@typed/template 0.2.0 → 0.3.0

package/src/internal/tokenizer.ts

@@ -1,338 +0,0 @@
-import * as Token from "../Token.js"
-import {
-  getAllTextUntilElementClose,
-  getAttributeWithoutQuotes,
-  getAttributeWithQuotes,
-  getBooleanAttribute,
-  getClosingTag,
-  getComment,
-  getOpeningTag,
-  getOpeningTagEnd,
-  getPart,
-  getSelfClosingTagEnd,
-  getTextUntilCloseBrace,
-  getTextUntilPart,
-  getWhitespace,
-  PART_REGEX,
-  PART_STRING
-} from "./chunks.js"
-import type { TextChunk } from "./chunks.js"
-
-export function tokenize(template: ReadonlyArray<string>): Iterable<Token.Token> {
-  return new Tokenizer(template)
-}
-
-function templateWithParts(template: ReadonlyArray<string>): string {
-  const length = template.length
-  const lastIndex = length - 1
-
-  let output = ""
-
-  for (let i = 0; i < length; i++) {
-    const str = template[i]
-
-    if (i === lastIndex) {
-      output += str
-    } else {
-      output += str + PART_STRING(i)
-    }
-  }
-
-  return output
-}
-
-class Tokenizer implements Iterable<Token.Token> {
-  protected input: string = ""
-  protected pos: number = 0
-  protected context: "text" | "element" | "self-closing" | "text-only" = "text"
-  protected currentTag: Stack<string> | null = null
-
-  constructor(private readonly template: ReadonlyArray<string>) {}
-
-  *[Symbol.iterator](): Generator<Token.Token> {
-    this.init()
-
-    // eslint-disable-next-line no-constant-condition
-    while (this.pos < this.input.length) {
-      switch (this.context) {
-        case "element":
-          yield* this.nextElementToken()
-          break
-        case "self-closing":
-          yield* this.nextSelfClosingToken()
-          break
-        case "text-only":
-          yield* this.nextTextOnlyToken()
-          break
-        default:
-          yield* this.nextTextToken()
-      }
-    }
-  }
-
-  private init(): void {
-    // Convert our template into a string for easier tokenization
-    // Remove any whitespace at the beginning or end of the template since we don't care about it.
-    this.input = this.template.length === 1 ? this.template[0] : templateWithParts(this.template)
-  }
-
-  private *nextTextToken(): Generator<Token.Token> {
-    const char = this.nextChar()
-    const isOpenBracket = char === "<"
-
-    let next: TextChunk | undefined
-
-    if (isOpenBracket && (next = this.chunk(getOpeningTag))) {
-      const name = next.match[2]
-
-      this.pushTag(name)
-      this.context = Token.SELF_CLOSING_TAGS.has(name)
-        ? "self-closing"
-        : Token.TEXT_ONLY_NODES_REGEX.has(name)
-        ? "text-only"
-        : "element"
-
-      yield new Token.OpeningTagToken(name)
-    } else if (isOpenBracket && (next = this.chunk(getClosingTag))) {
-      yield new Token.ClosingTagToken(next.match[3])
-      this.popTag()
-    } else if (isOpenBracket && (next = this.chunk(getComment))) {
-      yield* this.parseComment(next.match[2])
-    } else if ((next = this.chunk(getPart))) {
-      yield new Token.PartToken(parseInt(next.match[2], 10))
-    } else if ((next = this.chunk(getWhitespace))) {
-      yield new Token.TextToken(next.match[1])
-    } else if ((next = this.chunk(getTextUntilCloseBrace))) {
-      yield* parseTextAndParts(next.match[1])
-    } else {
-      yield new Token.TextToken(this.takeNextChar())
-    }
-  }
-
-  private *nextElementToken(): Generator<Token.Token> {
-    let next: TextChunk | undefined
-
-    if ((next = this.chunk(getAttributeWithQuotes))) {
-      yield* this.parseAttribute(next.match[2], next.match[4])
-    } else if ((next = this.chunk(getAttributeWithoutQuotes))) {
-      yield* this.parseAttribute(next.match[2], next.match[4])
-    } else if ((next = this.chunk(getOpeningTagEnd))) {
-      this.context = "text"
-
-      yield new Token.OpeningTagEndToken(this.currentTag!.value)
-    } else if ((next = this.chunk(getSelfClosingTagEnd))) {
-      yield new Token.OpeningTagEndToken(this.currentTag!.value, false)
-      yield new Token.ClosingTagToken(this.currentTag!.value)
-      this.popTag()
-      this.context = "text"
-    } else if ((next = this.chunk(getBooleanAttribute))) {
-      yield new Token.BooleanAttributeToken(next.match[2])
-    } else if ((next = this.chunk(getWhitespace))) {
-      // Continue
-    } else {
-      this.invalidTemplate()
-    }
-  }
-
-  private *nextSelfClosingToken(): Generator<Token.Token> {
-    let next: TextChunk | undefined
-
-    if ((next = this.chunk(getAttributeWithQuotes))) {
-      yield* this.parseAttribute(next.match[2], next.match[4])
-    } else if ((next = this.chunk(getAttributeWithoutQuotes))) {
-      yield* this.parseAttribute(next.match[2], next.match[4])
-    } else if ((next = this.chunk(getSelfClosingTagEnd))) {
-      yield new Token.OpeningTagEndToken(this.currentTag!.value, true)
-      this.popTag()
-      this.context = "text"
-    } else if ((next = this.chunk(getBooleanAttribute))) {
-      yield new Token.BooleanAttributeToken(next.match[2])
-    } else {
-      this.invalidTemplate()
-    }
-  }
-
-  private *nextTextOnlyToken(): Generator<Token.Token> {
-    let next: TextChunk | undefined
-
-    if ((next = this.chunk(getAttributeWithQuotes))) {
-      yield* this.parseAttribute(next.match[2], next.match[4])
-    } else if ((next = this.chunk(getAttributeWithoutQuotes))) {
-      yield* this.parseAttribute(next.match[2], next.match[4])
-    } else if ((next = this.chunk(getOpeningTagEnd))) {
-      yield new Token.OpeningTagEndToken(this.currentTag!.value)
-    } else if ((next = this.chunk(getClosingTag))) {
-      yield new Token.ClosingTagToken(next.match[3])
-      this.context = "text"
-    } else if ((next = this.chunk(getAllTextUntilElementClose(this.currentTag!.value)))) {
-      const text = next.match[1]
-
-      if (PART_REGEX.test(text)) {
-        yield* parseTextAndParts(text)
-      } else {
-        yield new Token.TextToken(text)
-      }
-      this.context = "text"
-    } else {
-      this.invalidTemplate()
-    }
-  }
-
-  private *parseComment(comment: string): Generator<Token.Token> {
-    const parts = Array.from(parseTextAndParts(comment))
-
-    if (parts.length === 1 && parts[0]._tag === "text") {
-      yield new Token.CommentToken(parts[0].value)
-      return
-    }
-
-    yield new Token.CommentStartToken("<!--")
-
-    for (const part of parts) {
-      yield part
-    }
-
-    yield new Token.CommentEndToken("-->")
-  }
-
-  private *parseAttribute(name: string, value: string) {
-    const parts = Array.from(parseTextAndParts(value))
-
-    if (value === "") {
-      yield new Token.BooleanAttributeToken(name)
-      return
-    }
-
-    if (parts.length === 1 && parts[0]._tag === "text") {
-      yield getAttributeToken(name, value)
-      return
-    }
-
-    yield getAttributeTokenPartial(name, "start")
-
-    for (const part of parts) {
-      yield part
-    }
-
-    yield getAttributeTokenPartial(name, "end")
-  }
-
-  private nextChar(): string {
-    return this.input.charAt(this.pos)
-  }
-
-  private takeNextChar(): string {
-    const text = this.input.substring(this.pos, this.pos + 1)
-    this.pos += 1
-
-    return text
-  }
-
-  private chunk(f: (str: string, pos: number) => TextChunk | undefined): TextChunk | undefined {
-    const chunk = f(this.input, this.pos)
-
-    if (chunk) {
-      this.pos += chunk.length
-    }
-
-    return chunk
-  }
-
-  private pushTag(name: string): void {
-    this.currentTag = new Stack(name, this.currentTag)
-  }
-
-  private popTag(): void {
-    if (this.currentTag) {
-      this.currentTag = this.currentTag.previous
-    }
-  }
-
-  protected invalidTemplate() {
-    let message = `Invalid template ${this.input}`
-    message += "\n\nFrom:\n\n"
-    message += this.input.substring(this.pos)
-
-    throw new SyntaxError(message)
-  }
-}
-
-class Stack<A> {
-  constructor(
-    readonly value: A,
-    readonly previous: Stack<A> | null
-  ) {}
-}
-
-function* parseTextAndParts(s: string): Generator<Token.TextToken | Token.PartToken> {
-  let next: TextChunk | undefined
-  let pos: number = 0
-
-  while (pos < s.length) {
-    if ((next = getPart(s, pos))) {
-      yield new Token.PartToken(parseInt(next.match[2], 10))
-      pos += next.length
-    } else if ((next = getTextUntilPart(s, pos))) {
-      yield new Token.TextToken(next.match[1])
-
-      pos += next.length
-    } else {
-      yield new Token.TextToken(s.substring(pos))
-
-      break
-    }
-  }
-}
-
-function getAttributeTokenPartial(name: string, ctx: "start" | "end") {
-  switch (name[0]) {
-    case "?":
-      return ctx === "start"
-        ? new Token.BooleanAttributeStartToken(name.substring(1))
-        : new Token.BooleanAttributeEndToken(name.substring(1))
-    case ".": {
-      const propertyName = name.substring(1)
-
-      switch (propertyName) {
-        case "data":
-          return ctx === "start"
-            ? new Token.DataAttributeStartToken()
-            : new Token.DataAttributeEndToken()
-        default:
-          return ctx === "start"
-            ? new Token.PropertyAttributeStartToken(propertyName)
-            : new Token.PropertyAttributeEndToken(propertyName)
-      }
-    }
-    case "@":
-      return ctx === "start"
-        ? new Token.EventAttributeStartToken(name.substring(1))
-        : new Token.EventAttributeEndToken(name.substring(1))
-    case "o":
-      if (name[1] === "n") {
-        return ctx === "start"
-          ? new Token.EventAttributeStartToken(name.substring(2))
-          : new Token.EventAttributeEndToken(name.substring(2))
-      }
-  }
-
-  switch (name.toLowerCase()) {
-    case "class":
-    case "classname":
-      return ctx === "start"
-        ? new Token.ClassNameAttributeStartToken()
-        : new Token.ClassNameAttributeEndToken()
-    case "ref":
-      return ctx === "start" ? new Token.RefAttributeStartToken() : new Token.RefAttributeEndToken()
-  }
-
-  return ctx === "start" ? new Token.AttributeStartToken(name) : new Token.AttributeEndToken(name)
-}
-
-function getAttributeToken(name: string, value: string) {
-  if (name[0] === "?") {
-    return new Token.BooleanAttributeToken(name.substring(1))
-  }
-
-  return new Token.AttributeToken(name, value)
-}