for conformance.
"Interpreter" mode runs so poorly for GoToSocial's use case that it's simply not feasible to run a GoToSocial instance in a stable manner on platforms that aren't 64-bit Linux or 64-bit FreeBSD, as all the memory and CPU get gobbled up by media processing.
However! To enable folks to run **experimental, unsupported deployments of GoToSocial**, we expose the `nowasm` build tag, which can be used to compile a build of GoToSocial that does not use Wazero or WASM at all.
A GoToSocial binary built with `nowasm` will use the [modernc version of SQLite](https://pkg.go.dev/modernc.org/sqlite) instead of the WASM one, and will use on-system `ffmpeg` and `ffprobe` binaries for media processing.
To build GoToSocial with the `nowasm` tag, you can pass the tag into our convenience `build.sh` script like so:
```bash
GO_BUILDTAGS=nowasm ./scripts/build.sh
```
In order to run a version of GoToSocial built in this way, you must ensure that `ffmpeg` and `ffprobe` are installed on the host. This is usually as simple as running a command like `doas -u root pkg_add ffmpeg` (OpenBSD), or `sudo apt install ffmpeg` (Debian etc.).
!!! Danger "No really though, it's unsupported"
Again, if running builds of GoToSocial with `nowasm` works for your OS/Arch combination, that's great, but we do not support such builds and we won't be able to help debugging why something doesn't work.

View file

@ -1,6 +1,6 @@
# Advanced # Advanced
In this section we touch on a number of more advanced topics, primarily related around deploying, operating and tuning GoToSocial. In this section we touch on a number of more advanced topics, primarily related around building, deploying, operating and tuning GoToSocial.
We consider these topics advanced because applying them incorrectly does have the possibility of causing client and federation issues. Applying any of these configuration changes may also make it harder for you to debug an issue with your GoToSocial instance if you don't understand the changes that you're making. We consider these topics advanced because applying them incorrectly does have the possibility of causing client and federation issues. Applying any of these configuration changes may also make it harder for you to debug an issue with your GoToSocial instance if you don't understand the changes that you're making.

View file

@ -1,47 +1,11 @@
# Releases # Releases
GoToSocial can be installed in a number of different ways. We publish official binary releases as well as container images. A number of third-party packages are maintained by different distributions and some people have created additional deployment tooling to make it easy to deploy GoToSocial yourself. GoToSocial can be installed in a number of different ways. We publish official binary releases as well as container images.
## Binary releases A number of third-party packages are maintained by different distributions, and some people have created additional deployment tooling to make it easy to deploy GoToSocial yourself.
We publish binary builds for Linux to [our GitHub project](https://github.com/superseriousbusiness/gotosocial/releases): {%
include "../../README.md"
* 32-bit Intel/AMD (i386/x86) start='<!--releases-start-->'
* 64-bit Intel/AMD (amd64/x86_64) end='<!--releases-end-->'
* 32-bit ARM (v6 and v7) %}
* 64-bit ARM64
For FreeBSD we publish:
* 64-bit Intel/AMD (amd64/x86_64)
## Containers
We also publish container images [on the Docker Hub](https://hub.docker.com/r/superseriousbusiness/gotosocial).
Containers are released for the same Linux platforms as our binary releases, with the exception of 32-bit Intel/AMD.
## Snapshots
We publish snapshot binary builds and Docker images of whatever is currently on main.
We always recommend using a stable release instead, but if you want to live on the edge (at your own risk!) then see the [snapshots](https://github.com/superseriousbusiness/gotosocial#snapshots) section on our GitHub repo for more information.
## Third-party
Some folks have created distribution packages for GoToSocial or additional tooling to aid in installing GoToSocial.
### Distribution packages
These packages are not maintained by GoToSocial, so please direct questions and issues to the repository maintainers (and donate to them!).
[![Packaging status](https://repology.org/badge/vertical-allrepos/gotosocial.svg)](https://repology.org/project/gotosocial/versions)
### Deployment tools
You can deploy your own instance of GoToSocial with the help of:
- [YunoHost GoToSocial Packaging](https://github.com/YunoHost-Apps/gotosocial_ynh) by [OniriCorpe](https://github.com/OniriCorpe).
- [Ansible Playbook (MASH)](https://github.com/mother-of-all-self-hosting/mash-playbook): The playbook supports a many services, including GoToSocial. [Documentation](https://github.com/mother-of-all-self-hosting/mash-playbook/blob/main/docs/services/gotosocial.md)
- GoToSocial Helm Charts:
- [GoToSocial Helm Chart](https://github.com/fSocietySocial/charts/tree/main/charts/gotosocial) by [0hlov3](https://github.com/0hlov3).

6
go.mod
View file

@ -42,7 +42,7 @@ require (
github.com/k3a/html2text v1.2.1 github.com/k3a/html2text v1.2.1
github.com/microcosm-cc/bluemonday v1.0.27 github.com/microcosm-cc/bluemonday v1.0.27
github.com/miekg/dns v1.1.62 github.com/miekg/dns v1.1.62
github.com/minio/minio-go/v7 v7.0.77 github.com/minio/minio-go/v7 v7.0.78
github.com/mitchellh/mapstructure v1.5.0 github.com/mitchellh/mapstructure v1.5.0
github.com/ncruces/go-sqlite3 v0.19.0 github.com/ncruces/go-sqlite3 v0.19.0
github.com/oklog/ulid v1.3.1 github.com/oklog/ulid v1.3.1
@ -63,7 +63,7 @@ require (
github.com/uptrace/bun/dialect/sqlitedialect v1.2.1 github.com/uptrace/bun/dialect/sqlitedialect v1.2.1
github.com/uptrace/bun/extra/bunotel v1.2.1 github.com/uptrace/bun/extra/bunotel v1.2.1
github.com/wagslane/go-password-validator v0.3.0 github.com/wagslane/go-password-validator v0.3.0
github.com/yuin/goldmark v1.7.4 github.com/yuin/goldmark v1.7.6
go.opentelemetry.io/otel v1.29.0 go.opentelemetry.io/otel v1.29.0
go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracegrpc v1.29.0 go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracegrpc v1.29.0
go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.29.0 go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.29.0
@ -158,7 +158,7 @@ require (
github.com/jinzhu/inflection v1.0.0 // indirect github.com/jinzhu/inflection v1.0.0 // indirect
github.com/josharian/intern v1.0.0 // indirect github.com/josharian/intern v1.0.0 // indirect
github.com/json-iterator/go v1.1.12 // indirect github.com/json-iterator/go v1.1.12 // indirect
github.com/klauspost/compress v1.17.9 // indirect github.com/klauspost/compress v1.17.11 // indirect
github.com/klauspost/cpuid/v2 v2.2.8 // indirect github.com/klauspost/cpuid/v2 v2.2.8 // indirect
github.com/kr/pretty v0.3.1 // indirect github.com/kr/pretty v0.3.1 // indirect
github.com/kr/text v0.2.0 // indirect github.com/kr/text v0.2.0 // indirect

12
go.sum
View file

@ -381,8 +381,8 @@ github.com/k3a/html2text v1.2.1/go.mod h1:ieEXykM67iT8lTvEWBh6fhpH4B23kB9OMKPdIB
github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck= github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck=
github.com/klauspost/compress v1.10.4/go.mod h1:aoV0uJVorq1K+umq18yTdKaF57EivdYsUV+/s2qKfXs= github.com/klauspost/compress v1.10.4/go.mod h1:aoV0uJVorq1K+umq18yTdKaF57EivdYsUV+/s2qKfXs=
github.com/klauspost/compress v1.10.10/go.mod h1:aoV0uJVorq1K+umq18yTdKaF57EivdYsUV+/s2qKfXs= github.com/klauspost/compress v1.10.10/go.mod h1:aoV0uJVorq1K+umq18yTdKaF57EivdYsUV+/s2qKfXs=
github.com/klauspost/compress v1.17.9 h1:6KIumPrER1LHsvBVuDa0r5xaG0Es51mhhB9BQB2qeMA= github.com/klauspost/compress v1.17.11 h1:In6xLpyWOi1+C7tXUUWv2ot1QvBjxevKAaI6IXrJmUc=
github.com/klauspost/compress v1.17.9/go.mod h1:Di0epgTjJY877eYKx5yC51cX2A2Vl2ibi7bDH9ttBbw= github.com/klauspost/compress v1.17.11/go.mod h1:pMDklpSncoRMuLFrf1W9Ss9KT+0rH90U12bZKk7uwG0=
github.com/klauspost/cpuid/v2 v2.0.1/go.mod h1:FInQzS24/EEf25PyTYn52gqo7WaD8xa0213Md/qVLRg= github.com/klauspost/cpuid/v2 v2.0.1/go.mod h1:FInQzS24/EEf25PyTYn52gqo7WaD8xa0213Md/qVLRg=
github.com/klauspost/cpuid/v2 v2.0.9/go.mod h1:FInQzS24/EEf25PyTYn52gqo7WaD8xa0213Md/qVLRg= github.com/klauspost/cpuid/v2 v2.0.9/go.mod h1:FInQzS24/EEf25PyTYn52gqo7WaD8xa0213Md/qVLRg=
github.com/klauspost/cpuid/v2 v2.2.8 h1:+StwCXwm9PdpiEkPyzBXIy+M9KUb4ODm0Zarf1kS5BM= github.com/klauspost/cpuid/v2 v2.2.8 h1:+StwCXwm9PdpiEkPyzBXIy+M9KUb4ODm0Zarf1kS5BM=
@ -413,8 +413,8 @@ github.com/miekg/dns v1.1.62 h1:cN8OuEF1/x5Rq6Np+h1epln8OiyPWV+lROx9LxcGgIQ=
github.com/miekg/dns v1.1.62/go.mod h1:mvDlcItzm+br7MToIKqkglaGhlFMHJ9DTNNWONWXbNQ= github.com/miekg/dns v1.1.62/go.mod h1:mvDlcItzm+br7MToIKqkglaGhlFMHJ9DTNNWONWXbNQ=
github.com/minio/md5-simd v1.1.2 h1:Gdi1DZK69+ZVMoNHRXJyNcxrMA4dSxoYHZSQbirFg34= github.com/minio/md5-simd v1.1.2 h1:Gdi1DZK69+ZVMoNHRXJyNcxrMA4dSxoYHZSQbirFg34=
github.com/minio/md5-simd v1.1.2/go.mod h1:MzdKDxYpY2BT9XQFocsiZf/NKVtR7nkE4RoEpN+20RM= github.com/minio/md5-simd v1.1.2/go.mod h1:MzdKDxYpY2BT9XQFocsiZf/NKVtR7nkE4RoEpN+20RM=
github.com/minio/minio-go/v7 v7.0.77 h1:GaGghJRg9nwDVlNbwYjSDJT1rqltQkBFDsypWX1v3Bw= github.com/minio/minio-go/v7 v7.0.78 h1:LqW2zy52fxnI4gg8C2oZviTaKHcBV36scS+RzJnxUFs=
github.com/minio/minio-go/v7 v7.0.77/go.mod h1:AVM3IUN6WwKzmwBxVdjzhH8xq+f57JSbbvzqvUzR6eg= github.com/minio/minio-go/v7 v7.0.78/go.mod h1:84gmIilaX4zcvAWWzJ5Z1WI5axN+hAbM5w25xf8xvC0=
github.com/mitchellh/copystructure v1.0.0/go.mod h1:SNtv71yrdKgLRyLFxmLdkAbkKEFWgYaq1OVrnRcwhnw= github.com/mitchellh/copystructure v1.0.0/go.mod h1:SNtv71yrdKgLRyLFxmLdkAbkKEFWgYaq1OVrnRcwhnw=
github.com/mitchellh/copystructure v1.2.0 h1:vpKXTN4ewci03Vljg/q9QvCGUDttBOGBIa15WveJJGw= github.com/mitchellh/copystructure v1.2.0 h1:vpKXTN4ewci03Vljg/q9QvCGUDttBOGBIa15WveJJGw=
github.com/mitchellh/copystructure v1.2.0/go.mod h1:qLl+cE2AmVv+CoeAwDPye/v+N2HKCj9FbZEVFJRxO9s= github.com/mitchellh/copystructure v1.2.0/go.mod h1:qLl+cE2AmVv+CoeAwDPye/v+N2HKCj9FbZEVFJRxO9s=
@ -621,8 +621,8 @@ github.com/yuin/goldmark v1.1.25/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9de
github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74= github.com/yuin/goldmark v1.1.27/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
github.com/yuin/goldmark v1.1.32/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74= github.com/yuin/goldmark v1.1.32/go.mod h1:3hX8gzYuyVAZsxl0MRgGTJEmQBFcNTphYh9decYSb74=
github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY= github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=
github.com/yuin/goldmark v1.7.4 h1:BDXOHExt+A7gwPCJgPIIq7ENvceR7we7rOS9TNoLZeg= github.com/yuin/goldmark v1.7.6 h1:cZgJxVh5mL5cu8KOnwxvFJy5TFB0BHUskZZyq7TYbDg=
github.com/yuin/goldmark v1.7.4/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E= github.com/yuin/goldmark v1.7.6/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E=
gitlab.com/NyaaaWhatsUpDoc/sqlite v1.33.1-concurrency-workaround h1:pFMJnlc1PuH+jcVz4vz53vcpnoZG+NqFBr3qikDmEB4= gitlab.com/NyaaaWhatsUpDoc/sqlite v1.33.1-concurrency-workaround h1:pFMJnlc1PuH+jcVz4vz53vcpnoZG+NqFBr3qikDmEB4=
gitlab.com/NyaaaWhatsUpDoc/sqlite v1.33.1-concurrency-workaround/go.mod h1:pXV2xHxhzXZsgT/RtTFAPY6JJDEvOTcTdwADQCCWD4k= gitlab.com/NyaaaWhatsUpDoc/sqlite v1.33.1-concurrency-workaround/go.mod h1:pXV2xHxhzXZsgT/RtTFAPY6JJDEvOTcTdwADQCCWD4k=
go.mongodb.org/mongo-driver v1.14.0 h1:P98w8egYRjYe3XDjxhYJagTokP/H6HzlsnojRgZRd80= go.mongodb.org/mongo-driver v1.14.0 h1:P98w8egYRjYe3XDjxhYJagTokP/H6HzlsnojRgZRd80=

View file

@ -15,7 +15,7 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build !moderncsqlite3 //go:build !moderncsqlite3 && !nowasm
package sqlite package sqlite

View file

@ -15,7 +15,7 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build moderncsqlite3 //go:build moderncsqlite3 || nowasm
package sqlite package sqlite

View file

@ -15,7 +15,7 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build !moderncsqlite3 //go:build !moderncsqlite3 && !nowasm
package sqlite package sqlite

View file

@ -15,7 +15,7 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build moderncsqlite3 //go:build moderncsqlite3 || nowasm
package sqlite package sqlite

View file

@ -0,0 +1,28 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package ffmpeg
import (
"codeberg.org/gruf/go-ffmpreg/wasm"
)
// Args encapsulates the passing of common
// configuration options to run an instance
// of a compiled WebAssembly module that is
// run in a typical CLI manner.
type Args = wasm.Args

View file

@ -0,0 +1,142 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build nowasm
package ffmpeg
import (
"context"
"fmt"
"io"
"io/fs"
"os/exec"
"codeberg.org/gruf/go-ffmpreg/wasm"
"github.com/tetratelabs/wazero"
"github.com/tetratelabs/wazero/sys"
)
func init() {
fmt.Println("!! you are using an unsupported build configuration of gotosocial with WebAssembly disabled !!")
fmt.Println("!! please do not file bug reports regarding media processing with this configuration !!")
fmt.Println("!! it is also less secure; this does not enforce version checks on ffmpeg / ffprobe versions !!")
}
// runCmd will run 'name' with the given arguments, returning exit code or error.
func runCmd(ctx context.Context, name string, args wasm.Args) (uint32, error) {
cmd := exec.CommandContext(ctx, name, args.Args...) //nolint:gosec
// Set provided std files.
cmd.Stdin = args.Stdin
cmd.Stdout = args.Stdout
cmd.Stderr = args.Stderr
if args.Config != nil {
// Gather some information
// from module config func.
var cfg falseModuleConfig
_ = args.Config(&cfg)
// Extract from conf.
cmd.Env = cfg.env
}
// Run prepared command, catching err type.
switch err := cmd.Run(); err := err.(type) {
// Extract code from
// any exit error type.
case *exec.ExitError:
rc := err.ExitCode()
return uint32(rc), err
default:
return 0, err
}
}
type falseModuleConfig struct{ env []string }
func (cfg *falseModuleConfig) WithArgs(...string) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithEnv(key string, value string) wazero.ModuleConfig {
cfg.env = append(cfg.env, key+"="+value)
return cfg // noop
}
func (cfg *falseModuleConfig) WithFS(fs.FS) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithFSConfig(wazero.FSConfig) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithName(string) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithStartFunctions(...string) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithStderr(io.Writer) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithStdin(io.Reader) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithStdout(io.Writer) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithWalltime(sys.Walltime, sys.ClockResolution) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithSysWalltime() wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithNanotime(sys.Nanotime, sys.ClockResolution) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithSysNanotime() wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithNanosleep(sys.Nanosleep) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithOsyield(sys.Osyield) wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithSysNanosleep() wazero.ModuleConfig {
return cfg // noop
}
func (cfg *falseModuleConfig) WithRandSource(io.Reader) wazero.ModuleConfig {
return cfg // noop
}

View file

@ -15,10 +15,14 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build !nowasm
package ffmpeg package ffmpeg
import ( import (
"context" "context"
"codeberg.org/gruf/go-ffmpreg/wasm"
) )
// ffmpegRunner limits the number of // ffmpegRunner limits the number of
@ -36,5 +40,7 @@ func InitFfmpeg(ctx context.Context, max int) error {
// Ffmpeg runs the given arguments with an instance of ffmpeg. // Ffmpeg runs the given arguments with an instance of ffmpeg.
func Ffmpeg(ctx context.Context, args Args) (uint32, error) { func Ffmpeg(ctx context.Context, args Args) (uint32, error) {
return ffmpegRunner.Run(ctx, ffmpeg, args) return ffmpegRunner.Run(ctx, func() (uint32, error) {
return wasm.Run(ctx, runtime, ffmpeg, args)
})
} }

View file

@ -0,0 +1,49 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build nowasm
package ffmpeg
import (
"context"
"os/exec"
)
// ffmpegRunner limits the number of
// ffmpeg WebAssembly instances that
// may be concurrently running, in
// order to reduce memory usage.
var ffmpegRunner runner
// InitFfmpeg looks for a local copy of ffmpeg in path, and prepares
// the runner to only allow max given concurrent running instances.
func InitFfmpeg(ctx context.Context, max int) error {
_, err := exec.LookPath("ffmpeg")
if err != nil {
return err
}
ffmpegRunner.Init(max)
return nil
}
// Ffmpeg runs the given arguments with an instance of ffmpeg.
func Ffmpeg(ctx context.Context, args Args) (uint32, error) {
return ffmpegRunner.Run(ctx, func() (uint32, error) {
return runCmd(ctx, "ffmpeg", args)
})
}

View file

@ -15,10 +15,14 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build !nowasm
package ffmpeg package ffmpeg
import ( import (
"context" "context"
"codeberg.org/gruf/go-ffmpreg/wasm"
) )
// ffprobeRunner limits the number of // ffprobeRunner limits the number of
@ -36,5 +40,7 @@ func InitFfprobe(ctx context.Context, max int) error {
// Ffprobe runs the given arguments with an instance of ffprobe. // Ffprobe runs the given arguments with an instance of ffprobe.
func Ffprobe(ctx context.Context, args Args) (uint32, error) { func Ffprobe(ctx context.Context, args Args) (uint32, error) {
return ffprobeRunner.Run(ctx, ffprobe, args) return ffmpegRunner.Run(ctx, func() (uint32, error) {
return wasm.Run(ctx, runtime, ffprobe, args)
})
} }

View file

@ -0,0 +1,49 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build nowasm
package ffmpeg
import (
"context"
"os/exec"
)
// ffprobeRunner limits the number of
// ffprobe WebAssembly instances that
// may be concurrently running, in
// order to reduce memory usage.
var ffprobeRunner runner
// InitFfprobe looks for a local copy of ffprobe in path, and prepares
// the runner to only allow max given concurrent running instances.
func InitFfprobe(ctx context.Context, max int) error {
_, err := exec.LookPath("ffprobe")
if err != nil {
return err
}
ffprobeRunner.Init(max)
return nil
}
// Ffprobe runs the given arguments with an instance of ffprobe.
func Ffprobe(ctx context.Context, args Args) (uint32, error) {
return ffprobeRunner.Run(ctx, func() (uint32, error) {
return runCmd(ctx, "ffprobe", args)
})
}

View file

@ -19,9 +19,6 @@
import ( import (
"context" "context"
"codeberg.org/gruf/go-ffmpreg/wasm"
"github.com/tetratelabs/wazero"
) )
// runner simply abstracts away the complexities // runner simply abstracts away the complexities
@ -53,7 +50,7 @@ func (r *runner) Init(n int) {
// Run will attempt to pass the given compiled WebAssembly module with args to run(), waiting on // Run will attempt to pass the given compiled WebAssembly module with args to run(), waiting on
// the receiving runner until a free slot is available to run an instance, (if a limit is enabled). // the receiving runner until a free slot is available to run an instance, (if a limit is enabled).
func (r *runner) Run(ctx context.Context, cmod wazero.CompiledModule, args Args) (uint32, error) { func (r *runner) Run(ctx context.Context, run func() (uint32, error)) (uint32, error) {
select { select {
// Context canceled. // Context canceled.
case <-ctx.Done(): case <-ctx.Done():
@ -66,6 +63,6 @@ func (r *runner) Run(ctx context.Context, cmod wazero.CompiledModule, args Args)
// Release slot back to pool on end. // Release slot back to pool on end.
defer func() { r.pool <- struct{}{} }() defer func() { r.pool <- struct{}{} }()
// Pass to main module runner function. // Call run.
return wasm.Run(ctx, runtime, cmod, args) return run()
} }

View file

@ -15,6 +15,8 @@
// You should have received a copy of the GNU Affero General Public License // You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>. // along with this program. If not, see <http://www.gnu.org/licenses/>.
//go:build !nowasm
package ffmpeg package ffmpeg
import ( import (
@ -41,12 +43,6 @@
ffprobe wazero.CompiledModule ffprobe wazero.CompiledModule
) )
// Args encapsulates the passing of common
// configuration options to run an instance
// of a compiled WebAssembly module that is
// run in a typical CLI manner.
type Args = wasm.Args
// compileFfmpeg ensures the ffmpeg WebAssembly has been // compileFfmpeg ensures the ffmpeg WebAssembly has been
// pre-compiled into memory. If already compiled is a no-op. // pre-compiled into memory. If already compiled is a no-op.
func compileFfmpeg(ctx context.Context) error { func compileFfmpeg(ctx context.Context) error {

View file

@ -100,7 +100,7 @@ nav:
- "configuration/advanced.md" - "configuration/advanced.md"
- "configuration/observability.md" - "configuration/observability.md"
- "Advanced": - "Advanced":
- "advanced/index.md" - "Overview": "advanced/index.md"
- "advanced/host-account-domain.md" - "advanced/host-account-domain.md"
- "advanced/outgoing-proxy.md" - "advanced/outgoing-proxy.md"
- "Caching": - "Caching":
@ -117,6 +117,8 @@ nav:
- "advanced/metrics.md" - "advanced/metrics.md"
- "advanced/replicating-sqlite.md" - "advanced/replicating-sqlite.md"
- "advanced/sqlite-networked-storage.md" - "advanced/sqlite-networked-storage.md"
- "Advanced builds":
- "advanced/builds/nowasm.md"
- "Admin": - "Admin":
- "admin/settings.md" - "admin/settings.md"

View file

@ -15,14 +15,16 @@ GO_GCFLAGS=${GO_GCFLAGS-}
GO_BUILDTAGS="${GO_BUILDTAGS} debugenv" GO_BUILDTAGS="${GO_BUILDTAGS} debugenv"
# Available Go build tags, with explanation, followed by benefits of enabling it: # Available Go build tags, with explanation, followed by benefits of enabling it:
# - kvformat: enables prettier output of log fields (slightly better performance) # - kvformat: enables prettier output of log fields (slightly better performance)
# - timetzdata: embed timezone database inside binary (allow setting local time inside Docker containers, at cost of 450KB) # - timetzdata: embed timezone database inside binary (allow setting local time inside Docker containers, at cost of 450KB)
# - notracing: disables compiling-in otel tracing support (reduced binary size, better performance) # - notracing: disables compiling-in otel tracing support (reduced binary size, better performance)
# - nometrics: disables compiling-in otel metrics support (reduced binary size, better performance) # - nometrics: disables compiling-in otel metrics support (reduced binary size, better performance)
# - noerrcaller: disables caller function prefix in errors (slightly better performance, at cost of err readability) # - noerrcaller: disables caller function prefix in errors (slightly better performance, at cost of err readability)
# - debug: enables /debug/pprof endpoint (adds debug, at performance cost) # - debug: enables /debug/pprof endpoint (adds debug, at performance cost)
# - debugenv: enables /debug/pprof endpoint if DEBUG=1 env during runtime (adds debug, at performance cost) # - debugenv: enables /debug/pprof endpoint if DEBUG=1 env during runtime (adds debug, at performance cost)
# - moderncsqlite3: reverts to using the C-to-Go transpiled SQLite driver (disables the WASM-based SQLite driver) # - moderncsqlite3: reverts to using the C-to-Go transpiled SQLite driver (disables the WASM-based SQLite driver)
# - nowasm: [UNSUPPORTED] removes all WebAssembly from builds including
# ffmpeg, ffprobe and SQLite (instead falling back to modernc).
log_exec env CGO_ENABLED=0 go build -trimpath -v \ log_exec env CGO_ENABLED=0 go build -trimpath -v \
-tags "${GO_BUILDTAGS}" \ -tags "${GO_BUILDTAGS}" \
-ldflags="${GO_LDFLAGS}" \ -ldflags="${GO_LDFLAGS}" \

View file

@ -1,5 +1,5 @@
# This is an example goreleaser.yaml file with some sane defaults. version: 2
# Make sure to check the documentation at http://goreleaser.com
before: before:
hooks: hooks:
- ./gen.sh - ./gen.sh
@ -99,7 +99,7 @@ archives:
checksum: checksum:
name_template: 'checksums.txt' name_template: 'checksums.txt'
snapshot: snapshot:
name_template: "{{ .Tag }}-next" version_template: "{{ .Tag }}-next"
changelog: changelog:
sort: asc sort: asc
filters: filters:

View file

@ -16,6 +16,27 @@ This package provides various compression algorithms.
# changelog # changelog
* Sep 23rd, 2024 - [1.17.10](https://github.com/klauspost/compress/releases/tag/v1.17.10)
* gzhttp: Add TransportAlwaysDecompress option. https://github.com/klauspost/compress/pull/978
* gzhttp: Add supported decompress request body by @mirecl in https://github.com/klauspost/compress/pull/1002
* s2: Add EncodeBuffer buffer recycling callback https://github.com/klauspost/compress/pull/982
* zstd: Improve memory usage on small streaming encodes https://github.com/klauspost/compress/pull/1007
* flate: read data written with partial flush by @vajexal in https://github.com/klauspost/compress/pull/996
* Jun 12th, 2024 - [1.17.9](https://github.com/klauspost/compress/releases/tag/v1.17.9)
* s2: Reduce ReadFrom temporary allocations https://github.com/klauspost/compress/pull/949
* flate, zstd: Shave some bytes off amd64 matchLen by @greatroar in https://github.com/klauspost/compress/pull/963
* Upgrade zip/zlib to 1.22.4 upstream https://github.com/klauspost/compress/pull/970 https://github.com/klauspost/compress/pull/971
* zstd: BuildDict fails with RLE table https://github.com/klauspost/compress/pull/951
* Apr 9th, 2024 - [1.17.8](https://github.com/klauspost/compress/releases/tag/v1.17.8)
* zstd: Reject blocks where reserved values are not 0 https://github.com/klauspost/compress/pull/885
* zstd: Add RLE detection+encoding https://github.com/klauspost/compress/pull/938
* Feb 21st, 2024 - [1.17.7](https://github.com/klauspost/compress/releases/tag/v1.17.7)
* s2: Add AsyncFlush method: Complete the block without flushing by @Jille in https://github.com/klauspost/compress/pull/927
* s2: Fix literal+repeat exceeds dst crash https://github.com/klauspost/compress/pull/930
* Feb 5th, 2024 - [1.17.6](https://github.com/klauspost/compress/releases/tag/v1.17.6) * Feb 5th, 2024 - [1.17.6](https://github.com/klauspost/compress/releases/tag/v1.17.6)
* zstd: Fix incorrect repeat coding in best mode https://github.com/klauspost/compress/pull/923 * zstd: Fix incorrect repeat coding in best mode https://github.com/klauspost/compress/pull/923
* s2: Fix DecodeConcurrent deadlock on errors https://github.com/klauspost/compress/pull/925 * s2: Fix DecodeConcurrent deadlock on errors https://github.com/klauspost/compress/pull/925
@ -81,7 +102,7 @@ https://github.com/klauspost/compress/pull/919 https://github.com/klauspost/comp
* zstd: Various minor improvements by @greatroar in https://github.com/klauspost/compress/pull/788 https://github.com/klauspost/compress/pull/794 https://github.com/klauspost/compress/pull/795 * zstd: Various minor improvements by @greatroar in https://github.com/klauspost/compress/pull/788 https://github.com/klauspost/compress/pull/794 https://github.com/klauspost/compress/pull/795
* s2: Fix huge block overflow https://github.com/klauspost/compress/pull/779 * s2: Fix huge block overflow https://github.com/klauspost/compress/pull/779
* s2: Allow CustomEncoder fallback https://github.com/klauspost/compress/pull/780 * s2: Allow CustomEncoder fallback https://github.com/klauspost/compress/pull/780
* gzhttp: Suppport ResponseWriter Unwrap() in gzhttp handler by @jgimenez in https://github.com/klauspost/compress/pull/799 * gzhttp: Support ResponseWriter Unwrap() in gzhttp handler by @jgimenez in https://github.com/klauspost/compress/pull/799
* Mar 13, 2023 - [v1.16.1](https://github.com/klauspost/compress/releases/tag/v1.16.1) * Mar 13, 2023 - [v1.16.1](https://github.com/klauspost/compress/releases/tag/v1.16.1)
* zstd: Speed up + improve best encoder by @greatroar in https://github.com/klauspost/compress/pull/776 * zstd: Speed up + improve best encoder by @greatroar in https://github.com/klauspost/compress/pull/776
@ -136,7 +157,7 @@ https://github.com/klauspost/compress/pull/919 https://github.com/klauspost/comp
* zstd: Add [WithDecodeAllCapLimit](https://pkg.go.dev/github.com/klauspost/compress@v1.15.10/zstd#WithDecodeAllCapLimit) https://github.com/klauspost/compress/pull/649 * zstd: Add [WithDecodeAllCapLimit](https://pkg.go.dev/github.com/klauspost/compress@v1.15.10/zstd#WithDecodeAllCapLimit) https://github.com/klauspost/compress/pull/649
* Add Go 1.19 - deprecate Go 1.16 https://github.com/klauspost/compress/pull/651 * Add Go 1.19 - deprecate Go 1.16 https://github.com/klauspost/compress/pull/651
* flate: Improve level 5+6 compression https://github.com/klauspost/compress/pull/656 * flate: Improve level 5+6 compression https://github.com/klauspost/compress/pull/656
* zstd: Improve "better" compresssion https://github.com/klauspost/compress/pull/657 * zstd: Improve "better" compression https://github.com/klauspost/compress/pull/657
* s2: Improve "best" compression https://github.com/klauspost/compress/pull/658 * s2: Improve "best" compression https://github.com/klauspost/compress/pull/658
* s2: Improve "better" compression. https://github.com/klauspost/compress/pull/635 * s2: Improve "better" compression. https://github.com/klauspost/compress/pull/635
* s2: Slightly faster non-assembly decompression https://github.com/klauspost/compress/pull/646 * s2: Slightly faster non-assembly decompression https://github.com/klauspost/compress/pull/646
@ -339,7 +360,7 @@ While the release has been extensively tested, it is recommended to testing when
* s2: Fix binaries. * s2: Fix binaries.
* Feb 25, 2021 (v1.11.8) * Feb 25, 2021 (v1.11.8)
* s2: Fixed occational out-of-bounds write on amd64. Upgrade recommended. * s2: Fixed occasional out-of-bounds write on amd64. Upgrade recommended.
* s2: Add AMD64 assembly for better mode. 25-50% faster. [#315](https://github.com/klauspost/compress/pull/315) * s2: Add AMD64 assembly for better mode. 25-50% faster. [#315](https://github.com/klauspost/compress/pull/315)
* s2: Less upfront decoder allocation. [#322](https://github.com/klauspost/compress/pull/322) * s2: Less upfront decoder allocation. [#322](https://github.com/klauspost/compress/pull/322)
* zstd: Faster "compression" of incompressible data. [#314](https://github.com/klauspost/compress/pull/314) * zstd: Faster "compression" of incompressible data. [#314](https://github.com/klauspost/compress/pull/314)
@ -518,7 +539,7 @@ While the release has been extensively tested, it is recommended to testing when
* Feb 19, 2016: Faster bit writer, level -2 is 15% faster, level 1 is 4% faster. * Feb 19, 2016: Faster bit writer, level -2 is 15% faster, level 1 is 4% faster.
* Feb 19, 2016: Handle small payloads faster in level 1-3. * Feb 19, 2016: Handle small payloads faster in level 1-3.
* Feb 19, 2016: Added faster level 2 + 3 compression modes. * Feb 19, 2016: Added faster level 2 + 3 compression modes.
* Feb 19, 2016: [Rebalanced compression levels](https://blog.klauspost.com/rebalancing-deflate-compression-levels/), so there is a more even progresssion in terms of compression. New default level is 5. * Feb 19, 2016: [Rebalanced compression levels](https://blog.klauspost.com/rebalancing-deflate-compression-levels/), so there is a more even progression in terms of compression. New default level is 5.
* Feb 14, 2016: Snappy: Merge upstream changes. * Feb 14, 2016: Snappy: Merge upstream changes.
* Feb 14, 2016: Snappy: Fix aggressive skipping. * Feb 14, 2016: Snappy: Fix aggressive skipping.
* Feb 14, 2016: Snappy: Update benchmark. * Feb 14, 2016: Snappy: Update benchmark.

View file

@ -15,7 +15,7 @@
// It is possible, but by no way guaranteed that corrupt data will // It is possible, but by no way guaranteed that corrupt data will
// return an error. // return an error.
// It is up to the caller to verify integrity of the returned data. // It is up to the caller to verify integrity of the returned data.
// Use a predefined Scrach to set maximum acceptable output size. // Use a predefined Scratch to set maximum acceptable output size.
func Decompress(b []byte, s *Scratch) ([]byte, error) { func Decompress(b []byte, s *Scratch) ([]byte, error) {
s, err := s.prepare(b) s, err := s.prepare(b)
if err != nil { if err != nil {

View file

@ -1136,7 +1136,7 @@ func (s *Scratch) matches(ct cTable, w io.Writer) {
errs++ errs++
} }
if errs > 0 { if errs > 0 {
fmt.Fprintf(w, "%d errros in base, stopping\n", errs) fmt.Fprintf(w, "%d errors in base, stopping\n", errs)
continue continue
} }
// Ensure that all combinations are covered. // Ensure that all combinations are covered.
@ -1152,7 +1152,7 @@ func (s *Scratch) matches(ct cTable, w io.Writer) {
errs++ errs++
} }
if errs > 20 { if errs > 20 {
fmt.Fprintf(w, "%d errros, stopping\n", errs) fmt.Fprintf(w, "%d errors, stopping\n", errs)
break break
} }
} }

View file

@ -9,6 +9,9 @@
"encoding/binary" "encoding/binary"
"math" "math"
"math/bits" "math/bits"
"sync"
"github.com/klauspost/compress/internal/race"
) )
// Encode returns the encoded form of src. The returned slice may be a sub- // Encode returns the encoded form of src. The returned slice may be a sub-
@ -52,6 +55,8 @@ func Encode(dst, src []byte) []byte {
return dst[:d] return dst[:d]
} }
var estblockPool [2]sync.Pool
// EstimateBlockSize will perform a very fast compression // EstimateBlockSize will perform a very fast compression
// without outputting the result and return the compressed output size. // without outputting the result and return the compressed output size.
// The function returns -1 if no improvement could be achieved. // The function returns -1 if no improvement could be achieved.
@ -61,9 +66,25 @@ func EstimateBlockSize(src []byte) (d int) {
return -1 return -1
} }
if len(src) <= 1024 { if len(src) <= 1024 {
d = calcBlockSizeSmall(src) const sz, pool = 2048, 0
tmp, ok := estblockPool[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer estblockPool[pool].Put(tmp)
d = calcBlockSizeSmall(src, tmp)
} else { } else {
d = calcBlockSize(src) const sz, pool = 32768, 1
tmp, ok := estblockPool[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer estblockPool[pool].Put(tmp)
d = calcBlockSize(src, tmp)
} }
if d == 0 { if d == 0 {

View file

@ -3,10 +3,16 @@
package s2 package s2
import "github.com/klauspost/compress/internal/race" import (
"sync"
"github.com/klauspost/compress/internal/race"
)
const hasAmd64Asm = true const hasAmd64Asm = true
var encPools [4]sync.Pool
// encodeBlock encodes a non-empty src to a guaranteed-large-enough dst. It // encodeBlock encodes a non-empty src to a guaranteed-large-enough dst. It
// assumes that the varint-encoded length of the decompressed bytes has already // assumes that the varint-encoded length of the decompressed bytes has already
// been written. // been written.
@ -29,23 +35,60 @@ func encodeBlock(dst, src []byte) (d int) {
) )
if len(src) >= 4<<20 { if len(src) >= 4<<20 {
return encodeBlockAsm(dst, src) const sz, pool = 65536, 0
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeBlockAsm(dst, src, tmp)
} }
if len(src) >= limit12B { if len(src) >= limit12B {
return encodeBlockAsm4MB(dst, src) const sz, pool = 65536, 0
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeBlockAsm4MB(dst, src, tmp)
} }
if len(src) >= limit10B { if len(src) >= limit10B {
return encodeBlockAsm12B(dst, src) const sz, pool = 16384, 1
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeBlockAsm12B(dst, src, tmp)
} }
if len(src) >= limit8B { if len(src) >= limit8B {
return encodeBlockAsm10B(dst, src) const sz, pool = 4096, 2
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeBlockAsm10B(dst, src, tmp)
} }
if len(src) < minNonLiteralBlockSize { if len(src) < minNonLiteralBlockSize {
return 0 return 0
} }
return encodeBlockAsm8B(dst, src) const sz, pool = 1024, 3
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeBlockAsm8B(dst, src, tmp)
} }
var encBetterPools [5]sync.Pool
// encodeBlockBetter encodes a non-empty src to a guaranteed-large-enough dst. It // encodeBlockBetter encodes a non-empty src to a guaranteed-large-enough dst. It
// assumes that the varint-encoded length of the decompressed bytes has already // assumes that the varint-encoded length of the decompressed bytes has already
// been written. // been written.
@ -68,21 +111,59 @@ func encodeBlockBetter(dst, src []byte) (d int) {
) )
if len(src) > 4<<20 { if len(src) > 4<<20 {
return encodeBetterBlockAsm(dst, src) const sz, pool = 589824, 0
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeBetterBlockAsm(dst, src, tmp)
} }
if len(src) >= limit12B { if len(src) >= limit12B {
return encodeBetterBlockAsm4MB(dst, src) const sz, pool = 589824, 0
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeBetterBlockAsm4MB(dst, src, tmp)
} }
if len(src) >= limit10B { if len(src) >= limit10B {
return encodeBetterBlockAsm12B(dst, src) const sz, pool = 81920, 0
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeBetterBlockAsm12B(dst, src, tmp)
} }
if len(src) >= limit8B { if len(src) >= limit8B {
return encodeBetterBlockAsm10B(dst, src) const sz, pool = 20480, 1
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeBetterBlockAsm10B(dst, src, tmp)
} }
if len(src) < minNonLiteralBlockSize { if len(src) < minNonLiteralBlockSize {
return 0 return 0
} }
return encodeBetterBlockAsm8B(dst, src)
const sz, pool = 5120, 2
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeBetterBlockAsm8B(dst, src, tmp)
} }
// encodeBlockSnappy encodes a non-empty src to a guaranteed-large-enough dst. It // encodeBlockSnappy encodes a non-empty src to a guaranteed-large-enough dst. It
@ -105,22 +186,57 @@ func encodeBlockSnappy(dst, src []byte) (d int) {
// Use 8 bit table when less than... // Use 8 bit table when less than...
limit8B = 512 limit8B = 512
) )
if len(src) >= 64<<10 { if len(src) > 65536 {
return encodeSnappyBlockAsm(dst, src) const sz, pool = 65536, 0
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeSnappyBlockAsm(dst, src, tmp)
} }
if len(src) >= limit12B { if len(src) >= limit12B {
return encodeSnappyBlockAsm64K(dst, src) const sz, pool = 65536, 0
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeSnappyBlockAsm64K(dst, src, tmp)
} }
if len(src) >= limit10B { if len(src) >= limit10B {
return encodeSnappyBlockAsm12B(dst, src) const sz, pool = 16384, 1
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeSnappyBlockAsm12B(dst, src, tmp)
} }
if len(src) >= limit8B { if len(src) >= limit8B {
return encodeSnappyBlockAsm10B(dst, src) const sz, pool = 4096, 2
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeSnappyBlockAsm10B(dst, src, tmp)
} }
if len(src) < minNonLiteralBlockSize { if len(src) < minNonLiteralBlockSize {
return 0 return 0
} }
return encodeSnappyBlockAsm8B(dst, src) const sz, pool = 1024, 3
tmp, ok := encPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encPools[pool].Put(tmp)
return encodeSnappyBlockAsm8B(dst, src, tmp)
} }
// encodeBlockSnappy encodes a non-empty src to a guaranteed-large-enough dst. It // encodeBlockSnappy encodes a non-empty src to a guaranteed-large-enough dst. It
@ -143,20 +259,59 @@ func encodeBlockBetterSnappy(dst, src []byte) (d int) {
// Use 8 bit table when less than... // Use 8 bit table when less than...
limit8B = 512 limit8B = 512
) )
if len(src) >= 64<<10 { if len(src) > 65536 {
return encodeSnappyBetterBlockAsm(dst, src) const sz, pool = 589824, 0
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeSnappyBetterBlockAsm(dst, src, tmp)
} }
if len(src) >= limit12B { if len(src) >= limit12B {
return encodeSnappyBetterBlockAsm64K(dst, src) const sz, pool = 294912, 4
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeSnappyBetterBlockAsm64K(dst, src, tmp)
} }
if len(src) >= limit10B { if len(src) >= limit10B {
return encodeSnappyBetterBlockAsm12B(dst, src) const sz, pool = 81920, 0
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeSnappyBetterBlockAsm12B(dst, src, tmp)
} }
if len(src) >= limit8B { if len(src) >= limit8B {
return encodeSnappyBetterBlockAsm10B(dst, src) const sz, pool = 20480, 1
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeSnappyBetterBlockAsm10B(dst, src, tmp)
} }
if len(src) < minNonLiteralBlockSize { if len(src) < minNonLiteralBlockSize {
return 0 return 0
} }
return encodeSnappyBetterBlockAsm8B(dst, src)
const sz, pool = 5120, 2
tmp, ok := encBetterPools[pool].Get().(*[sz]byte)
if !ok {
tmp = &[sz]byte{}
}
race.WriteSlice(tmp[:])
defer encBetterPools[pool].Put(tmp)
return encodeSnappyBetterBlockAsm8B(dst, src, tmp)
} }

View file

@ -317,7 +317,7 @@ func matchLen(a []byte, b []byte) int {
} }
// input must be > inputMargin // input must be > inputMargin
func calcBlockSize(src []byte) (d int) { func calcBlockSize(src []byte, _ *[32768]byte) (d int) {
// Initialize the hash table. // Initialize the hash table.
const ( const (
tableBits = 13 tableBits = 13
@ -503,7 +503,7 @@ func calcBlockSize(src []byte) (d int) {
} }
// length must be > inputMargin. // length must be > inputMargin.
func calcBlockSizeSmall(src []byte) (d int) { func calcBlockSizeSmall(src []byte, _ *[2048]byte) (d int) {
// Initialize the hash table. // Initialize the hash table.
const ( const (
tableBits = 9 tableBits = 9

View file

@ -11,154 +11,154 @@ func _dummy_()
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBlockAsm(dst []byte, src []byte) int func encodeBlockAsm(dst []byte, src []byte, tmp *[65536]byte) int
// encodeBlockAsm4MB encodes a non-empty src to a guaranteed-large-enough dst. // encodeBlockAsm4MB encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4194304 bytes. // Maximum input 4194304 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBlockAsm4MB(dst []byte, src []byte) int func encodeBlockAsm4MB(dst []byte, src []byte, tmp *[65536]byte) int
// encodeBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst. // encodeBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 16383 bytes. // Maximum input 16383 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBlockAsm12B(dst []byte, src []byte) int func encodeBlockAsm12B(dst []byte, src []byte, tmp *[16384]byte) int
// encodeBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst. // encodeBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4095 bytes. // Maximum input 4095 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBlockAsm10B(dst []byte, src []byte) int func encodeBlockAsm10B(dst []byte, src []byte, tmp *[4096]byte) int
// encodeBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst. // encodeBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 511 bytes. // Maximum input 511 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBlockAsm8B(dst []byte, src []byte) int func encodeBlockAsm8B(dst []byte, src []byte, tmp *[1024]byte) int
// encodeBetterBlockAsm encodes a non-empty src to a guaranteed-large-enough dst. // encodeBetterBlockAsm encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4294967295 bytes. // Maximum input 4294967295 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBetterBlockAsm(dst []byte, src []byte) int func encodeBetterBlockAsm(dst []byte, src []byte, tmp *[589824]byte) int
// encodeBetterBlockAsm4MB encodes a non-empty src to a guaranteed-large-enough dst. // encodeBetterBlockAsm4MB encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4194304 bytes. // Maximum input 4194304 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBetterBlockAsm4MB(dst []byte, src []byte) int func encodeBetterBlockAsm4MB(dst []byte, src []byte, tmp *[589824]byte) int
// encodeBetterBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst. // encodeBetterBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 16383 bytes. // Maximum input 16383 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBetterBlockAsm12B(dst []byte, src []byte) int func encodeBetterBlockAsm12B(dst []byte, src []byte, tmp *[81920]byte) int
// encodeBetterBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst. // encodeBetterBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4095 bytes. // Maximum input 4095 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBetterBlockAsm10B(dst []byte, src []byte) int func encodeBetterBlockAsm10B(dst []byte, src []byte, tmp *[20480]byte) int
// encodeBetterBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst. // encodeBetterBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 511 bytes. // Maximum input 511 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeBetterBlockAsm8B(dst []byte, src []byte) int func encodeBetterBlockAsm8B(dst []byte, src []byte, tmp *[5120]byte) int
// encodeSnappyBlockAsm encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBlockAsm encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4294967295 bytes. // Maximum input 4294967295 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBlockAsm(dst []byte, src []byte) int func encodeSnappyBlockAsm(dst []byte, src []byte, tmp *[65536]byte) int
// encodeSnappyBlockAsm64K encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBlockAsm64K encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 65535 bytes. // Maximum input 65535 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBlockAsm64K(dst []byte, src []byte) int func encodeSnappyBlockAsm64K(dst []byte, src []byte, tmp *[65536]byte) int
// encodeSnappyBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 16383 bytes. // Maximum input 16383 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBlockAsm12B(dst []byte, src []byte) int func encodeSnappyBlockAsm12B(dst []byte, src []byte, tmp *[16384]byte) int
// encodeSnappyBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4095 bytes. // Maximum input 4095 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBlockAsm10B(dst []byte, src []byte) int func encodeSnappyBlockAsm10B(dst []byte, src []byte, tmp *[4096]byte) int
// encodeSnappyBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 511 bytes. // Maximum input 511 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBlockAsm8B(dst []byte, src []byte) int func encodeSnappyBlockAsm8B(dst []byte, src []byte, tmp *[1024]byte) int
// encodeSnappyBetterBlockAsm encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBetterBlockAsm encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4294967295 bytes. // Maximum input 4294967295 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBetterBlockAsm(dst []byte, src []byte) int func encodeSnappyBetterBlockAsm(dst []byte, src []byte, tmp *[589824]byte) int
// encodeSnappyBetterBlockAsm64K encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBetterBlockAsm64K encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 65535 bytes. // Maximum input 65535 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBetterBlockAsm64K(dst []byte, src []byte) int func encodeSnappyBetterBlockAsm64K(dst []byte, src []byte, tmp *[294912]byte) int
// encodeSnappyBetterBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBetterBlockAsm12B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 16383 bytes. // Maximum input 16383 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBetterBlockAsm12B(dst []byte, src []byte) int func encodeSnappyBetterBlockAsm12B(dst []byte, src []byte, tmp *[81920]byte) int
// encodeSnappyBetterBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBetterBlockAsm10B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4095 bytes. // Maximum input 4095 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBetterBlockAsm10B(dst []byte, src []byte) int func encodeSnappyBetterBlockAsm10B(dst []byte, src []byte, tmp *[20480]byte) int
// encodeSnappyBetterBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst. // encodeSnappyBetterBlockAsm8B encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 511 bytes. // Maximum input 511 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func encodeSnappyBetterBlockAsm8B(dst []byte, src []byte) int func encodeSnappyBetterBlockAsm8B(dst []byte, src []byte, tmp *[5120]byte) int
// calcBlockSize encodes a non-empty src to a guaranteed-large-enough dst. // calcBlockSize encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 4294967295 bytes. // Maximum input 4294967295 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func calcBlockSize(src []byte) int func calcBlockSize(src []byte, tmp *[32768]byte) int
// calcBlockSizeSmall encodes a non-empty src to a guaranteed-large-enough dst. // calcBlockSizeSmall encodes a non-empty src to a guaranteed-large-enough dst.
// Maximum input 1024 bytes. // Maximum input 1024 bytes.
// It assumes that the varint-encoded length of the decompressed bytes has already been written. // It assumes that the varint-encoded length of the decompressed bytes has already been written.
// //
//go:noescape //go:noescape
func calcBlockSizeSmall(src []byte) int func calcBlockSizeSmall(src []byte, tmp *[2048]byte) int
// emitLiteral writes a literal chunk and returns the number of bytes written. // emitLiteral writes a literal chunk and returns the number of bytes written.
// //

File diff suppressed because it is too large Load diff

View file

@ -83,11 +83,14 @@ type Writer struct {
snappy bool snappy bool
flushOnWrite bool flushOnWrite bool
appendIndex bool appendIndex bool
bufferCB func([]byte)
level uint8 level uint8
} }
type result struct { type result struct {
b []byte b []byte
// return when writing
ret []byte
// Uncompressed start offset // Uncompressed start offset
startOffset int64 startOffset int64
} }
@ -146,6 +149,10 @@ func (w *Writer) Reset(writer io.Writer) {
for write := range toWrite { for write := range toWrite {
// Wait for the data to be available. // Wait for the data to be available.
input := <-write input := <-write
if input.ret != nil && w.bufferCB != nil {
w.bufferCB(input.ret)
input.ret = nil
}
in := input.b in := input.b
if len(in) > 0 { if len(in) > 0 {
if w.err(nil) == nil { if w.err(nil) == nil {
@ -341,7 +348,8 @@ func (w *Writer) AddSkippableBlock(id uint8, data []byte) (err error) {
// but the input buffer cannot be written to by the caller // but the input buffer cannot be written to by the caller
// until Flush or Close has been called when concurrency != 1. // until Flush or Close has been called when concurrency != 1.
// //
// If you cannot control that, use the regular Write function. // Use the WriterBufferDone to receive a callback when the buffer is done
// Processing.
// //
// Note that input is not buffered. // Note that input is not buffered.
// This means that each write will result in discrete blocks being created. // This means that each write will result in discrete blocks being created.
@ -364,6 +372,9 @@ func (w *Writer) EncodeBuffer(buf []byte) (err error) {
} }
if w.concurrency == 1 { if w.concurrency == 1 {
_, err := w.writeSync(buf) _, err := w.writeSync(buf)
if w.bufferCB != nil {
w.bufferCB(buf)
}
return err return err
} }
@ -378,7 +389,7 @@ func (w *Writer) EncodeBuffer(buf []byte) (err error) {
hWriter <- result{startOffset: w.uncompWritten, b: magicChunkBytes} hWriter <- result{startOffset: w.uncompWritten, b: magicChunkBytes}
} }
} }
orgBuf := buf
for len(buf) > 0 { for len(buf) > 0 {
// Cut input. // Cut input.
uncompressed := buf uncompressed := buf
@ -397,6 +408,9 @@ func (w *Writer) EncodeBuffer(buf []byte) (err error) {
startOffset: w.uncompWritten, startOffset: w.uncompWritten,
} }
w.uncompWritten += int64(len(uncompressed)) w.uncompWritten += int64(len(uncompressed))
if len(buf) == 0 && w.bufferCB != nil {
res.ret = orgBuf
}
go func() { go func() {
race.ReadSlice(uncompressed) race.ReadSlice(uncompressed)
@ -922,7 +936,7 @@ func WriterBetterCompression() WriterOption {
} }
// WriterBestCompression will enable better compression. // WriterBestCompression will enable better compression.
// EncodeBetter compresses better than Encode but typically with a // EncodeBest compresses better than Encode but typically with a
// big speed decrease on compression. // big speed decrease on compression.
func WriterBestCompression() WriterOption { func WriterBestCompression() WriterOption {
return func(w *Writer) error { return func(w *Writer) error {
@ -941,6 +955,17 @@ func WriterUncompressed() WriterOption {
} }
} }
// WriterBufferDone will perform a callback when EncodeBuffer has finished
// writing a buffer to the output and the buffer can safely be reused.
// If the buffer was split into several blocks, it will be sent after the last block.
// Callbacks will not be done concurrently.
func WriterBufferDone(fn func(b []byte)) WriterOption {
return func(w *Writer) error {
w.bufferCB = fn
return nil
}
}
// WriterBlockSize allows to override the default block size. // WriterBlockSize allows to override the default block size.
// Blocks will be this size or smaller. // Blocks will be this size or smaller.
// Minimum size is 4KB and maximum size is 4MB. // Minimum size is 4KB and maximum size is 4MB.

View file

@ -598,7 +598,9 @@ func (b *blockDec) prepareSequences(in []byte, hist *history) (err error) {
printf("RLE set to 0x%x, code: %v", symb, v) printf("RLE set to 0x%x, code: %v", symb, v)
} }
case compModeFSE: case compModeFSE:
println("Reading table for", tableIndex(i)) if debugDecoder {
println("Reading table for", tableIndex(i))
}
if seq.fse == nil || seq.fse.preDefined { if seq.fse == nil || seq.fse.preDefined {
seq.fse = fseDecoderPool.Get().(*fseDecoder) seq.fse = fseDecoderPool.Get().(*fseDecoder)
} }

View file

@ -179,9 +179,9 @@ func (e *betterFastEncoder) Encode(blk *blockEnc, src []byte) {
if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) { if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) {
// Consider history as well. // Consider history as well.
var seq seq var seq seq
lenght := 4 + e.matchlen(s+4+repOff, repIndex+4, src) length := 4 + e.matchlen(s+4+repOff, repIndex+4, src)
seq.matchLen = uint32(lenght - zstdMinMatch) seq.matchLen = uint32(length - zstdMinMatch)
// We might be able to match backwards. // We might be able to match backwards.
// Extend as long as we can. // Extend as long as we can.
@ -210,12 +210,12 @@ func (e *betterFastEncoder) Encode(blk *blockEnc, src []byte) {
// Index match start+1 (long) -> s - 1 // Index match start+1 (long) -> s - 1
index0 := s + repOff index0 := s + repOff
s += lenght + repOff s += length + repOff
nextEmit = s nextEmit = s
if s >= sLimit { if s >= sLimit {
if debugEncoder { if debugEncoder {
println("repeat ended", s, lenght) println("repeat ended", s, length)
} }
break encodeLoop break encodeLoop
@ -241,9 +241,9 @@ func (e *betterFastEncoder) Encode(blk *blockEnc, src []byte) {
if false && repIndex >= 0 && load6432(src, repIndex) == load6432(src, s+repOff) { if false && repIndex >= 0 && load6432(src, repIndex) == load6432(src, s+repOff) {
// Consider history as well. // Consider history as well.
var seq seq var seq seq
lenght := 8 + e.matchlen(s+8+repOff2, repIndex+8, src) length := 8 + e.matchlen(s+8+repOff2, repIndex+8, src)
seq.matchLen = uint32(lenght - zstdMinMatch) seq.matchLen = uint32(length - zstdMinMatch)
// We might be able to match backwards. // We might be able to match backwards.
// Extend as long as we can. // Extend as long as we can.
@ -270,11 +270,11 @@ func (e *betterFastEncoder) Encode(blk *blockEnc, src []byte) {
} }
blk.sequences = append(blk.sequences, seq) blk.sequences = append(blk.sequences, seq)
s += lenght + repOff2 s += length + repOff2
nextEmit = s nextEmit = s
if s >= sLimit { if s >= sLimit {
if debugEncoder { if debugEncoder {
println("repeat ended", s, lenght) println("repeat ended", s, length)
} }
break encodeLoop break encodeLoop
@ -708,9 +708,9 @@ func (e *betterFastEncoderDict) Encode(blk *blockEnc, src []byte) {
if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) { if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) {
// Consider history as well. // Consider history as well.
var seq seq var seq seq
lenght := 4 + e.matchlen(s+4+repOff, repIndex+4, src) length := 4 + e.matchlen(s+4+repOff, repIndex+4, src)
seq.matchLen = uint32(lenght - zstdMinMatch) seq.matchLen = uint32(length - zstdMinMatch)
// We might be able to match backwards. // We might be able to match backwards.
// Extend as long as we can. // Extend as long as we can.
@ -738,12 +738,12 @@ func (e *betterFastEncoderDict) Encode(blk *blockEnc, src []byte) {
blk.sequences = append(blk.sequences, seq) blk.sequences = append(blk.sequences, seq)
// Index match start+1 (long) -> s - 1 // Index match start+1 (long) -> s - 1
s += lenght + repOff s += length + repOff
nextEmit = s nextEmit = s
if s >= sLimit { if s >= sLimit {
if debugEncoder { if debugEncoder {
println("repeat ended", s, lenght) println("repeat ended", s, length)
} }
break encodeLoop break encodeLoop
@ -772,9 +772,9 @@ func (e *betterFastEncoderDict) Encode(blk *blockEnc, src []byte) {
if false && repIndex >= 0 && load6432(src, repIndex) == load6432(src, s+repOff) { if false && repIndex >= 0 && load6432(src, repIndex) == load6432(src, s+repOff) {
// Consider history as well. // Consider history as well.
var seq seq var seq seq
lenght := 8 + e.matchlen(s+8+repOff2, repIndex+8, src) length := 8 + e.matchlen(s+8+repOff2, repIndex+8, src)
seq.matchLen = uint32(lenght - zstdMinMatch) seq.matchLen = uint32(length - zstdMinMatch)
// We might be able to match backwards. // We might be able to match backwards.
// Extend as long as we can. // Extend as long as we can.
@ -801,11 +801,11 @@ func (e *betterFastEncoderDict) Encode(blk *blockEnc, src []byte) {
} }
blk.sequences = append(blk.sequences, seq) blk.sequences = append(blk.sequences, seq)
s += lenght + repOff2 s += length + repOff2
nextEmit = s nextEmit = s
if s >= sLimit { if s >= sLimit {
if debugEncoder { if debugEncoder {
println("repeat ended", s, lenght) println("repeat ended", s, length)
} }
break encodeLoop break encodeLoop

View file

@ -138,9 +138,9 @@ func (e *doubleFastEncoder) Encode(blk *blockEnc, src []byte) {
if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) { if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) {
// Consider history as well. // Consider history as well.
var seq seq var seq seq
lenght := 4 + e.matchlen(s+4+repOff, repIndex+4, src) length := 4 + e.matchlen(s+4+repOff, repIndex+4, src)
seq.matchLen = uint32(lenght - zstdMinMatch) seq.matchLen = uint32(length - zstdMinMatch)
// We might be able to match backwards. // We might be able to match backwards.
// Extend as long as we can. // Extend as long as we can.
@ -166,11 +166,11 @@ func (e *doubleFastEncoder) Encode(blk *blockEnc, src []byte) {
println("repeat sequence", seq, "next s:", s) println("repeat sequence", seq, "next s:", s)
} }
blk.sequences = append(blk.sequences, seq) blk.sequences = append(blk.sequences, seq)
s += lenght + repOff s += length + repOff
nextEmit = s nextEmit = s
if s >= sLimit { if s >= sLimit {
if debugEncoder { if debugEncoder {
println("repeat ended", s, lenght) println("repeat ended", s, length)
} }
break encodeLoop break encodeLoop
@ -798,9 +798,9 @@ func (e *doubleFastEncoderDict) Encode(blk *blockEnc, src []byte) {
if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) { if repIndex >= 0 && load3232(src, repIndex) == uint32(cv>>(repOff*8)) {
// Consider history as well. // Consider history as well.
var seq seq var seq seq
lenght := 4 + e.matchlen(s+4+repOff, repIndex+4, src) length := 4 + e.matchlen(s+4+repOff, repIndex+4, src)
seq.matchLen = uint32(lenght - zstdMinMatch) seq.matchLen = uint32(length - zstdMinMatch)
// We might be able to match backwards. // We might be able to match backwards.
// Extend as long as we can. // Extend as long as we can.
@ -826,11 +826,11 @@ func (e *doubleFastEncoderDict) Encode(blk *blockEnc, src []byte) {
println("repeat sequence", seq, "next s:", s) println("repeat sequence", seq, "next s:", s)
} }
blk.sequences = append(blk.sequences, seq) blk.sequences = append(blk.sequences, seq)
s += lenght + repOff s += length + repOff
nextEmit = s nextEmit = s
if s >= sLimit { if s >= sLimit {
if debugEncoder { if debugEncoder {
println("repeat ended", s, lenght) println("repeat ended", s, length)
} }
break encodeLoop break encodeLoop

View file

@ -6,6 +6,7 @@
import ( import (
"crypto/rand" "crypto/rand"
"errors"
"fmt" "fmt"
"io" "io"
"math" "math"
@ -149,6 +150,9 @@ func (e *Encoder) ResetContentSize(w io.Writer, size int64) {
// and write CRC if requested. // and write CRC if requested.
func (e *Encoder) Write(p []byte) (n int, err error) { func (e *Encoder) Write(p []byte) (n int, err error) {
s := &e.state s := &e.state
if s.eofWritten {
return 0, ErrEncoderClosed
}
for len(p) > 0 { for len(p) > 0 {
if len(p)+len(s.filling) < e.o.blockSize { if len(p)+len(s.filling) < e.o.blockSize {
if e.o.crc { if e.o.crc {
@ -202,7 +206,7 @@ func (e *Encoder) nextBlock(final bool) error {
return nil return nil
} }
if final && len(s.filling) > 0 { if final && len(s.filling) > 0 {
s.current = e.EncodeAll(s.filling, s.current[:0]) s.current = e.encodeAll(s.encoder, s.filling, s.current[:0])
var n2 int var n2 int
n2, s.err = s.w.Write(s.current) n2, s.err = s.w.Write(s.current)
if s.err != nil { if s.err != nil {
@ -288,6 +292,9 @@ func (e *Encoder) nextBlock(final bool) error {
s.filling, s.current, s.previous = s.previous[:0], s.filling, s.current s.filling, s.current, s.previous = s.previous[:0], s.filling, s.current
s.nInput += int64(len(s.current)) s.nInput += int64(len(s.current))
s.wg.Add(1) s.wg.Add(1)
if final {
s.eofWritten = true
}
go func(src []byte) { go func(src []byte) {
if debugEncoder { if debugEncoder {
println("Adding block,", len(src), "bytes, final:", final) println("Adding block,", len(src), "bytes, final:", final)
@ -303,9 +310,6 @@ func (e *Encoder) nextBlock(final bool) error {
blk := enc.Block() blk := enc.Block()
enc.Encode(blk, src) enc.Encode(blk, src)
blk.last = final blk.last = final
if final {
s.eofWritten = true
}
// Wait for pending writes. // Wait for pending writes.
s.wWg.Wait() s.wWg.Wait()
if s.writeErr != nil { if s.writeErr != nil {
@ -401,12 +405,20 @@ func (e *Encoder) Flush() error {
if len(s.filling) > 0 { if len(s.filling) > 0 {
err := e.nextBlock(false) err := e.nextBlock(false)
if err != nil { if err != nil {
// Ignore Flush after Close.
if errors.Is(s.err, ErrEncoderClosed) {
return nil
}
return err return err
} }
} }
s.wg.Wait() s.wg.Wait()
s.wWg.Wait() s.wWg.Wait()
if s.err != nil { if s.err != nil {
// Ignore Flush after Close.
if errors.Is(s.err, ErrEncoderClosed) {
return nil
}
return s.err return s.err
} }
return s.writeErr return s.writeErr
@ -422,6 +434,9 @@ func (e *Encoder) Close() error {
} }
err := e.nextBlock(true) err := e.nextBlock(true)
if err != nil { if err != nil {
if errors.Is(s.err, ErrEncoderClosed) {
return nil
}
return err return err
} }
if s.frameContentSize > 0 { if s.frameContentSize > 0 {
@ -459,6 +474,11 @@ func (e *Encoder) Close() error {
} }
_, s.err = s.w.Write(frame) _, s.err = s.w.Write(frame)
} }
if s.err == nil {
s.err = ErrEncoderClosed
return nil
}
return s.err return s.err
} }
@ -469,6 +489,15 @@ func (e *Encoder) Close() error {
// Data compressed with EncodeAll can be decoded with the Decoder, // Data compressed with EncodeAll can be decoded with the Decoder,
// using either a stream or DecodeAll. // using either a stream or DecodeAll.
func (e *Encoder) EncodeAll(src, dst []byte) []byte { func (e *Encoder) EncodeAll(src, dst []byte) []byte {
e.init.Do(e.initialize)
enc := <-e.encoders
defer func() {
e.encoders <- enc
}()
return e.encodeAll(enc, src, dst)
}
func (e *Encoder) encodeAll(enc encoder, src, dst []byte) []byte {
if len(src) == 0 { if len(src) == 0 {
if e.o.fullZero { if e.o.fullZero {
// Add frame header. // Add frame header.
@ -491,13 +520,7 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
} }
return dst return dst
} }
e.init.Do(e.initialize)
enc := <-e.encoders
defer func() {
// Release encoder reference to last block.
// If a non-single block is needed the encoder will reset again.
e.encoders <- enc
}()
// Use single segments when above minimum window and below window size. // Use single segments when above minimum window and below window size.
single := len(src) <= e.o.windowSize && len(src) > MinWindowSize single := len(src) <= e.o.windowSize && len(src) > MinWindowSize
if e.o.single != nil { if e.o.single != nil {

View file

@ -146,7 +146,9 @@ func (d *frameDec) reset(br byteBuffer) error {
} }
return err return err
} }
printf("raw: %x, mantissa: %d, exponent: %d\n", wd, wd&7, wd>>3) if debugDecoder {
printf("raw: %x, mantissa: %d, exponent: %d\n", wd, wd&7, wd>>3)
}
windowLog := 10 + (wd >> 3) windowLog := 10 + (wd >> 3)
windowBase := uint64(1) << windowLog windowBase := uint64(1) << windowLog
windowAdd := (windowBase / 8) * uint64(wd&0x7) windowAdd := (windowBase / 8) * uint64(wd&0x7)

View file

@ -146,7 +146,7 @@ func (s *sequenceDecs) decodeSyncSimple(hist []byte) (bool, error) {
return true, fmt.Errorf("output bigger than max block size (%d)", maxBlockSize) return true, fmt.Errorf("output bigger than max block size (%d)", maxBlockSize)
default: default:
return true, fmt.Errorf("sequenceDecs_decode returned erronous code %d", errCode) return true, fmt.Errorf("sequenceDecs_decode returned erroneous code %d", errCode)
} }
s.seqSize += ctx.litRemain s.seqSize += ctx.litRemain
@ -292,7 +292,7 @@ func (s *sequenceDecs) decode(seqs []seqVals) error {
return io.ErrUnexpectedEOF return io.ErrUnexpectedEOF
} }
return fmt.Errorf("sequenceDecs_decode_amd64 returned erronous code %d", errCode) return fmt.Errorf("sequenceDecs_decode_amd64 returned erroneous code %d", errCode)
} }
if ctx.litRemain < 0 { if ctx.litRemain < 0 {

View file

@ -1814,7 +1814,7 @@ TEXT ·sequenceDecs_decodeSync_amd64(SB), $64-32
MOVQ 40(SP), AX MOVQ 40(SP), AX
ADDQ AX, 48(SP) ADDQ AX, 48(SP)
// Calculate poiter to s.out[cap(s.out)] (a past-end pointer) // Calculate pointer to s.out[cap(s.out)] (a past-end pointer)
ADDQ R10, 32(SP) ADDQ R10, 32(SP)
// outBase += outPosition // outBase += outPosition
@ -2376,7 +2376,7 @@ TEXT ·sequenceDecs_decodeSync_bmi2(SB), $64-32
MOVQ 40(SP), CX MOVQ 40(SP), CX
ADDQ CX, 48(SP) ADDQ CX, 48(SP)
// Calculate poiter to s.out[cap(s.out)] (a past-end pointer) // Calculate pointer to s.out[cap(s.out)] (a past-end pointer)
ADDQ R9, 32(SP) ADDQ R9, 32(SP)
// outBase += outPosition // outBase += outPosition
@ -2896,7 +2896,7 @@ TEXT ·sequenceDecs_decodeSync_safe_amd64(SB), $64-32
MOVQ 40(SP), AX MOVQ 40(SP), AX
ADDQ AX, 48(SP) ADDQ AX, 48(SP)
// Calculate poiter to s.out[cap(s.out)] (a past-end pointer) // Calculate pointer to s.out[cap(s.out)] (a past-end pointer)
ADDQ R10, 32(SP) ADDQ R10, 32(SP)
// outBase += outPosition // outBase += outPosition
@ -3560,7 +3560,7 @@ TEXT ·sequenceDecs_decodeSync_safe_bmi2(SB), $64-32
MOVQ 40(SP), CX MOVQ 40(SP), CX
ADDQ CX, 48(SP) ADDQ CX, 48(SP)
// Calculate poiter to s.out[cap(s.out)] (a past-end pointer) // Calculate pointer to s.out[cap(s.out)] (a past-end pointer)
ADDQ R9, 32(SP) ADDQ R9, 32(SP)
// outBase += outPosition // outBase += outPosition

View file

@ -88,6 +88,10 @@
// Close has been called. // Close has been called.
ErrDecoderClosed = errors.New("decoder used after Close") ErrDecoderClosed = errors.New("decoder used after Close")
// ErrEncoderClosed will be returned if the Encoder was used after
// Close has been called.
ErrEncoderClosed = errors.New("encoder used after Close")
// ErrDecoderNilInput is returned when a nil Reader was provided // ErrDecoderNilInput is returned when a nil Reader was provided
// and an operation other than Reset/DecodeAll/Close was attempted. // and an operation other than Reset/DecodeAll/Close was attempted.
ErrDecoderNilInput = errors.New("nil input provided as reader") ErrDecoderNilInput = errors.New("nil input provided as reader")

View file

@ -32,6 +32,10 @@ functional-test:
@GO111MODULE=on go build -race functional_tests.go @GO111MODULE=on go build -race functional_tests.go
@SERVER_ENDPOINT=localhost:9000 ACCESS_KEY=minioadmin SECRET_KEY=minioadmin ENABLE_HTTPS=1 MINT_MODE=full ./functional_tests @SERVER_ENDPOINT=localhost:9000 ACCESS_KEY=minioadmin SECRET_KEY=minioadmin ENABLE_HTTPS=1 MINT_MODE=full ./functional_tests
functional-test-notls:
@GO111MODULE=on go build -race functional_tests.go
@SERVER_ENDPOINT=localhost:9000 ACCESS_KEY=minioadmin SECRET_KEY=minioadmin ENABLE_HTTPS=0 MINT_MODE=full ./functional_tests
clean: clean:
@echo "Cleaning up all the generated files" @echo "Cleaning up all the generated files"
@find . -name '*.test' | xargs rm -fv @find . -name '*.test' | xargs rm -fv

View file

@ -45,6 +45,8 @@
ReplicationStatusFailed ReplicationStatus = "FAILED" ReplicationStatusFailed ReplicationStatus = "FAILED"
// ReplicationStatusReplica indicates object is a replica of a source // ReplicationStatusReplica indicates object is a replica of a source
ReplicationStatusReplica ReplicationStatus = "REPLICA" ReplicationStatusReplica ReplicationStatus = "REPLICA"
// ReplicationStatusReplicaEdge indicates object is a replica of a edge source
ReplicationStatusReplicaEdge ReplicationStatus = "REPLICA-EDGE"
) )
// Empty returns true if no replication status set. // Empty returns true if no replication status set.

View file

@ -128,7 +128,7 @@ type Options struct {
// Global constants. // Global constants.
const ( const (
libraryName = "minio-go" libraryName = "minio-go"
libraryVersion = "v7.0.77" libraryVersion = "v7.0.78"
) )
// User Agent should always following the below style. // User Agent should always following the below style.

View file

@ -3565,16 +3565,10 @@ func validateObjectAttributeRequest(OA *minio.ObjectAttributes, opts *minio.Obje
} }
} }
hasFullObjectChecksum := true hasFullObjectChecksum := (OA.Checksum.ChecksumCRC32 != "" ||
if OA.Checksum.ChecksumCRC32 == "" { OA.Checksum.ChecksumCRC32C != "" ||
if OA.Checksum.ChecksumCRC32C == "" { OA.Checksum.ChecksumSHA1 != "" ||
if OA.Checksum.ChecksumSHA1 == "" { OA.Checksum.ChecksumSHA256 != "")
if OA.Checksum.ChecksumSHA256 == "" {
hasFullObjectChecksum = false
}
}
}
}
if test.HasFullChecksum { if test.HasFullChecksum {
if !hasFullObjectChecksum { if !hasFullObjectChecksum {

View file

@ -25,6 +25,7 @@
"io" "io"
"net/http" "net/http"
"net/url" "net/url"
"os"
"strconv" "strconv"
"strings" "strings"
"time" "time"
@ -85,29 +86,59 @@ type STSWebIdentity struct {
// assuming. // assuming.
RoleARN string RoleARN string
// Policy is the policy where the credentials should be limited too.
Policy string
// roleSessionName is the identifier for the assumed role session. // roleSessionName is the identifier for the assumed role session.
roleSessionName string roleSessionName string
} }
// NewSTSWebIdentity returns a pointer to a new // NewSTSWebIdentity returns a pointer to a new
// Credentials object wrapping the STSWebIdentity. // Credentials object wrapping the STSWebIdentity.
func NewSTSWebIdentity(stsEndpoint string, getWebIDTokenExpiry func() (*WebIdentityToken, error)) (*Credentials, error) { func NewSTSWebIdentity(stsEndpoint string, getWebIDTokenExpiry func() (*WebIdentityToken, error), opts ...func(*STSWebIdentity)) (*Credentials, error) {
if stsEndpoint == "" { if stsEndpoint == "" {
return nil, errors.New("STS endpoint cannot be empty") return nil, errors.New("STS endpoint cannot be empty")
} }
if getWebIDTokenExpiry == nil { if getWebIDTokenExpiry == nil {
return nil, errors.New("Web ID token and expiry retrieval function should be defined") return nil, errors.New("Web ID token and expiry retrieval function should be defined")
} }
return New(&STSWebIdentity{ i := &STSWebIdentity{
Client: &http.Client{ Client: &http.Client{
Transport: http.DefaultTransport, Transport: http.DefaultTransport,
}, },
STSEndpoint: stsEndpoint, STSEndpoint: stsEndpoint,
GetWebIDTokenExpiry: getWebIDTokenExpiry, GetWebIDTokenExpiry: getWebIDTokenExpiry,
}), nil }
for _, o := range opts {
o(i)
}
return New(i), nil
} }
func getWebIdentityCredentials(clnt *http.Client, endpoint, roleARN, roleSessionName string, // NewKubernetesIdentity returns a pointer to a new
// Credentials object using the Kubernetes service account
func NewKubernetesIdentity(stsEndpoint string, opts ...func(*STSWebIdentity)) (*Credentials, error) {
return NewSTSWebIdentity(stsEndpoint, func() (*WebIdentityToken, error) {
token, err := os.ReadFile("/var/run/secrets/kubernetes.io/serviceaccount/token")
if err != nil {
return nil, err
}
return &WebIdentityToken{
Token: string(token),
}, nil
}, opts...)
}
// WithPolicy option will enforce that the returned credentials
// will be scoped down to the specified policy
func WithPolicy(policy string) func(*STSWebIdentity) {
return func(i *STSWebIdentity) {
i.Policy = policy
}
}
func getWebIdentityCredentials(clnt *http.Client, endpoint, roleARN, roleSessionName string, policy string,
getWebIDTokenExpiry func() (*WebIdentityToken, error), getWebIDTokenExpiry func() (*WebIdentityToken, error),
) (AssumeRoleWithWebIdentityResponse, error) { ) (AssumeRoleWithWebIdentityResponse, error) {
idToken, err := getWebIDTokenExpiry() idToken, err := getWebIDTokenExpiry()
@ -133,6 +164,9 @@ func getWebIdentityCredentials(clnt *http.Client, endpoint, roleARN, roleSession
if idToken.Expiry > 0 { if idToken.Expiry > 0 {
v.Set("DurationSeconds", fmt.Sprintf("%d", idToken.Expiry)) v.Set("DurationSeconds", fmt.Sprintf("%d", idToken.Expiry))
} }
if policy != "" {
v.Set("Policy", policy)
}
v.Set("Version", STSVersion) v.Set("Version", STSVersion)
u, err := url.Parse(endpoint) u, err := url.Parse(endpoint)
@ -183,7 +217,7 @@ func getWebIdentityCredentials(clnt *http.Client, endpoint, roleARN, roleSession
// Retrieve retrieves credentials from the MinIO service. // Retrieve retrieves credentials from the MinIO service.
// Error will be returned if the request fails. // Error will be returned if the request fails.
func (m *STSWebIdentity) Retrieve() (Value, error) { func (m *STSWebIdentity) Retrieve() (Value, error) {
a, err := getWebIdentityCredentials(m.Client, m.STSEndpoint, m.RoleARN, m.roleSessionName, m.GetWebIDTokenExpiry) a, err := getWebIdentityCredentials(m.Client, m.STSEndpoint, m.RoleARN, m.roleSessionName, m.Policy, m.GetWebIDTokenExpiry)
if err != nil { if err != nil {
return Value{}, err return Value{}, err
} }

View file

@ -69,7 +69,7 @@ func (err errTag) Error() string {
// https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/Using_Tags.html#tag-restrictions // https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/Using_Tags.html#tag-restrictions
// borrowed from this article and also testing various ASCII characters following regex // borrowed from this article and also testing various ASCII characters following regex
// is supported by AWS S3 for both tags and values. // is supported by AWS S3 for both tags and values.
var validTagKeyValue = regexp.MustCompile(`^[a-zA-Z0-9-+\-._:/@ ]+$`) var validTagKeyValue = regexp.MustCompile(`^[a-zA-Z0-9-+\-._:/@ =]+$`)
func checkKey(key string) error { func checkKey(key string) error {
if len(key) == 0 { if len(key) == 0 {

View file

@ -1,6 +1,7 @@
package ast package ast
import ( import (
"bytes"
"fmt" "fmt"
"strings" "strings"
@ -47,6 +48,15 @@ func (b *BaseBlock) SetLines(v *textm.Segments) {
b.lines = v b.lines = v
} }
// Text implements Node.Text.
func (b *BaseBlock) Text(source []byte) []byte {
var buf bytes.Buffer
for _, line := range b.Lines().Sliced(0, b.Lines().Len()) {
buf.Write(line.Value(source))
}
return buf.Bytes()
}
// A Document struct is a root node of Markdown text. // A Document struct is a root node of Markdown text.
type Document struct { type Document struct {
BaseBlock BaseBlock

View file

@ -184,11 +184,11 @@ func (b *tableParagraphTransformer) Transform(node *gast.Paragraph, reader text.
func (b *tableParagraphTransformer) parseRow(segment text.Segment, func (b *tableParagraphTransformer) parseRow(segment text.Segment,
alignments []ast.Alignment, isHeader bool, reader text.Reader, pc parser.Context) *ast.TableRow { alignments []ast.Alignment, isHeader bool, reader text.Reader, pc parser.Context) *ast.TableRow {
source := reader.Source() source := reader.Source()
segment = segment.TrimLeftSpace(source)
segment = segment.TrimRightSpace(source)
line := segment.Value(source) line := segment.Value(source)
pos := 0 pos := 0
pos += util.TrimLeftSpaceLength(line)
limit := len(line) limit := len(line)
limit -= util.TrimRightSpaceLength(line)
row := ast.NewTableRow(alignments) row := ast.NewTableRow(alignments)
if len(line) > 0 && line[pos] == '|' { if len(line) > 0 && line[pos] == '|' {
pos++ pos++

View file

@ -28,12 +28,13 @@ func (b *blockquoteParser) process(reader text.Reader) bool {
reader.Advance(pos) reader.Advance(pos)
return true return true
} }
if line[pos] == ' ' || line[pos] == '\t' {
pos++
}
reader.Advance(pos) reader.Advance(pos)
if line[pos-1] == '\t' { if line[pos] == ' ' || line[pos] == '\t' {
reader.SetPadding(2) padding := 0
if line[pos] == '\t' {
padding = util.TabWidth(reader.LineOffset()) - 1
}
reader.AdvanceAndSetPadding(1, padding)
} }
return true return true
} }

View file

@ -58,7 +58,7 @@ func (b *listItemParser) Continue(node ast.Node, reader text.Reader, pc Context)
} }
offset := lastOffset(node.Parent()) offset := lastOffset(node.Parent())
isEmpty := node.ChildCount() == 0 isEmpty := node.ChildCount() == 0 && pc.Get(emptyListItemWithBlankLines) != nil
indent, _ := util.IndentWidth(line, reader.LineOffset()) indent, _ := util.IndentWidth(line, reader.LineOffset())
if (isEmpty || indent < offset) && indent < 4 { if (isEmpty || indent < offset) && indent < 4 {
_, typ := matchesListItem(line, true) _, typ := matchesListItem(line, true)

View file

@ -878,10 +878,17 @@ func (p *parser) Parse(reader text.Reader, opts ...ParseOption) ast.Node {
blockReader := text.NewBlockReader(reader.Source(), nil) blockReader := text.NewBlockReader(reader.Source(), nil)
p.walkBlock(root, func(node ast.Node) { p.walkBlock(root, func(node ast.Node) {
p.parseBlock(blockReader, node, pc) p.parseBlock(blockReader, node, pc)
lines := node.Lines()
if lines != nil && lines.Len() != 0 {
s := lines.At(lines.Len() - 1)
s.EOB = true
lines.Set(lines.Len()-1, s)
}
}) })
for _, at := range p.astTransformers { for _, at := range p.astTransformers {
at.Transform(root, reader, pc) at.Transform(root, reader, pc)
} }
// root.Dump(reader.Source(), 0) // root.Dump(reader.Source(), 0)
return root return root
} }
@ -1256,4 +1263,5 @@ func (p *parser) parseBlock(block text.BlockReader, parent ast.Node, pc Context)
for _, ip := range p.closeBlockers { for _, ip := range p.closeBlockers {
ip.CloseBlock(parent, block, pc) ip.CloseBlock(parent, block, pc)
} }
} }

View file

@ -2,6 +2,7 @@
import ( import (
"bytes" "bytes"
"github.com/yuin/goldmark/util" "github.com/yuin/goldmark/util"
) )
@ -18,6 +19,9 @@ type Segment struct {
// Padding is a padding length of the segment. // Padding is a padding length of the segment.
Padding int Padding int
// EOB is true if the segment is end of the block.
EOB bool
} }
// NewSegment return a new Segment. // NewSegment return a new Segment.
@ -40,12 +44,18 @@ func NewSegmentPadding(start, stop, n int) Segment {
// Value returns a value of the segment. // Value returns a value of the segment.
func (t *Segment) Value(buffer []byte) []byte { func (t *Segment) Value(buffer []byte) []byte {
var result []byte
if t.Padding == 0 { if t.Padding == 0 {
return buffer[t.Start:t.Stop] result = buffer[t.Start:t.Stop]
} else {
result = make([]byte, 0, t.Padding+t.Stop-t.Start+1)
result = append(result, bytes.Repeat(space, t.Padding)...)
result = append(result, buffer[t.Start:t.Stop]...)
} }
result := make([]byte, 0, t.Padding+t.Stop-t.Start+1) if t.EOB && len(result) > 0 && result[len(result)-1] != '\n' {
result = append(result, bytes.Repeat(space, t.Padding)...) result = append(result, '\n')
return append(result, buffer[t.Start:t.Stop]...) }
return result
} }
// Len returns a length of the segment. // Len returns a length of the segment.

File diff suppressed because it is too large Load diff

View file

@ -166,7 +166,13 @@ func IndentPositionPadding(bs []byte, currentPos, paddingv, width int) (pos, pad
w := 0 w := 0
i := 0 i := 0
l := len(bs) l := len(bs)
p := paddingv
for ; i < l; i++ { for ; i < l; i++ {
if p > 0 {
p--
w++
continue
}
if bs[i] == '\t' && w < width { if bs[i] == '\t' && w < width {
w += TabWidth(currentPos + w) w += TabWidth(currentPos + w)
} else if bs[i] == ' ' && w < width { } else if bs[i] == ' ' && w < width {

View file

@ -1,5 +1,5 @@
//go:build !appengine && !js //go:build !appengine && !js && !go1.21
// +build !appengine,!js // +build !appengine,!js,!go1.21
package util package util

View file

@ -0,0 +1,18 @@
//go:build !appengine && !js && go1.21
// +build !appengine,!js,go1.21
package util
import (
"unsafe"
)
// BytesToReadOnlyString returns a string converted from given bytes.
func BytesToReadOnlyString(b []byte) string {
return unsafe.String(unsafe.SliceData(b), len(b))
}
// StringToReadOnlyBytes returns bytes converted from given string.
func StringToReadOnlyBytes(s string) []byte {
return unsafe.Slice(unsafe.StringData(s), len(s))
}

10
vendor/modules.txt vendored
View file

@ -441,8 +441,8 @@ github.com/json-iterator/go
# github.com/k3a/html2text v1.2.1 # github.com/k3a/html2text v1.2.1
## explicit; go 1.16 ## explicit; go 1.16
github.com/k3a/html2text github.com/k3a/html2text
# github.com/klauspost/compress v1.17.9 # github.com/klauspost/compress v1.17.11
## explicit; go 1.20 ## explicit; go 1.21
github.com/klauspost/compress github.com/klauspost/compress
github.com/klauspost/compress/fse github.com/klauspost/compress/fse
github.com/klauspost/compress/huff0 github.com/klauspost/compress/huff0
@ -486,8 +486,8 @@ github.com/miekg/dns
# github.com/minio/md5-simd v1.1.2 # github.com/minio/md5-simd v1.1.2
## explicit; go 1.14 ## explicit; go 1.14
github.com/minio/md5-simd github.com/minio/md5-simd
# github.com/minio/minio-go/v7 v7.0.77 # github.com/minio/minio-go/v7 v7.0.78
## explicit; go 1.21 ## explicit; go 1.22
github.com/minio/minio-go/v7 github.com/minio/minio-go/v7
github.com/minio/minio-go/v7/pkg/cors github.com/minio/minio-go/v7/pkg/cors
github.com/minio/minio-go/v7/pkg/credentials github.com/minio/minio-go/v7/pkg/credentials
@ -954,7 +954,7 @@ github.com/vmihailenco/tagparser/v2/internal/parser
# github.com/wagslane/go-password-validator v0.3.0 # github.com/wagslane/go-password-validator v0.3.0
## explicit; go 1.16 ## explicit; go 1.16
github.com/wagslane/go-password-validator github.com/wagslane/go-password-validator
# github.com/yuin/goldmark v1.7.4 # github.com/yuin/goldmark v1.7.6
## explicit; go 1.19 ## explicit; go 1.19
github.com/yuin/goldmark github.com/yuin/goldmark
github.com/yuin/goldmark/ast github.com/yuin/goldmark/ast