gotosocial/internal/regexes/regexes.go
2021-12-20 18:42:19 +01:00

136 lines
6 KiB
Go

/*
GoToSocial
Copyright (C) 2021-2022 GoToSocial Authors admin@gotosocial.org
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 regexes
import (
"fmt"
"regexp"
)
const (
users = "users"
actors = "actors"
statuses = "statuses"
inbox = "inbox"
outbox = "outbox"
followers = "followers"
following = "following"
liked = "liked"
// collections = "collections"
// featured = "featured"
publicKey = "main-key"
follow = "follow"
// update = "updates"
blocks = "blocks"
)
const (
maximumUsernameLength = 64
maximumEmojiShortcodeLength = 30
maximumHashtagLength = 30
)
var (
mentionName = `^@(\w+)(?:@([a-zA-Z0-9_\-\.:]+))?$`
// MentionName captures the username and domain part from a mention string
// such as @whatever_user@example.org, returning whatever_user and example.org (without the @ symbols)
MentionName = regexp.MustCompile(mentionName)
// mention regex can be played around with here: https://regex101.com/r/G1oGR0/1
mentionFinder = `(?:^|\s)(@\w+(?:@[a-zA-Z0-9_\-\.]+)?)`
// MentionFinder extracts mentions from a piece of text.
MentionFinder = regexp.MustCompile(mentionFinder)
// hashtag regex can be played with here: https://regex101.com/r/bPxeca/1
hashtagFinder = fmt.Sprintf(`(?:^|\n|\s)(#[a-zA-Z0-9]{1,%d})(?:\b)`, maximumHashtagLength)
// HashtagFinder finds possible hashtags in a string.
// It returns just the string part of the hashtag, not the # symbol.
HashtagFinder = regexp.MustCompile(hashtagFinder)
emojiShortcode = fmt.Sprintf(`\w{2,%d}`, maximumEmojiShortcodeLength)
// EmojiShortcode validates an emoji name.
EmojiShortcode = regexp.MustCompile(fmt.Sprintf("^%s$", emojiShortcode))
// emoji regex can be played with here: https://regex101.com/r/478XGM/1
emojiFinderString = fmt.Sprintf(`(?:\B)?:(%s):(?:\B)?`, emojiShortcode)
// EmojiFinder extracts emoji strings from a piece of text.
EmojiFinder = regexp.MustCompile(emojiFinderString)
// usernameString defines an acceptable username on this instance
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$`, usernameString))
userPathString = fmt.Sprintf(`^?/%s/(%s)$`, users, usernameString)
// 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, usernameString, publicKey)
// 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$`, users, usernameString, inbox)
// InboxPath parses a path that validates and captures the username part from eg /users/example_username/inbox
InboxPath = regexp.MustCompile(inboxPath)
outboxPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameString, outbox)
// 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)$`, actors, usernameString)
// ActorPath parses a path that validates and captures the username part from eg /actors/example_username
ActorPath = regexp.MustCompile(actorPath)
followersPath = fmt.Sprintf(`^/?%s/(%s)/%s$`, users, usernameString, followers)
// 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$`, users, usernameString, following)
// FollowingPath parses a path that validates and captures the username part from eg /users/example_username/following
FollowingPath = regexp.MustCompile(followingPath)
followPath = fmt.Sprintf(`^/?%s/(%s)/%s/(%s)$`, users, usernameString, follow, ulid)
// 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, usernameString, liked)
// 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)$`, users, usernameString, liked, ulid)
// LikePath parses a path that validates and captures the username part and the ulid part
// from eg /users/example_username/like/01F7XT5JZW1WMVSW1KADS8PVDH
LikePath = regexp.MustCompile(likePath)
statusesPath = fmt.Sprintf(`^/?%s/(%s)/%s/(%s)$`, users, usernameString, statuses, ulid)
// 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, usernameString, blocks, ulid)
// 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)
)