gotosocial/internal/regexes/regexes.go

180 lines
7.8 KiB
Go
Raw Normal View History

// 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/>.
2021-09-01 16:29:25 +00:00
package regexes
import (
[bugfix] Fix existing bio text showing as HTML (#531) * fix existing bio text showing as HTML - updated replaced mentions to include instance - strips HTML from account source note in Verify handler - update text formatter to use buffers for string writes Signed-off-by: kim <grufwub@gmail.com> * go away linter Signed-off-by: kim <grufwub@gmail.com> * change buf reset location, change html mention tags Signed-off-by: kim <grufwub@gmail.com> * reduce FindLinks code complexity Signed-off-by: kim <grufwub@gmail.com> * fix HTML to text conversion Signed-off-by: kim <grufwub@gmail.com> * Update internal/regexes/regexes.go Co-authored-by: Mina Galić <mina.galic@puppet.com> * use improved html2text lib with more options Signed-off-by: kim <grufwub@gmail.com> * fix to produce actual plaintext from html Signed-off-by: kim <grufwub@gmail.com> * fix span tags instead written as space Signed-off-by: kim <grufwub@gmail.com> * performance improvements to regex replacements, fix link replace logic for un-html-ing in the future Signed-off-by: kim <grufwub@gmail.com> * fix tag/mention replacements to use input string, fix link replace to not include scheme Signed-off-by: kim <grufwub@gmail.com> * use matched input string for link replace href text Signed-off-by: kim <grufwub@gmail.com> * remove unused code (to appease linter :sobs:) Signed-off-by: kim <grufwub@gmail.com> * improve hashtagFinger regex to be more compliant Signed-off-by: kim <grufwub@gmail.com> * update breakReplacer to include both unix and windows line endings Signed-off-by: kim <grufwub@gmail.com> * add NoteRaw field to Account to store plaintext account bio, add migration for this, set for sensitive accounts Signed-off-by: kim <grufwub@gmail.com> * drop unnecessary code Signed-off-by: kim <grufwub@gmail.com> * update text package tests to fix logic changes Signed-off-by: kim <grufwub@gmail.com> * add raw note content testing to account update and account verify Signed-off-by: kim <grufwub@gmail.com> * remove unused modules Signed-off-by: kim <grufwub@gmail.com> * fix emoji regex Signed-off-by: kim <grufwub@gmail.com> * fix replacement of hashtags Signed-off-by: kim <grufwub@gmail.com> * update code comment Signed-off-by: kim <grufwub@gmail.com> Co-authored-by: Mina Galić <mina.galic@puppet.com>
2022-05-07 15:55:27 +00:00
"bytes"
2021-09-01 16:29:25 +00:00
"fmt"
"regexp"
[bugfix] Fix existing bio text showing as HTML (#531) * fix existing bio text showing as HTML - updated replaced mentions to include instance - strips HTML from account source note in Verify handler - update text formatter to use buffers for string writes Signed-off-by: kim <grufwub@gmail.com> * go away linter Signed-off-by: kim <grufwub@gmail.com> * change buf reset location, change html mention tags Signed-off-by: kim <grufwub@gmail.com> * reduce FindLinks code complexity Signed-off-by: kim <grufwub@gmail.com> * fix HTML to text conversion Signed-off-by: kim <grufwub@gmail.com> * Update internal/regexes/regexes.go Co-authored-by: Mina Galić <mina.galic@puppet.com> * use improved html2text lib with more options Signed-off-by: kim <grufwub@gmail.com> * fix to produce actual plaintext from html Signed-off-by: kim <grufwub@gmail.com> * fix span tags instead written as space Signed-off-by: kim <grufwub@gmail.com> * performance improvements to regex replacements, fix link replace logic for un-html-ing in the future Signed-off-by: kim <grufwub@gmail.com> * fix tag/mention replacements to use input string, fix link replace to not include scheme Signed-off-by: kim <grufwub@gmail.com> * use matched input string for link replace href text Signed-off-by: kim <grufwub@gmail.com> * remove unused code (to appease linter :sobs:) Signed-off-by: kim <grufwub@gmail.com> * improve hashtagFinger regex to be more compliant Signed-off-by: kim <grufwub@gmail.com> * update breakReplacer to include both unix and windows line endings Signed-off-by: kim <grufwub@gmail.com> * add NoteRaw field to Account to store plaintext account bio, add migration for this, set for sensitive accounts Signed-off-by: kim <grufwub@gmail.com> * drop unnecessary code Signed-off-by: kim <grufwub@gmail.com> * update text package tests to fix logic changes Signed-off-by: kim <grufwub@gmail.com> * add raw note content testing to account update and account verify Signed-off-by: kim <grufwub@gmail.com> * remove unused modules Signed-off-by: kim <grufwub@gmail.com> * fix emoji regex Signed-off-by: kim <grufwub@gmail.com> * fix replacement of hashtags Signed-off-by: kim <grufwub@gmail.com> * update code comment Signed-off-by: kim <grufwub@gmail.com> Co-authored-by: Mina Galić <mina.galic@puppet.com>
2022-05-07 15:55:27 +00:00
"sync"
"mvdan.cc/xurls/v2"
2021-09-01 16:29:25 +00:00
)
const (
2021-09-02 10:24:18 +00:00
users = "users"
actors = "actors"
statuses = "statuses"
inbox = "inbox"
outbox = "outbox"
followers = "followers"
following = "following"
liked = "liked"
publicKey = "main-key"
follow = "follow"
blocks = "blocks"
reports = "reports"
2021-09-01 16:29:25 +00:00
)
const (
maximumUsernameLength = 64
maximumEmojiShortcodeLength = 30
)
var (
[bugfix] Fix existing bio text showing as HTML (#531) * fix existing bio text showing as HTML - updated replaced mentions to include instance - strips HTML from account source note in Verify handler - update text formatter to use buffers for string writes Signed-off-by: kim <grufwub@gmail.com> * go away linter Signed-off-by: kim <grufwub@gmail.com> * change buf reset location, change html mention tags Signed-off-by: kim <grufwub@gmail.com> * reduce FindLinks code complexity Signed-off-by: kim <grufwub@gmail.com> * fix HTML to text conversion Signed-off-by: kim <grufwub@gmail.com> * Update internal/regexes/regexes.go Co-authored-by: Mina Galić <mina.galic@puppet.com> * use improved html2text lib with more options Signed-off-by: kim <grufwub@gmail.com> * fix to produce actual plaintext from html Signed-off-by: kim <grufwub@gmail.com> * fix span tags instead written as space Signed-off-by: kim <grufwub@gmail.com> * performance improvements to regex replacements, fix link replace logic for un-html-ing in the future Signed-off-by: kim <grufwub@gmail.com> * fix tag/mention replacements to use input string, fix link replace to not include scheme Signed-off-by: kim <grufwub@gmail.com> * use matched input string for link replace href text Signed-off-by: kim <grufwub@gmail.com> * remove unused code (to appease linter :sobs:) Signed-off-by: kim <grufwub@gmail.com> * improve hashtagFinger regex to be more compliant Signed-off-by: kim <grufwub@gmail.com> * update breakReplacer to include both unix and windows line endings Signed-off-by: kim <grufwub@gmail.com> * add NoteRaw field to Account to store plaintext account bio, add migration for this, set for sensitive accounts Signed-off-by: kim <grufwub@gmail.com> * drop unnecessary code Signed-off-by: kim <grufwub@gmail.com> * update text package tests to fix logic changes Signed-off-by: kim <grufwub@gmail.com> * add raw note content testing to account update and account verify Signed-off-by: kim <grufwub@gmail.com> * remove unused modules Signed-off-by: kim <grufwub@gmail.com> * fix emoji regex Signed-off-by: kim <grufwub@gmail.com> * fix replacement of hashtags Signed-off-by: kim <grufwub@gmail.com> * update code comment Signed-off-by: kim <grufwub@gmail.com> Co-authored-by: Mina Galić <mina.galic@puppet.com>
2022-05-07 15:55:27 +00:00
schemes = `(http|https)://`
// LinkScheme captures http/https schemes in URLs.
LinkScheme = func() *regexp.Regexp {
rgx, err := xurls.StrictMatchingScheme(schemes)
if err != nil {
panic(err)
}
return rgx
}()
mentionName = `^@([\w\-\.]+)(?:@([\w\-\.:]+))?$`
2021-09-01 16:29:25 +00:00
// MentionName captures the username and domain part from a mention string
// such as @whatever_user@example.org, returning whatever_user and example.org (without the @ symbols)
MentionName = regexp.MustCompile(mentionName)
// mention regex can be played around with here: https://regex101.com/r/P0vpYG/1
mentionFinder = `(?:^|\s)(@\w+(?:@[a-zA-Z0-9_\-\.]+)?)`
2021-09-01 16:29:25 +00:00
// MentionFinder extracts mentions from a piece of text.
MentionFinder = regexp.MustCompile(mentionFinder)
emojiShortcode = fmt.Sprintf(`\w{2,%d}`, maximumEmojiShortcodeLength)
// EmojiShortcode validates an emoji name.
EmojiShortcode = regexp.MustCompile(fmt.Sprintf("^%s$", emojiShortcode))
// emoji regex can be played with here: https://regex101.com/r/478XGM/1
[bugfix] Fix existing bio text showing as HTML (#531) * fix existing bio text showing as HTML - updated replaced mentions to include instance - strips HTML from account source note in Verify handler - update text formatter to use buffers for string writes Signed-off-by: kim <grufwub@gmail.com> * go away linter Signed-off-by: kim <grufwub@gmail.com> * change buf reset location, change html mention tags Signed-off-by: kim <grufwub@gmail.com> * reduce FindLinks code complexity Signed-off-by: kim <grufwub@gmail.com> * fix HTML to text conversion Signed-off-by: kim <grufwub@gmail.com> * Update internal/regexes/regexes.go Co-authored-by: Mina Galić <mina.galic@puppet.com> * use improved html2text lib with more options Signed-off-by: kim <grufwub@gmail.com> * fix to produce actual plaintext from html Signed-off-by: kim <grufwub@gmail.com> * fix span tags instead written as space Signed-off-by: kim <grufwub@gmail.com> * performance improvements to regex replacements, fix link replace logic for un-html-ing in the future Signed-off-by: kim <grufwub@gmail.com> * fix tag/mention replacements to use input string, fix link replace to not include scheme Signed-off-by: kim <grufwub@gmail.com> * use matched input string for link replace href text Signed-off-by: kim <grufwub@gmail.com> * remove unused code (to appease linter :sobs:) Signed-off-by: kim <grufwub@gmail.com> * improve hashtagFinger regex to be more compliant Signed-off-by: kim <grufwub@gmail.com> * update breakReplacer to include both unix and windows line endings Signed-off-by: kim <grufwub@gmail.com> * add NoteRaw field to Account to store plaintext account bio, add migration for this, set for sensitive accounts Signed-off-by: kim <grufwub@gmail.com> * drop unnecessary code Signed-off-by: kim <grufwub@gmail.com> * update text package tests to fix logic changes Signed-off-by: kim <grufwub@gmail.com> * add raw note content testing to account update and account verify Signed-off-by: kim <grufwub@gmail.com> * remove unused modules Signed-off-by: kim <grufwub@gmail.com> * fix emoji regex Signed-off-by: kim <grufwub@gmail.com> * fix replacement of hashtags Signed-off-by: kim <grufwub@gmail.com> * update code comment Signed-off-by: kim <grufwub@gmail.com> Co-authored-by: Mina Galić <mina.galic@puppet.com>
2022-05-07 15:55:27 +00:00
emojiFinderString = fmt.Sprintf(`(?:\b)?:(%s):(?:\b)?`, emojiShortcode)
2021-09-01 16:29:25 +00:00
// EmojiFinder extracts emoji strings from a piece of text.
EmojiFinder = regexp.MustCompile(emojiFinderString)
// usernameString defines an acceptable username for a new account on this instance
2021-09-01 16:29:25 +00:00
usernameString = fmt.Sprintf(`[a-z0-9_]{2,%d}`, maximumUsernameLength)
// Username can be used to validate usernames of new signups
Username = regexp.MustCompile(fmt.Sprintf(`^%s$`, usernameString))
// usernameStringRelaxed is like usernameString, but also allows the '.' character,
// so it can also be used to match the instance account, which will have a username
// like 'example.org', and it has no upper length limit, so will work for long domains.
usernameStringRelaxed = `[a-z0-9_\.]{2,}`
userPathString = fmt.Sprintf(`^/?%s/(%s)$`, users, usernameStringRelaxed)
2021-09-01 16:29:25 +00:00
// UserPath parses a path that validates and captures the username part from eg /users/example_username
UserPath = regexp.MustCompile(userPathString)
publicKeyPath = fmt.Sprintf(`^/?%s/(%s)/%s`, users, usernameStringRelaxed, publicKey)
2021-09-01 16:29:25 +00:00
// PublicKeyPath parses a path that validates and captures the username part from eg /users/example_username/main-key
PublicKeyPath = regexp.MustCompile(publicKeyPath)
inboxPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameStringRelaxed, inbox)
2021-09-01 16:29:25 +00:00
// InboxPath parses a path that validates and captures the username part from eg /users/example_username/inbox
InboxPath = regexp.MustCompile(inboxPath)
outboxPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameStringRelaxed, outbox)
2021-09-01 16:29:25 +00:00
// OutboxPath parses a path that validates and captures the username part from eg /users/example_username/outbox
OutboxPath = regexp.MustCompile(outboxPath)
actorPath = fmt.Sprintf(`^/?%s/(%s)$`, actors, usernameStringRelaxed)
2021-09-01 16:29:25 +00:00
// ActorPath parses a path that validates and captures the username part from eg /actors/example_username
ActorPath = regexp.MustCompile(actorPath)
followersPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameStringRelaxed, followers)
2021-09-01 16:29:25 +00:00
// FollowersPath parses a path that validates and captures the username part from eg /users/example_username/followers
FollowersPath = regexp.MustCompile(followersPath)
followingPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameStringRelaxed, following)
2021-09-01 16:29:25 +00:00
// FollowingPath parses a path that validates and captures the username part from eg /users/example_username/following
FollowingPath = regexp.MustCompile(followingPath)
followPath = fmt.Sprintf(`^/?%s/(%s)/%s/(%s)$`, users, usernameStringRelaxed, follow, ulid)
2021-09-01 16:29:25 +00:00
// FollowPath parses a path that validates and captures the username part and the ulid part
// from eg /users/example_username/follow/01F7XT5JZW1WMVSW1KADS8PVDH
FollowPath = regexp.MustCompile(followPath)
ulid = `[0123456789ABCDEFGHJKMNPQRSTVWXYZ]{26}`
// ULID parses and validate a ULID.
ULID = regexp.MustCompile(fmt.Sprintf(`^%s$`, ulid))
likedPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameStringRelaxed, liked)
2021-09-01 16:29:25 +00:00
// LikedPath parses a path that validates and captures the username part from eg /users/example_username/liked
LikedPath = regexp.MustCompile(likedPath)
likePath = fmt.Sprintf(`^/?%s/(%s)/%s/(%s)$`, users, usernameStringRelaxed, liked, ulid)
2021-09-01 16:29:25 +00:00
// LikePath parses a path that validates and captures the username part and the ulid part
// from eg /users/example_username/like/01F7XT5JZW1WMVSW1KADS8PVDH
LikePath = regexp.MustCompile(likePath)
statusesPath = fmt.Sprintf(`^/?%s/(%s)/%s/(%s)$`, users, usernameStringRelaxed, statuses, ulid)
2021-09-01 16:29:25 +00:00
// StatusesPath parses a path that validates and captures the username part and the ulid part
// from eg /users/example_username/statuses/01F7XT5JZW1WMVSW1KADS8PVDH
// The regex can be played with here: https://regex101.com/r/G9zuxQ/1
StatusesPath = regexp.MustCompile(statusesPath)
blockPath = fmt.Sprintf(`^/?%s/(%s)/%s/(%s)$`, users, usernameStringRelaxed, blocks, ulid)
2021-09-01 16:29:25 +00:00
// BlockPath parses a path that validates and captures the username part and the ulid part
// from eg /users/example_username/blocks/01F7XT5JZW1WMVSW1KADS8PVDH
BlockPath = regexp.MustCompile(blockPath)
reportPath = fmt.Sprintf(`^/?%s/(%s)$`, reports, ulid)
// ReportPath parses a path that validates and captures the ulid part
// from eg /reports/01GP3AWY4CRDVRNZKW0TEAMB5R
ReportPath = regexp.MustCompile(reportPath)
filePath = fmt.Sprintf(`^(%s)/([a-z]+)/([a-z]+)/(%s)\.([a-z]+)$`, ulid, ulid)
// FilePath parses a file storage path of the form [ACCOUNT_ID]/[MEDIA_TYPE]/[MEDIA_SIZE]/[FILE_NAME]
// eg 01F8MH1H7YV1Z7D2C8K2730QBF/attachment/small/01F8MH8RMYQ6MSNY3JM2XT1CQ5.jpeg
// It captures the account id, media type, media size, file name, and file extension, eg
// `01F8MH1H7YV1Z7D2C8K2730QBF`, `attachment`, `small`, `01F8MH8RMYQ6MSNY3JM2XT1CQ5`, `jpeg`.
FilePath = regexp.MustCompile(filePath)
// MisskeyReportNotes captures a list of Note URIs from report content created by Misskey.
// https://regex101.com/r/EnTOBV/1
MisskeyReportNotes = regexp.MustCompile(`(?m)(?:^Note: ((?:http|https):\/\/.*)$)`)
2021-09-01 16:29:25 +00:00
)
[bugfix] Fix existing bio text showing as HTML (#531) * fix existing bio text showing as HTML - updated replaced mentions to include instance - strips HTML from account source note in Verify handler - update text formatter to use buffers for string writes Signed-off-by: kim <grufwub@gmail.com> * go away linter Signed-off-by: kim <grufwub@gmail.com> * change buf reset location, change html mention tags Signed-off-by: kim <grufwub@gmail.com> * reduce FindLinks code complexity Signed-off-by: kim <grufwub@gmail.com> * fix HTML to text conversion Signed-off-by: kim <grufwub@gmail.com> * Update internal/regexes/regexes.go Co-authored-by: Mina Galić <mina.galic@puppet.com> * use improved html2text lib with more options Signed-off-by: kim <grufwub@gmail.com> * fix to produce actual plaintext from html Signed-off-by: kim <grufwub@gmail.com> * fix span tags instead written as space Signed-off-by: kim <grufwub@gmail.com> * performance improvements to regex replacements, fix link replace logic for un-html-ing in the future Signed-off-by: kim <grufwub@gmail.com> * fix tag/mention replacements to use input string, fix link replace to not include scheme Signed-off-by: kim <grufwub@gmail.com> * use matched input string for link replace href text Signed-off-by: kim <grufwub@gmail.com> * remove unused code (to appease linter :sobs:) Signed-off-by: kim <grufwub@gmail.com> * improve hashtagFinger regex to be more compliant Signed-off-by: kim <grufwub@gmail.com> * update breakReplacer to include both unix and windows line endings Signed-off-by: kim <grufwub@gmail.com> * add NoteRaw field to Account to store plaintext account bio, add migration for this, set for sensitive accounts Signed-off-by: kim <grufwub@gmail.com> * drop unnecessary code Signed-off-by: kim <grufwub@gmail.com> * update text package tests to fix logic changes Signed-off-by: kim <grufwub@gmail.com> * add raw note content testing to account update and account verify Signed-off-by: kim <grufwub@gmail.com> * remove unused modules Signed-off-by: kim <grufwub@gmail.com> * fix emoji regex Signed-off-by: kim <grufwub@gmail.com> * fix replacement of hashtags Signed-off-by: kim <grufwub@gmail.com> * update code comment Signed-off-by: kim <grufwub@gmail.com> Co-authored-by: Mina Galić <mina.galic@puppet.com>
2022-05-07 15:55:27 +00:00
// bufpool is a memory pool of byte buffers for use in our regex utility functions.
var bufpool = sync.Pool{
New: func() any {
buf := bytes.NewBuffer(make([]byte, 0, 512))
return buf
},
}
// ReplaceAllStringFunc will call through to .ReplaceAllStringFunc in the provided regex, but provide you a clean byte buffer for optimized string writes.
func ReplaceAllStringFunc(rgx *regexp.Regexp, src string, repl func(match string, buf *bytes.Buffer) string) string {
buf := bufpool.Get().(*bytes.Buffer) //nolint
defer bufpool.Put(buf)
return rgx.ReplaceAllStringFunc(src, func(match string) string {
buf.Reset() // reset use
return repl(match, buf)
})
}