about summary refs log tree commit diff
path: root/src/lib/strings
diff options
context:
space:
mode:
Diffstat (limited to 'src/lib/strings')
-rw-r--r--src/lib/strings/rich-text-detection.ts59
-rw-r--r--src/lib/strings/rich-text-sanitize.ts32
-rw-r--r--src/lib/strings/rich-text.ts216
3 files changed, 0 insertions, 307 deletions
diff --git a/src/lib/strings/rich-text-detection.ts b/src/lib/strings/rich-text-detection.ts
index 386ed48e1..51d09ec5d 100644
--- a/src/lib/strings/rich-text-detection.ts
+++ b/src/lib/strings/rich-text-detection.ts
@@ -1,64 +1,5 @@
-import {AppBskyFeedPost} from '@atproto/api'
-type Entity = AppBskyFeedPost.Entity
 import {isValidDomain} from './url-helpers'
 
-export function extractEntities(
-  text: string,
-  knownHandles?: Set<string>,
-): Entity[] | undefined {
-  let match
-  let ents: Entity[] = []
-  {
-    // mentions
-    const re = /(^|\s|\()(@)([a-zA-Z0-9.-]+)(\b)/g
-    while ((match = re.exec(text))) {
-      if (knownHandles && !knownHandles.has(match[3])) {
-        continue // not a known handle
-      } else if (!match[3].includes('.')) {
-        continue // probably not a handle
-      }
-      const start = text.indexOf(match[3], match.index) - 1
-      ents.push({
-        type: 'mention',
-        value: match[3],
-        index: {start, end: start + match[3].length + 1},
-      })
-    }
-  }
-  {
-    // links
-    const re =
-      /(^|\s|\()((https?:\/\/[\S]+)|((?<domain>[a-z][a-z0-9]*(\.[a-z0-9]+)+)[\S]*))/gim
-    while ((match = re.exec(text))) {
-      let value = match[2]
-      if (!value.startsWith('http')) {
-        const domain = match.groups?.domain
-        if (!domain || !isValidDomain(domain)) {
-          continue
-        }
-        value = `https://${value}`
-      }
-      const start = text.indexOf(match[2], match.index)
-      const index = {start, end: start + match[2].length}
-      // strip ending puncuation
-      if (/[.,;!?]$/.test(value)) {
-        value = value.slice(0, -1)
-        index.end--
-      }
-      if (/[)]$/.test(value) && !value.includes('(')) {
-        value = value.slice(0, -1)
-        index.end--
-      }
-      ents.push({
-        type: 'link',
-        value,
-        index,
-      })
-    }
-  }
-  return ents.length > 0 ? ents : undefined
-}
-
 interface DetectedLink {
   link: string
 }
diff --git a/src/lib/strings/rich-text-sanitize.ts b/src/lib/strings/rich-text-sanitize.ts
deleted file mode 100644
index 0b5895707..000000000
--- a/src/lib/strings/rich-text-sanitize.ts
+++ /dev/null
@@ -1,32 +0,0 @@
-import {RichText} from './rich-text'
-
-const EXCESS_SPACE_RE = /[\r\n]([\u00AD\u2060\u200D\u200C\u200B\s]*[\r\n]){2,}/
-const REPLACEMENT_STR = '\n\n'
-
-export function removeExcessNewlines(richText: RichText): RichText {
-  return clean(richText, EXCESS_SPACE_RE, REPLACEMENT_STR)
-}
-
-// TODO: check on whether this works correctly with multi-byte codepoints
-export function clean(
-  richText: RichText,
-  targetRegexp: RegExp,
-  replacementString: string,
-): RichText {
-  richText = richText.clone()
-
-  let match = richText.text.match(targetRegexp)
-  while (match && typeof match.index !== 'undefined') {
-    const oldText = richText.text
-    const removeStartIndex = match.index
-    const removeEndIndex = removeStartIndex + match[0].length
-    richText.delete(removeStartIndex, removeEndIndex)
-    if (richText.text === oldText) {
-      break // sanity check
-    }
-    richText.insert(removeStartIndex, replacementString)
-    match = richText.text.match(targetRegexp)
-  }
-
-  return richText
-}
diff --git a/src/lib/strings/rich-text.ts b/src/lib/strings/rich-text.ts
deleted file mode 100644
index 1df2144e0..000000000
--- a/src/lib/strings/rich-text.ts
+++ /dev/null
@@ -1,216 +0,0 @@
-/*
-= Rich Text Manipulation
-
-When we sanitize rich text, we have to update the entity indices as the
-text is modified. This can be modeled as inserts() and deletes() of the
-rich text string. The possible scenarios are outlined below, along with
-their expected behaviors.
-
-NOTE: Slices are start inclusive, end exclusive
-
-== richTextInsert()
-
-Target string:
-
-   0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l l o   w o r l d   // string value
-       ^-------^           // target slice {start: 2, end: 7}
-
-Scenarios:
-
-A: ^                       // insert "test" at 0
-B:        ^                // insert "test" at 4
-C:                 ^       // insert "test" at 8
-
-A = before           -> move both by num added
-B = inner            -> move end by num added
-C = after            -> noop
-
-Results:
-
-A: 0 1 2 3 4 5 6 7 8 910   // string indices
-   t e s t h e l l o   w   // string value
-               ^-------^   // target slice {start: 6, end: 11}
-
-B: 0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l l t e s t o   w   // string value
-       ^---------------^   // target slice {start: 2, end: 11}
-
-C: 0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l l o   w o t e s   // string value
-       ^-------^           // target slice {start: 2, end: 7}
-
-== richTextDelete()
-
-Target string:
-
-   0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l l o   w o r l d   // string value
-       ^-------^           // target slice {start: 2, end: 7}
-
-Scenarios:
-
-A: ^---------------^       // remove slice {start: 0, end: 9}
-B:               ^-----^   // remove slice {start: 7, end: 11}
-C:         ^-----------^   // remove slice {start: 4, end: 11}
-D:       ^-^               // remove slice {start: 3, end: 5}
-E:   ^-----^               // remove slice {start: 1, end: 5}
-F: ^-^                     // remove slice {start: 0, end: 2}
-
-A = entirely outer   -> delete slice
-B = entirely after   -> noop
-C = partially after  -> move end to remove-start
-D = entirely inner   -> move end by num removed
-E = partially before -> move start to remove-start index, move end by num removed
-F = entirely before  -> move both by num removed
-
-Results:
-
-A: 0 1 2 3 4 5 6 7 8 910   // string indices
-   l d                     // string value
-                           // target slice (deleted)
-
-B: 0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l l o   w           // string value
-       ^-------^           // target slice {start: 2, end: 7}
-
-C: 0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l l                 // string value
-       ^-^                 // target slice {start: 2, end: 4}
-
-D: 0 1 2 3 4 5 6 7 8 910   // string indices
-   h e l   w o r l d       // string value
-       ^---^               // target slice {start: 2, end: 5}
-
-E: 0 1 2 3 4 5 6 7 8 910   // string indices
-   h   w o r l d           // string value
-     ^-^                   // target slice {start: 1, end: 3}
-
-F: 0 1 2 3 4 5 6 7 8 910   // string indices
-   l l o   w o r l d       // string value
-   ^-------^               // target slice {start: 0, end: 5}
- */
-
-import cloneDeep from 'lodash.clonedeep'
-import {AppBskyFeedPost} from '@atproto/api'
-import {removeExcessNewlines} from './rich-text-sanitize'
-
-export type Entity = AppBskyFeedPost.Entity
-export interface RichTextOpts {
-  cleanNewlines?: boolean
-}
-
-export class RichText {
-  constructor(
-    public text: string,
-    public entities?: Entity[],
-    opts?: RichTextOpts,
-  ) {
-    if (opts?.cleanNewlines) {
-      removeExcessNewlines(this).copyInto(this)
-    }
-  }
-
-  clone() {
-    return new RichText(this.text, cloneDeep(this.entities))
-  }
-
-  copyInto(target: RichText) {
-    target.text = this.text
-    target.entities = cloneDeep(this.entities)
-  }
-
-  insert(insertIndex: number, insertText: string) {
-    this.text =
-      this.text.slice(0, insertIndex) +
-      insertText +
-      this.text.slice(insertIndex)
-
-    if (!this.entities?.length) {
-      return this
-    }
-
-    const numCharsAdded = insertText.length
-    for (const ent of this.entities) {
-      // see comment at top of file for labels of each scenario
-      // scenario A (before)
-      if (insertIndex <= ent.index.start) {
-        // move both by num added
-        ent.index.start += numCharsAdded
-        ent.index.end += numCharsAdded
-      }
-      // scenario B (inner)
-      else if (insertIndex >= ent.index.start && insertIndex < ent.index.end) {
-        // move end by num added
-        ent.index.end += numCharsAdded
-      }
-      // scenario C (after)
-      // noop
-    }
-    return this
-  }
-
-  delete(removeStartIndex: number, removeEndIndex: number) {
-    this.text =
-      this.text.slice(0, removeStartIndex) + this.text.slice(removeEndIndex)
-
-    if (!this.entities?.length) {
-      return this
-    }
-
-    const numCharsRemoved = removeEndIndex - removeStartIndex
-    for (const ent of this.entities) {
-      // see comment at top of file for labels of each scenario
-      // scenario A (entirely outer)
-      if (
-        removeStartIndex <= ent.index.start &&
-        removeEndIndex >= ent.index.end
-      ) {
-        // delete slice (will get removed in final pass)
-        ent.index.start = 0
-        ent.index.end = 0
-      }
-      // scenario B (entirely after)
-      else if (removeStartIndex > ent.index.end) {
-        // noop
-      }
-      // scenario C (partially after)
-      else if (
-        removeStartIndex > ent.index.start &&
-        removeStartIndex <= ent.index.end &&
-        removeEndIndex > ent.index.end
-      ) {
-        // move end to remove start
-        ent.index.end = removeStartIndex
-      }
-      // scenario D (entirely inner)
-      else if (
-        removeStartIndex >= ent.index.start &&
-        removeEndIndex <= ent.index.end
-      ) {
-        // move end by num removed
-        ent.index.end -= numCharsRemoved
-      }
-      // scenario E (partially before)
-      else if (
-        removeStartIndex < ent.index.start &&
-        removeEndIndex >= ent.index.start &&
-        removeEndIndex <= ent.index.end
-      ) {
-        // move start to remove-start index, move end by num removed
-        ent.index.start = removeStartIndex
-        ent.index.end -= numCharsRemoved
-      }
-      // scenario F (entirely before)
-      else if (removeEndIndex < ent.index.start) {
-        // move both by num removed
-        ent.index.start -= numCharsRemoved
-        ent.index.end -= numCharsRemoved
-      }
-    }
-
-    // filter out any entities that were made irrelevant
-    this.entities = this.entities.filter(ent => ent.index.start < ent.index.end)
-    return this
-  }
-}
generated by cgit-pink 1.4.1 (git 2.36.1) at 2025-10-11 14:40:46 +0000