Erster Commit
This commit is contained in:
663
node_modules/htmlparser2/src/Parser.ts
generated
vendored
Normal file
663
node_modules/htmlparser2/src/Parser.ts
generated
vendored
Normal file
@@ -0,0 +1,663 @@
|
||||
import Tokenizer, { type Callbacks, QuoteType } from "./Tokenizer.js";
|
||||
import { fromCodePoint } from "entities/decode";
|
||||
|
||||
const formTags = new Set([
|
||||
"input",
|
||||
"option",
|
||||
"optgroup",
|
||||
"select",
|
||||
"button",
|
||||
"datalist",
|
||||
"textarea",
|
||||
]);
|
||||
const pTag = new Set(["p"]);
|
||||
const tableSectionTags = new Set(["thead", "tbody"]);
|
||||
const ddtTags = new Set(["dd", "dt"]);
|
||||
const rtpTags = new Set(["rt", "rp"]);
|
||||
|
||||
const openImpliesClose = new Map<string, Set<string>>([
|
||||
["tr", new Set(["tr", "th", "td"])],
|
||||
["th", new Set(["th"])],
|
||||
["td", new Set(["thead", "th", "td"])],
|
||||
["body", new Set(["head", "link", "script"])],
|
||||
["li", new Set(["li"])],
|
||||
["p", pTag],
|
||||
["h1", pTag],
|
||||
["h2", pTag],
|
||||
["h3", pTag],
|
||||
["h4", pTag],
|
||||
["h5", pTag],
|
||||
["h6", pTag],
|
||||
["select", formTags],
|
||||
["input", formTags],
|
||||
["output", formTags],
|
||||
["button", formTags],
|
||||
["datalist", formTags],
|
||||
["textarea", formTags],
|
||||
["option", new Set(["option"])],
|
||||
["optgroup", new Set(["optgroup", "option"])],
|
||||
["dd", ddtTags],
|
||||
["dt", ddtTags],
|
||||
["address", pTag],
|
||||
["article", pTag],
|
||||
["aside", pTag],
|
||||
["blockquote", pTag],
|
||||
["details", pTag],
|
||||
["div", pTag],
|
||||
["dl", pTag],
|
||||
["fieldset", pTag],
|
||||
["figcaption", pTag],
|
||||
["figure", pTag],
|
||||
["footer", pTag],
|
||||
["form", pTag],
|
||||
["header", pTag],
|
||||
["hr", pTag],
|
||||
["main", pTag],
|
||||
["nav", pTag],
|
||||
["ol", pTag],
|
||||
["pre", pTag],
|
||||
["section", pTag],
|
||||
["table", pTag],
|
||||
["ul", pTag],
|
||||
["rt", rtpTags],
|
||||
["rp", rtpTags],
|
||||
["tbody", tableSectionTags],
|
||||
["tfoot", tableSectionTags],
|
||||
]);
|
||||
|
||||
const voidElements = new Set([
|
||||
"area",
|
||||
"base",
|
||||
"basefont",
|
||||
"br",
|
||||
"col",
|
||||
"command",
|
||||
"embed",
|
||||
"frame",
|
||||
"hr",
|
||||
"img",
|
||||
"input",
|
||||
"isindex",
|
||||
"keygen",
|
||||
"link",
|
||||
"meta",
|
||||
"param",
|
||||
"source",
|
||||
"track",
|
||||
"wbr",
|
||||
]);
|
||||
|
||||
const foreignContextElements = new Set(["math", "svg"]);
|
||||
|
||||
const htmlIntegrationElements = new Set([
|
||||
"mi",
|
||||
"mo",
|
||||
"mn",
|
||||
"ms",
|
||||
"mtext",
|
||||
"annotation-xml",
|
||||
"foreignobject",
|
||||
"desc",
|
||||
"title",
|
||||
]);
|
||||
|
||||
export interface ParserOptions {
|
||||
/**
|
||||
* Indicates whether special tags (`<script>`, `<style>`, and `<title>`) should get special treatment
|
||||
* and if "empty" tags (eg. `<br>`) can have children. If `false`, the content of special tags
|
||||
* will be text only. For feeds and other XML content (documents that don't consist of HTML),
|
||||
* set this to `true`.
|
||||
*
|
||||
* @default false
|
||||
*/
|
||||
xmlMode?: boolean;
|
||||
|
||||
/**
|
||||
* Decode entities within the document.
|
||||
*
|
||||
* @default true
|
||||
*/
|
||||
decodeEntities?: boolean;
|
||||
|
||||
/**
|
||||
* If set to true, all tags will be lowercased.
|
||||
*
|
||||
* @default !xmlMode
|
||||
*/
|
||||
lowerCaseTags?: boolean;
|
||||
|
||||
/**
|
||||
* If set to `true`, all attribute names will be lowercased. This has noticeable impact on speed.
|
||||
*
|
||||
* @default !xmlMode
|
||||
*/
|
||||
lowerCaseAttributeNames?: boolean;
|
||||
|
||||
/**
|
||||
* If set to true, CDATA sections will be recognized as text even if the xmlMode option is not enabled.
|
||||
* NOTE: If xmlMode is set to `true` then CDATA sections will always be recognized as text.
|
||||
*
|
||||
* @default xmlMode
|
||||
*/
|
||||
recognizeCDATA?: boolean;
|
||||
|
||||
/**
|
||||
* If set to `true`, self-closing tags will trigger the onclosetag event even if xmlMode is not set to `true`.
|
||||
* NOTE: If xmlMode is set to `true` then self-closing tags will always be recognized.
|
||||
*
|
||||
* @default xmlMode
|
||||
*/
|
||||
recognizeSelfClosing?: boolean;
|
||||
|
||||
/**
|
||||
* Allows the default tokenizer to be overwritten.
|
||||
*/
|
||||
Tokenizer?: typeof Tokenizer;
|
||||
}
|
||||
|
||||
export interface Handler {
|
||||
onparserinit(parser: Parser): void;
|
||||
|
||||
/**
|
||||
* Resets the handler back to starting state
|
||||
*/
|
||||
onreset(): void;
|
||||
|
||||
/**
|
||||
* Signals the handler that parsing is done
|
||||
*/
|
||||
onend(): void;
|
||||
onerror(error: Error): void;
|
||||
onclosetag(name: string, isImplied: boolean): void;
|
||||
onopentagname(name: string): void;
|
||||
/**
|
||||
*
|
||||
* @param name Name of the attribute
|
||||
* @param value Value of the attribute.
|
||||
* @param quote Quotes used around the attribute. `null` if the attribute has no quotes around the value, `undefined` if the attribute has no value.
|
||||
*/
|
||||
onattribute(
|
||||
name: string,
|
||||
value: string,
|
||||
quote?: string | undefined | null,
|
||||
): void;
|
||||
onopentag(
|
||||
name: string,
|
||||
attribs: { [s: string]: string },
|
||||
isImplied: boolean,
|
||||
): void;
|
||||
ontext(data: string): void;
|
||||
oncomment(data: string): void;
|
||||
oncdatastart(): void;
|
||||
oncdataend(): void;
|
||||
oncommentend(): void;
|
||||
onprocessinginstruction(name: string, data: string): void;
|
||||
}
|
||||
|
||||
const reNameEnd = /\s|\//;
|
||||
|
||||
export class Parser implements Callbacks {
|
||||
/** The start index of the last event. */
|
||||
public startIndex = 0;
|
||||
/** The end index of the last event. */
|
||||
public endIndex = 0;
|
||||
/**
|
||||
* Store the start index of the current open tag,
|
||||
* so we can update the start index for attributes.
|
||||
*/
|
||||
private openTagStart = 0;
|
||||
|
||||
private tagname = "";
|
||||
private attribname = "";
|
||||
private attribvalue = "";
|
||||
private attribs: null | { [key: string]: string } = null;
|
||||
private readonly stack: string[] = [];
|
||||
/** Determines whether self-closing tags are recognized. */
|
||||
private readonly foreignContext: boolean[];
|
||||
private readonly cbs: Partial<Handler>;
|
||||
private readonly lowerCaseTagNames: boolean;
|
||||
private readonly lowerCaseAttributeNames: boolean;
|
||||
private readonly recognizeSelfClosing: boolean;
|
||||
/** We are parsing HTML. Inverse of the `xmlMode` option. */
|
||||
private readonly htmlMode: boolean;
|
||||
private readonly tokenizer: Tokenizer;
|
||||
|
||||
private readonly buffers: string[] = [];
|
||||
private bufferOffset = 0;
|
||||
/** The index of the last written buffer. Used when resuming after a `pause()`. */
|
||||
private writeIndex = 0;
|
||||
/** Indicates whether the parser has finished running / `.end` has been called. */
|
||||
private ended = false;
|
||||
|
||||
constructor(
|
||||
cbs?: Partial<Handler> | null,
|
||||
private readonly options: ParserOptions = {},
|
||||
) {
|
||||
this.cbs = cbs ?? {};
|
||||
this.htmlMode = !this.options.xmlMode;
|
||||
this.lowerCaseTagNames = options.lowerCaseTags ?? this.htmlMode;
|
||||
this.lowerCaseAttributeNames =
|
||||
options.lowerCaseAttributeNames ?? this.htmlMode;
|
||||
this.recognizeSelfClosing =
|
||||
options.recognizeSelfClosing ?? !this.htmlMode;
|
||||
this.tokenizer = new (options.Tokenizer ?? Tokenizer)(
|
||||
this.options,
|
||||
this,
|
||||
);
|
||||
this.foreignContext = [!this.htmlMode];
|
||||
this.cbs.onparserinit?.(this);
|
||||
}
|
||||
|
||||
// Tokenizer event handlers
|
||||
|
||||
/** @internal */
|
||||
ontext(start: number, endIndex: number): void {
|
||||
const data = this.getSlice(start, endIndex);
|
||||
this.endIndex = endIndex - 1;
|
||||
this.cbs.ontext?.(data);
|
||||
this.startIndex = endIndex;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
ontextentity(cp: number, endIndex: number): void {
|
||||
this.endIndex = endIndex - 1;
|
||||
this.cbs.ontext?.(fromCodePoint(cp));
|
||||
this.startIndex = endIndex;
|
||||
}
|
||||
|
||||
/**
|
||||
* Checks if the current tag is a void element. Override this if you want
|
||||
* to specify your own additional void elements.
|
||||
*/
|
||||
protected isVoidElement(name: string): boolean {
|
||||
return this.htmlMode && voidElements.has(name);
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onopentagname(start: number, endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
|
||||
let name = this.getSlice(start, endIndex);
|
||||
|
||||
if (this.lowerCaseTagNames) {
|
||||
name = name.toLowerCase();
|
||||
}
|
||||
|
||||
this.emitOpenTag(name);
|
||||
}
|
||||
|
||||
private emitOpenTag(name: string) {
|
||||
this.openTagStart = this.startIndex;
|
||||
this.tagname = name;
|
||||
|
||||
const impliesClose = this.htmlMode && openImpliesClose.get(name);
|
||||
|
||||
if (impliesClose) {
|
||||
while (this.stack.length > 0 && impliesClose.has(this.stack[0])) {
|
||||
const element = this.stack.shift()!;
|
||||
this.cbs.onclosetag?.(element, true);
|
||||
}
|
||||
}
|
||||
if (!this.isVoidElement(name)) {
|
||||
this.stack.unshift(name);
|
||||
|
||||
if (this.htmlMode) {
|
||||
if (foreignContextElements.has(name)) {
|
||||
this.foreignContext.unshift(true);
|
||||
} else if (htmlIntegrationElements.has(name)) {
|
||||
this.foreignContext.unshift(false);
|
||||
}
|
||||
}
|
||||
}
|
||||
this.cbs.onopentagname?.(name);
|
||||
if (this.cbs.onopentag) this.attribs = {};
|
||||
}
|
||||
|
||||
private endOpenTag(isImplied: boolean) {
|
||||
this.startIndex = this.openTagStart;
|
||||
|
||||
if (this.attribs) {
|
||||
this.cbs.onopentag?.(this.tagname, this.attribs, isImplied);
|
||||
this.attribs = null;
|
||||
}
|
||||
if (this.cbs.onclosetag && this.isVoidElement(this.tagname)) {
|
||||
this.cbs.onclosetag(this.tagname, true);
|
||||
}
|
||||
|
||||
this.tagname = "";
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onopentagend(endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
this.endOpenTag(false);
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onclosetag(start: number, endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
|
||||
let name = this.getSlice(start, endIndex);
|
||||
|
||||
if (this.lowerCaseTagNames) {
|
||||
name = name.toLowerCase();
|
||||
}
|
||||
|
||||
if (
|
||||
this.htmlMode &&
|
||||
(foreignContextElements.has(name) ||
|
||||
htmlIntegrationElements.has(name))
|
||||
) {
|
||||
this.foreignContext.shift();
|
||||
}
|
||||
|
||||
if (!this.isVoidElement(name)) {
|
||||
const pos = this.stack.indexOf(name);
|
||||
if (pos !== -1) {
|
||||
for (let index = 0; index <= pos; index++) {
|
||||
const element = this.stack.shift()!;
|
||||
// We know the stack has sufficient elements.
|
||||
this.cbs.onclosetag?.(element, index !== pos);
|
||||
}
|
||||
} else if (this.htmlMode && name === "p") {
|
||||
// Implicit open before close
|
||||
this.emitOpenTag("p");
|
||||
this.closeCurrentTag(true);
|
||||
}
|
||||
} else if (this.htmlMode && name === "br") {
|
||||
// We can't use `emitOpenTag` for implicit open, as `br` would be implicitly closed.
|
||||
this.cbs.onopentagname?.("br");
|
||||
this.cbs.onopentag?.("br", {}, true);
|
||||
this.cbs.onclosetag?.("br", false);
|
||||
}
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onselfclosingtag(endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
if (this.recognizeSelfClosing || this.foreignContext[0]) {
|
||||
this.closeCurrentTag(false);
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
} else {
|
||||
// Ignore the fact that the tag is self-closing.
|
||||
this.onopentagend(endIndex);
|
||||
}
|
||||
}
|
||||
|
||||
private closeCurrentTag(isOpenImplied: boolean) {
|
||||
const name = this.tagname;
|
||||
this.endOpenTag(isOpenImplied);
|
||||
|
||||
// Self-closing tags will be on the top of the stack
|
||||
if (this.stack[0] === name) {
|
||||
// If the opening tag isn't implied, the closing tag has to be implied.
|
||||
this.cbs.onclosetag?.(name, !isOpenImplied);
|
||||
this.stack.shift();
|
||||
}
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onattribname(start: number, endIndex: number): void {
|
||||
this.startIndex = start;
|
||||
const name = this.getSlice(start, endIndex);
|
||||
|
||||
this.attribname = this.lowerCaseAttributeNames
|
||||
? name.toLowerCase()
|
||||
: name;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onattribdata(start: number, endIndex: number): void {
|
||||
this.attribvalue += this.getSlice(start, endIndex);
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onattribentity(cp: number): void {
|
||||
this.attribvalue += fromCodePoint(cp);
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onattribend(quote: QuoteType, endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
|
||||
this.cbs.onattribute?.(
|
||||
this.attribname,
|
||||
this.attribvalue,
|
||||
quote === QuoteType.Double
|
||||
? '"'
|
||||
: quote === QuoteType.Single
|
||||
? "'"
|
||||
: quote === QuoteType.NoValue
|
||||
? undefined
|
||||
: null,
|
||||
);
|
||||
|
||||
if (
|
||||
this.attribs &&
|
||||
!Object.prototype.hasOwnProperty.call(this.attribs, this.attribname)
|
||||
) {
|
||||
this.attribs[this.attribname] = this.attribvalue;
|
||||
}
|
||||
this.attribvalue = "";
|
||||
}
|
||||
|
||||
private getInstructionName(value: string) {
|
||||
const index = value.search(reNameEnd);
|
||||
let name = index < 0 ? value : value.substr(0, index);
|
||||
|
||||
if (this.lowerCaseTagNames) {
|
||||
name = name.toLowerCase();
|
||||
}
|
||||
|
||||
return name;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
ondeclaration(start: number, endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
const value = this.getSlice(start, endIndex);
|
||||
|
||||
if (this.cbs.onprocessinginstruction) {
|
||||
const name = this.getInstructionName(value);
|
||||
this.cbs.onprocessinginstruction(`!${name}`, `!${value}`);
|
||||
}
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onprocessinginstruction(start: number, endIndex: number): void {
|
||||
this.endIndex = endIndex;
|
||||
const value = this.getSlice(start, endIndex);
|
||||
|
||||
if (this.cbs.onprocessinginstruction) {
|
||||
const name = this.getInstructionName(value);
|
||||
this.cbs.onprocessinginstruction(`?${name}`, `?${value}`);
|
||||
}
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
oncomment(start: number, endIndex: number, offset: number): void {
|
||||
this.endIndex = endIndex;
|
||||
|
||||
this.cbs.oncomment?.(this.getSlice(start, endIndex - offset));
|
||||
this.cbs.oncommentend?.();
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
oncdata(start: number, endIndex: number, offset: number): void {
|
||||
this.endIndex = endIndex;
|
||||
const value = this.getSlice(start, endIndex - offset);
|
||||
|
||||
if (!this.htmlMode || this.options.recognizeCDATA) {
|
||||
this.cbs.oncdatastart?.();
|
||||
this.cbs.ontext?.(value);
|
||||
this.cbs.oncdataend?.();
|
||||
} else {
|
||||
this.cbs.oncomment?.(`[CDATA[${value}]]`);
|
||||
this.cbs.oncommentend?.();
|
||||
}
|
||||
|
||||
// Set `startIndex` for next node
|
||||
this.startIndex = endIndex + 1;
|
||||
}
|
||||
|
||||
/** @internal */
|
||||
onend(): void {
|
||||
if (this.cbs.onclosetag) {
|
||||
// Set the end index for all remaining tags
|
||||
this.endIndex = this.startIndex;
|
||||
for (let index = 0; index < this.stack.length; index++) {
|
||||
this.cbs.onclosetag(this.stack[index], true);
|
||||
}
|
||||
}
|
||||
this.cbs.onend?.();
|
||||
}
|
||||
|
||||
/**
|
||||
* Resets the parser to a blank state, ready to parse a new HTML document
|
||||
*/
|
||||
public reset(): void {
|
||||
this.cbs.onreset?.();
|
||||
this.tokenizer.reset();
|
||||
this.tagname = "";
|
||||
this.attribname = "";
|
||||
this.attribs = null;
|
||||
this.stack.length = 0;
|
||||
this.startIndex = 0;
|
||||
this.endIndex = 0;
|
||||
this.cbs.onparserinit?.(this);
|
||||
this.buffers.length = 0;
|
||||
this.foreignContext.length = 0;
|
||||
this.foreignContext.unshift(!this.htmlMode);
|
||||
this.bufferOffset = 0;
|
||||
this.writeIndex = 0;
|
||||
this.ended = false;
|
||||
}
|
||||
|
||||
/**
|
||||
* Resets the parser, then parses a complete document and
|
||||
* pushes it to the handler.
|
||||
*
|
||||
* @param data Document to parse.
|
||||
*/
|
||||
public parseComplete(data: string): void {
|
||||
this.reset();
|
||||
this.end(data);
|
||||
}
|
||||
|
||||
private getSlice(start: number, end: number) {
|
||||
while (start - this.bufferOffset >= this.buffers[0].length) {
|
||||
this.shiftBuffer();
|
||||
}
|
||||
|
||||
let slice = this.buffers[0].slice(
|
||||
start - this.bufferOffset,
|
||||
end - this.bufferOffset,
|
||||
);
|
||||
|
||||
while (end - this.bufferOffset > this.buffers[0].length) {
|
||||
this.shiftBuffer();
|
||||
slice += this.buffers[0].slice(0, end - this.bufferOffset);
|
||||
}
|
||||
|
||||
return slice;
|
||||
}
|
||||
|
||||
private shiftBuffer(): void {
|
||||
this.bufferOffset += this.buffers[0].length;
|
||||
this.writeIndex--;
|
||||
this.buffers.shift();
|
||||
}
|
||||
|
||||
/**
|
||||
* Parses a chunk of data and calls the corresponding callbacks.
|
||||
*
|
||||
* @param chunk Chunk to parse.
|
||||
*/
|
||||
public write(chunk: string): void {
|
||||
if (this.ended) {
|
||||
this.cbs.onerror?.(new Error(".write() after done!"));
|
||||
return;
|
||||
}
|
||||
|
||||
this.buffers.push(chunk);
|
||||
if (this.tokenizer.running) {
|
||||
this.tokenizer.write(chunk);
|
||||
this.writeIndex++;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Parses the end of the buffer and clears the stack, calls onend.
|
||||
*
|
||||
* @param chunk Optional final chunk to parse.
|
||||
*/
|
||||
public end(chunk?: string): void {
|
||||
if (this.ended) {
|
||||
this.cbs.onerror?.(new Error(".end() after done!"));
|
||||
return;
|
||||
}
|
||||
|
||||
if (chunk) this.write(chunk);
|
||||
this.ended = true;
|
||||
this.tokenizer.end();
|
||||
}
|
||||
|
||||
/**
|
||||
* Pauses parsing. The parser won't emit events until `resume` is called.
|
||||
*/
|
||||
public pause(): void {
|
||||
this.tokenizer.pause();
|
||||
}
|
||||
|
||||
/**
|
||||
* Resumes parsing after `pause` was called.
|
||||
*/
|
||||
public resume(): void {
|
||||
this.tokenizer.resume();
|
||||
|
||||
while (
|
||||
this.tokenizer.running &&
|
||||
this.writeIndex < this.buffers.length
|
||||
) {
|
||||
this.tokenizer.write(this.buffers[this.writeIndex++]);
|
||||
}
|
||||
|
||||
if (this.ended) this.tokenizer.end();
|
||||
}
|
||||
|
||||
/**
|
||||
* Alias of `write`, for backwards compatibility.
|
||||
*
|
||||
* @param chunk Chunk to parse.
|
||||
* @deprecated
|
||||
*/
|
||||
public parseChunk(chunk: string): void {
|
||||
this.write(chunk);
|
||||
}
|
||||
/**
|
||||
* Alias of `end`, for backwards compatibility.
|
||||
*
|
||||
* @param chunk Optional final chunk to parse.
|
||||
* @deprecated
|
||||
*/
|
||||
public done(chunk?: string): void {
|
||||
this.end(chunk);
|
||||
}
|
||||
}
|
||||
877
node_modules/htmlparser2/src/Tokenizer.ts
generated
vendored
Normal file
877
node_modules/htmlparser2/src/Tokenizer.ts
generated
vendored
Normal file
@@ -0,0 +1,877 @@
|
||||
import {
|
||||
EntityDecoder,
|
||||
DecodingMode,
|
||||
htmlDecodeTree,
|
||||
xmlDecodeTree,
|
||||
} from "entities/decode";
|
||||
|
||||
const enum CharCodes {
|
||||
Tab = 0x9, // "\t"
|
||||
NewLine = 0xa, // "\n"
|
||||
FormFeed = 0xc, // "\f"
|
||||
CarriageReturn = 0xd, // "\r"
|
||||
Space = 0x20, // " "
|
||||
ExclamationMark = 0x21, // "!"
|
||||
Number = 0x23, // "#"
|
||||
Amp = 0x26, // "&"
|
||||
SingleQuote = 0x27, // "'"
|
||||
DoubleQuote = 0x22, // '"'
|
||||
Dash = 0x2d, // "-"
|
||||
Slash = 0x2f, // "/"
|
||||
Zero = 0x30, // "0"
|
||||
Nine = 0x39, // "9"
|
||||
Semi = 0x3b, // ";"
|
||||
Lt = 0x3c, // "<"
|
||||
Eq = 0x3d, // "="
|
||||
Gt = 0x3e, // ">"
|
||||
Questionmark = 0x3f, // "?"
|
||||
UpperA = 0x41, // "A"
|
||||
LowerA = 0x61, // "a"
|
||||
UpperF = 0x46, // "F"
|
||||
LowerF = 0x66, // "f"
|
||||
UpperZ = 0x5a, // "Z"
|
||||
LowerZ = 0x7a, // "z"
|
||||
LowerX = 0x78, // "x"
|
||||
OpeningSquareBracket = 0x5b, // "["
|
||||
}
|
||||
|
||||
/** All the states the tokenizer can be in. */
|
||||
const enum State {
|
||||
Text = 1,
|
||||
BeforeTagName, // After <
|
||||
InTagName,
|
||||
InSelfClosingTag,
|
||||
BeforeClosingTagName,
|
||||
InClosingTagName,
|
||||
AfterClosingTagName,
|
||||
|
||||
// Attributes
|
||||
BeforeAttributeName,
|
||||
InAttributeName,
|
||||
AfterAttributeName,
|
||||
BeforeAttributeValue,
|
||||
InAttributeValueDq, // "
|
||||
InAttributeValueSq, // '
|
||||
InAttributeValueNq,
|
||||
|
||||
// Declarations
|
||||
BeforeDeclaration, // !
|
||||
InDeclaration,
|
||||
|
||||
// Processing instructions
|
||||
InProcessingInstruction, // ?
|
||||
|
||||
// Comments & CDATA
|
||||
BeforeComment,
|
||||
CDATASequence,
|
||||
InSpecialComment,
|
||||
InCommentLike,
|
||||
|
||||
// Special tags
|
||||
BeforeSpecialS, // Decide if we deal with `<script` or `<style`
|
||||
BeforeSpecialT, // Decide if we deal with `<title` or `<textarea`
|
||||
SpecialStartSequence,
|
||||
InSpecialTag,
|
||||
|
||||
InEntity,
|
||||
}
|
||||
|
||||
function isWhitespace(c: number): boolean {
|
||||
return (
|
||||
c === CharCodes.Space ||
|
||||
c === CharCodes.NewLine ||
|
||||
c === CharCodes.Tab ||
|
||||
c === CharCodes.FormFeed ||
|
||||
c === CharCodes.CarriageReturn
|
||||
);
|
||||
}
|
||||
|
||||
function isEndOfTagSection(c: number): boolean {
|
||||
return c === CharCodes.Slash || c === CharCodes.Gt || isWhitespace(c);
|
||||
}
|
||||
|
||||
function isASCIIAlpha(c: number): boolean {
|
||||
return (
|
||||
(c >= CharCodes.LowerA && c <= CharCodes.LowerZ) ||
|
||||
(c >= CharCodes.UpperA && c <= CharCodes.UpperZ)
|
||||
);
|
||||
}
|
||||
|
||||
export enum QuoteType {
|
||||
NoValue = 0,
|
||||
Unquoted = 1,
|
||||
Single = 2,
|
||||
Double = 3,
|
||||
}
|
||||
|
||||
export interface Callbacks {
|
||||
onattribdata(start: number, endIndex: number): void;
|
||||
onattribentity(codepoint: number): void;
|
||||
onattribend(quote: QuoteType, endIndex: number): void;
|
||||
onattribname(start: number, endIndex: number): void;
|
||||
oncdata(start: number, endIndex: number, endOffset: number): void;
|
||||
onclosetag(start: number, endIndex: number): void;
|
||||
oncomment(start: number, endIndex: number, endOffset: number): void;
|
||||
ondeclaration(start: number, endIndex: number): void;
|
||||
onend(): void;
|
||||
onopentagend(endIndex: number): void;
|
||||
onopentagname(start: number, endIndex: number): void;
|
||||
onprocessinginstruction(start: number, endIndex: number): void;
|
||||
onselfclosingtag(endIndex: number): void;
|
||||
ontext(start: number, endIndex: number): void;
|
||||
ontextentity(codepoint: number, endIndex: number): void;
|
||||
}
|
||||
|
||||
/**
|
||||
* Sequences used to match longer strings.
|
||||
*
|
||||
* We don't have `Script`, `Style`, or `Title` here. Instead, we re-use the *End
|
||||
* sequences with an increased offset.
|
||||
*/
|
||||
const Sequences = {
|
||||
Cdata: new Uint8Array([0x43, 0x44, 0x41, 0x54, 0x41, 0x5b]), // CDATA[
|
||||
CdataEnd: new Uint8Array([0x5d, 0x5d, 0x3e]), // ]]>
|
||||
CommentEnd: new Uint8Array([0x2d, 0x2d, 0x3e]), // `-->`
|
||||
ScriptEnd: new Uint8Array([0x3c, 0x2f, 0x73, 0x63, 0x72, 0x69, 0x70, 0x74]), // `</script`
|
||||
StyleEnd: new Uint8Array([0x3c, 0x2f, 0x73, 0x74, 0x79, 0x6c, 0x65]), // `</style`
|
||||
TitleEnd: new Uint8Array([0x3c, 0x2f, 0x74, 0x69, 0x74, 0x6c, 0x65]), // `</title`
|
||||
TextareaEnd: new Uint8Array([
|
||||
0x3c, 0x2f, 0x74, 0x65, 0x78, 0x74, 0x61, 0x72, 0x65, 0x61,
|
||||
]), // `</textarea`
|
||||
XmpEnd: new Uint8Array([0x3c, 0x2f, 0x78, 0x6d, 0x70]), // `</xmp`
|
||||
};
|
||||
|
||||
export default class Tokenizer {
|
||||
/** The current state the tokenizer is in. */
|
||||
private state = State.Text;
|
||||
/** The read buffer. */
|
||||
private buffer = "";
|
||||
/** The beginning of the section that is currently being read. */
|
||||
private sectionStart = 0;
|
||||
/** The index within the buffer that we are currently looking at. */
|
||||
private index = 0;
|
||||
/** The start of the last entity. */
|
||||
private entityStart = 0;
|
||||
/** Some behavior, eg. when decoding entities, is done while we are in another state. This keeps track of the other state type. */
|
||||
private baseState = State.Text;
|
||||
/** For special parsing behavior inside of script and style tags. */
|
||||
private isSpecial = false;
|
||||
/** Indicates whether the tokenizer has been paused. */
|
||||
public running = true;
|
||||
/** The offset of the current buffer. */
|
||||
private offset = 0;
|
||||
|
||||
private readonly xmlMode: boolean;
|
||||
private readonly decodeEntities: boolean;
|
||||
private readonly entityDecoder: EntityDecoder;
|
||||
|
||||
constructor(
|
||||
{
|
||||
xmlMode = false,
|
||||
decodeEntities = true,
|
||||
}: { xmlMode?: boolean; decodeEntities?: boolean },
|
||||
private readonly cbs: Callbacks,
|
||||
) {
|
||||
this.xmlMode = xmlMode;
|
||||
this.decodeEntities = decodeEntities;
|
||||
this.entityDecoder = new EntityDecoder(
|
||||
xmlMode ? xmlDecodeTree : htmlDecodeTree,
|
||||
(cp, consumed) => this.emitCodePoint(cp, consumed),
|
||||
);
|
||||
}
|
||||
|
||||
public reset(): void {
|
||||
this.state = State.Text;
|
||||
this.buffer = "";
|
||||
this.sectionStart = 0;
|
||||
this.index = 0;
|
||||
this.baseState = State.Text;
|
||||
this.currentSequence = undefined!;
|
||||
this.running = true;
|
||||
this.offset = 0;
|
||||
}
|
||||
|
||||
public write(chunk: string): void {
|
||||
this.offset += this.buffer.length;
|
||||
this.buffer = chunk;
|
||||
this.parse();
|
||||
}
|
||||
|
||||
public end(): void {
|
||||
if (this.running) this.finish();
|
||||
}
|
||||
|
||||
public pause(): void {
|
||||
this.running = false;
|
||||
}
|
||||
|
||||
public resume(): void {
|
||||
this.running = true;
|
||||
if (this.index < this.buffer.length + this.offset) {
|
||||
this.parse();
|
||||
}
|
||||
}
|
||||
|
||||
private stateText(c: number): void {
|
||||
if (
|
||||
c === CharCodes.Lt ||
|
||||
(!this.decodeEntities && this.fastForwardTo(CharCodes.Lt))
|
||||
) {
|
||||
if (this.index > this.sectionStart) {
|
||||
this.cbs.ontext(this.sectionStart, this.index);
|
||||
}
|
||||
this.state = State.BeforeTagName;
|
||||
this.sectionStart = this.index;
|
||||
} else if (this.decodeEntities && c === CharCodes.Amp) {
|
||||
this.startEntity();
|
||||
}
|
||||
}
|
||||
|
||||
private currentSequence: Uint8Array = undefined!;
|
||||
private sequenceIndex = 0;
|
||||
private stateSpecialStartSequence(c: number): void {
|
||||
const isEnd = this.sequenceIndex === this.currentSequence.length;
|
||||
const isMatch = isEnd
|
||||
? // If we are at the end of the sequence, make sure the tag name has ended
|
||||
isEndOfTagSection(c)
|
||||
: // Otherwise, do a case-insensitive comparison
|
||||
(c | 0x20) === this.currentSequence[this.sequenceIndex];
|
||||
|
||||
if (!isMatch) {
|
||||
this.isSpecial = false;
|
||||
} else if (!isEnd) {
|
||||
this.sequenceIndex++;
|
||||
return;
|
||||
}
|
||||
|
||||
this.sequenceIndex = 0;
|
||||
this.state = State.InTagName;
|
||||
this.stateInTagName(c);
|
||||
}
|
||||
|
||||
/** Look for an end tag. For <title> tags, also decode entities. */
|
||||
private stateInSpecialTag(c: number): void {
|
||||
if (this.sequenceIndex === this.currentSequence.length) {
|
||||
if (c === CharCodes.Gt || isWhitespace(c)) {
|
||||
const endOfText = this.index - this.currentSequence.length;
|
||||
|
||||
if (this.sectionStart < endOfText) {
|
||||
// Spoof the index so that reported locations match up.
|
||||
const actualIndex = this.index;
|
||||
this.index = endOfText;
|
||||
this.cbs.ontext(this.sectionStart, endOfText);
|
||||
this.index = actualIndex;
|
||||
}
|
||||
|
||||
this.isSpecial = false;
|
||||
this.sectionStart = endOfText + 2; // Skip over the `</`
|
||||
this.stateInClosingTagName(c);
|
||||
return; // We are done; skip the rest of the function.
|
||||
}
|
||||
|
||||
this.sequenceIndex = 0;
|
||||
}
|
||||
|
||||
if ((c | 0x20) === this.currentSequence[this.sequenceIndex]) {
|
||||
this.sequenceIndex += 1;
|
||||
} else if (this.sequenceIndex === 0) {
|
||||
if (this.currentSequence === Sequences.TitleEnd) {
|
||||
// We have to parse entities in <title> tags.
|
||||
if (this.decodeEntities && c === CharCodes.Amp) {
|
||||
this.startEntity();
|
||||
}
|
||||
} else if (this.fastForwardTo(CharCodes.Lt)) {
|
||||
// Outside of <title> tags, we can fast-forward.
|
||||
this.sequenceIndex = 1;
|
||||
}
|
||||
} else {
|
||||
// If we see a `<`, set the sequence index to 1; useful for eg. `<</script>`.
|
||||
this.sequenceIndex = Number(c === CharCodes.Lt);
|
||||
}
|
||||
}
|
||||
|
||||
private stateCDATASequence(c: number): void {
|
||||
if (c === Sequences.Cdata[this.sequenceIndex]) {
|
||||
if (++this.sequenceIndex === Sequences.Cdata.length) {
|
||||
this.state = State.InCommentLike;
|
||||
this.currentSequence = Sequences.CdataEnd;
|
||||
this.sequenceIndex = 0;
|
||||
this.sectionStart = this.index + 1;
|
||||
}
|
||||
} else {
|
||||
this.sequenceIndex = 0;
|
||||
this.state = State.InDeclaration;
|
||||
this.stateInDeclaration(c); // Reconsume the character
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* When we wait for one specific character, we can speed things up
|
||||
* by skipping through the buffer until we find it.
|
||||
*
|
||||
* @returns Whether the character was found.
|
||||
*/
|
||||
private fastForwardTo(c: number): boolean {
|
||||
while (++this.index < this.buffer.length + this.offset) {
|
||||
if (this.buffer.charCodeAt(this.index - this.offset) === c) {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
* We increment the index at the end of the `parse` loop,
|
||||
* so set it to `buffer.length - 1` here.
|
||||
*
|
||||
* TODO: Refactor `parse` to increment index before calling states.
|
||||
*/
|
||||
this.index = this.buffer.length + this.offset - 1;
|
||||
|
||||
return false;
|
||||
}
|
||||
|
||||
/**
|
||||
* Comments and CDATA end with `-->` and `]]>`.
|
||||
*
|
||||
* Their common qualities are:
|
||||
* - Their end sequences have a distinct character they start with.
|
||||
* - That character is then repeated, so we have to check multiple repeats.
|
||||
* - All characters but the start character of the sequence can be skipped.
|
||||
*/
|
||||
private stateInCommentLike(c: number): void {
|
||||
if (c === this.currentSequence[this.sequenceIndex]) {
|
||||
if (++this.sequenceIndex === this.currentSequence.length) {
|
||||
if (this.currentSequence === Sequences.CdataEnd) {
|
||||
this.cbs.oncdata(this.sectionStart, this.index, 2);
|
||||
} else {
|
||||
this.cbs.oncomment(this.sectionStart, this.index, 2);
|
||||
}
|
||||
|
||||
this.sequenceIndex = 0;
|
||||
this.sectionStart = this.index + 1;
|
||||
this.state = State.Text;
|
||||
}
|
||||
} else if (this.sequenceIndex === 0) {
|
||||
// Fast-forward to the first character of the sequence
|
||||
if (this.fastForwardTo(this.currentSequence[0])) {
|
||||
this.sequenceIndex = 1;
|
||||
}
|
||||
} else if (c !== this.currentSequence[this.sequenceIndex - 1]) {
|
||||
// Allow long sequences, eg. --->, ]]]>
|
||||
this.sequenceIndex = 0;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* HTML only allows ASCII alpha characters (a-z and A-Z) at the beginning of a tag name.
|
||||
*
|
||||
* XML allows a lot more characters here (@see https://www.w3.org/TR/REC-xml/#NT-NameStartChar).
|
||||
* We allow anything that wouldn't end the tag.
|
||||
*/
|
||||
private isTagStartChar(c: number) {
|
||||
return this.xmlMode ? !isEndOfTagSection(c) : isASCIIAlpha(c);
|
||||
}
|
||||
|
||||
private startSpecial(sequence: Uint8Array, offset: number) {
|
||||
this.isSpecial = true;
|
||||
this.currentSequence = sequence;
|
||||
this.sequenceIndex = offset;
|
||||
this.state = State.SpecialStartSequence;
|
||||
}
|
||||
|
||||
private stateBeforeTagName(c: number): void {
|
||||
if (c === CharCodes.ExclamationMark) {
|
||||
this.state = State.BeforeDeclaration;
|
||||
this.sectionStart = this.index + 1;
|
||||
} else if (c === CharCodes.Questionmark) {
|
||||
this.state = State.InProcessingInstruction;
|
||||
this.sectionStart = this.index + 1;
|
||||
} else if (this.isTagStartChar(c)) {
|
||||
const lower = c | 0x20;
|
||||
this.sectionStart = this.index;
|
||||
if (this.xmlMode) {
|
||||
this.state = State.InTagName;
|
||||
} else if (lower === Sequences.ScriptEnd[2]) {
|
||||
this.state = State.BeforeSpecialS;
|
||||
} else if (
|
||||
lower === Sequences.TitleEnd[2] ||
|
||||
lower === Sequences.XmpEnd[2]
|
||||
) {
|
||||
this.state = State.BeforeSpecialT;
|
||||
} else {
|
||||
this.state = State.InTagName;
|
||||
}
|
||||
} else if (c === CharCodes.Slash) {
|
||||
this.state = State.BeforeClosingTagName;
|
||||
} else {
|
||||
this.state = State.Text;
|
||||
this.stateText(c);
|
||||
}
|
||||
}
|
||||
private stateInTagName(c: number): void {
|
||||
if (isEndOfTagSection(c)) {
|
||||
this.cbs.onopentagname(this.sectionStart, this.index);
|
||||
this.sectionStart = -1;
|
||||
this.state = State.BeforeAttributeName;
|
||||
this.stateBeforeAttributeName(c);
|
||||
}
|
||||
}
|
||||
private stateBeforeClosingTagName(c: number): void {
|
||||
if (isWhitespace(c)) {
|
||||
// Ignore
|
||||
} else if (c === CharCodes.Gt) {
|
||||
this.state = State.Text;
|
||||
} else {
|
||||
this.state = this.isTagStartChar(c)
|
||||
? State.InClosingTagName
|
||||
: State.InSpecialComment;
|
||||
this.sectionStart = this.index;
|
||||
}
|
||||
}
|
||||
private stateInClosingTagName(c: number): void {
|
||||
if (c === CharCodes.Gt || isWhitespace(c)) {
|
||||
this.cbs.onclosetag(this.sectionStart, this.index);
|
||||
this.sectionStart = -1;
|
||||
this.state = State.AfterClosingTagName;
|
||||
this.stateAfterClosingTagName(c);
|
||||
}
|
||||
}
|
||||
private stateAfterClosingTagName(c: number): void {
|
||||
// Skip everything until ">"
|
||||
if (c === CharCodes.Gt || this.fastForwardTo(CharCodes.Gt)) {
|
||||
this.state = State.Text;
|
||||
this.sectionStart = this.index + 1;
|
||||
}
|
||||
}
|
||||
private stateBeforeAttributeName(c: number): void {
|
||||
if (c === CharCodes.Gt) {
|
||||
this.cbs.onopentagend(this.index);
|
||||
if (this.isSpecial) {
|
||||
this.state = State.InSpecialTag;
|
||||
this.sequenceIndex = 0;
|
||||
} else {
|
||||
this.state = State.Text;
|
||||
}
|
||||
this.sectionStart = this.index + 1;
|
||||
} else if (c === CharCodes.Slash) {
|
||||
this.state = State.InSelfClosingTag;
|
||||
} else if (!isWhitespace(c)) {
|
||||
this.state = State.InAttributeName;
|
||||
this.sectionStart = this.index;
|
||||
}
|
||||
}
|
||||
private stateInSelfClosingTag(c: number): void {
|
||||
if (c === CharCodes.Gt) {
|
||||
this.cbs.onselfclosingtag(this.index);
|
||||
this.state = State.Text;
|
||||
this.sectionStart = this.index + 1;
|
||||
this.isSpecial = false; // Reset special state, in case of self-closing special tags
|
||||
} else if (!isWhitespace(c)) {
|
||||
this.state = State.BeforeAttributeName;
|
||||
this.stateBeforeAttributeName(c);
|
||||
}
|
||||
}
|
||||
private stateInAttributeName(c: number): void {
|
||||
if (c === CharCodes.Eq || isEndOfTagSection(c)) {
|
||||
this.cbs.onattribname(this.sectionStart, this.index);
|
||||
this.sectionStart = this.index;
|
||||
this.state = State.AfterAttributeName;
|
||||
this.stateAfterAttributeName(c);
|
||||
}
|
||||
}
|
||||
private stateAfterAttributeName(c: number): void {
|
||||
if (c === CharCodes.Eq) {
|
||||
this.state = State.BeforeAttributeValue;
|
||||
} else if (c === CharCodes.Slash || c === CharCodes.Gt) {
|
||||
this.cbs.onattribend(QuoteType.NoValue, this.sectionStart);
|
||||
this.sectionStart = -1;
|
||||
this.state = State.BeforeAttributeName;
|
||||
this.stateBeforeAttributeName(c);
|
||||
} else if (!isWhitespace(c)) {
|
||||
this.cbs.onattribend(QuoteType.NoValue, this.sectionStart);
|
||||
this.state = State.InAttributeName;
|
||||
this.sectionStart = this.index;
|
||||
}
|
||||
}
|
||||
private stateBeforeAttributeValue(c: number): void {
|
||||
if (c === CharCodes.DoubleQuote) {
|
||||
this.state = State.InAttributeValueDq;
|
||||
this.sectionStart = this.index + 1;
|
||||
} else if (c === CharCodes.SingleQuote) {
|
||||
this.state = State.InAttributeValueSq;
|
||||
this.sectionStart = this.index + 1;
|
||||
} else if (!isWhitespace(c)) {
|
||||
this.sectionStart = this.index;
|
||||
this.state = State.InAttributeValueNq;
|
||||
this.stateInAttributeValueNoQuotes(c); // Reconsume token
|
||||
}
|
||||
}
|
||||
private handleInAttributeValue(c: number, quote: number) {
|
||||
if (
|
||||
c === quote ||
|
||||
(!this.decodeEntities && this.fastForwardTo(quote))
|
||||
) {
|
||||
this.cbs.onattribdata(this.sectionStart, this.index);
|
||||
this.sectionStart = -1;
|
||||
this.cbs.onattribend(
|
||||
quote === CharCodes.DoubleQuote
|
||||
? QuoteType.Double
|
||||
: QuoteType.Single,
|
||||
this.index + 1,
|
||||
);
|
||||
this.state = State.BeforeAttributeName;
|
||||
} else if (this.decodeEntities && c === CharCodes.Amp) {
|
||||
this.startEntity();
|
||||
}
|
||||
}
|
||||
private stateInAttributeValueDoubleQuotes(c: number): void {
|
||||
this.handleInAttributeValue(c, CharCodes.DoubleQuote);
|
||||
}
|
||||
private stateInAttributeValueSingleQuotes(c: number): void {
|
||||
this.handleInAttributeValue(c, CharCodes.SingleQuote);
|
||||
}
|
||||
private stateInAttributeValueNoQuotes(c: number): void {
|
||||
if (isWhitespace(c) || c === CharCodes.Gt) {
|
||||
this.cbs.onattribdata(this.sectionStart, this.index);
|
||||
this.sectionStart = -1;
|
||||
this.cbs.onattribend(QuoteType.Unquoted, this.index);
|
||||
this.state = State.BeforeAttributeName;
|
||||
this.stateBeforeAttributeName(c);
|
||||
} else if (this.decodeEntities && c === CharCodes.Amp) {
|
||||
this.startEntity();
|
||||
}
|
||||
}
|
||||
private stateBeforeDeclaration(c: number): void {
|
||||
if (c === CharCodes.OpeningSquareBracket) {
|
||||
this.state = State.CDATASequence;
|
||||
this.sequenceIndex = 0;
|
||||
} else {
|
||||
this.state =
|
||||
c === CharCodes.Dash
|
||||
? State.BeforeComment
|
||||
: State.InDeclaration;
|
||||
}
|
||||
}
|
||||
private stateInDeclaration(c: number): void {
|
||||
if (c === CharCodes.Gt || this.fastForwardTo(CharCodes.Gt)) {
|
||||
this.cbs.ondeclaration(this.sectionStart, this.index);
|
||||
this.state = State.Text;
|
||||
this.sectionStart = this.index + 1;
|
||||
}
|
||||
}
|
||||
private stateInProcessingInstruction(c: number): void {
|
||||
if (c === CharCodes.Gt || this.fastForwardTo(CharCodes.Gt)) {
|
||||
this.cbs.onprocessinginstruction(this.sectionStart, this.index);
|
||||
this.state = State.Text;
|
||||
this.sectionStart = this.index + 1;
|
||||
}
|
||||
}
|
||||
private stateBeforeComment(c: number): void {
|
||||
if (c === CharCodes.Dash) {
|
||||
this.state = State.InCommentLike;
|
||||
this.currentSequence = Sequences.CommentEnd;
|
||||
// Allow short comments (eg. <!-->)
|
||||
this.sequenceIndex = 2;
|
||||
this.sectionStart = this.index + 1;
|
||||
} else {
|
||||
this.state = State.InDeclaration;
|
||||
}
|
||||
}
|
||||
private stateInSpecialComment(c: number): void {
|
||||
if (c === CharCodes.Gt || this.fastForwardTo(CharCodes.Gt)) {
|
||||
this.cbs.oncomment(this.sectionStart, this.index, 0);
|
||||
this.state = State.Text;
|
||||
this.sectionStart = this.index + 1;
|
||||
}
|
||||
}
|
||||
private stateBeforeSpecialS(c: number): void {
|
||||
const lower = c | 0x20;
|
||||
if (lower === Sequences.ScriptEnd[3]) {
|
||||
this.startSpecial(Sequences.ScriptEnd, 4);
|
||||
} else if (lower === Sequences.StyleEnd[3]) {
|
||||
this.startSpecial(Sequences.StyleEnd, 4);
|
||||
} else {
|
||||
this.state = State.InTagName;
|
||||
this.stateInTagName(c); // Consume the token again
|
||||
}
|
||||
}
|
||||
|
||||
private stateBeforeSpecialT(c: number): void {
|
||||
const lower = c | 0x20;
|
||||
switch (lower) {
|
||||
case Sequences.TitleEnd[3]: {
|
||||
this.startSpecial(Sequences.TitleEnd, 4);
|
||||
|
||||
break;
|
||||
}
|
||||
case Sequences.TextareaEnd[3]: {
|
||||
this.startSpecial(Sequences.TextareaEnd, 4);
|
||||
|
||||
break;
|
||||
}
|
||||
case Sequences.XmpEnd[3]: {
|
||||
this.startSpecial(Sequences.XmpEnd, 4);
|
||||
|
||||
break;
|
||||
}
|
||||
default: {
|
||||
this.state = State.InTagName;
|
||||
this.stateInTagName(c); // Consume the token again
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
private startEntity() {
|
||||
this.baseState = this.state;
|
||||
this.state = State.InEntity;
|
||||
this.entityStart = this.index;
|
||||
this.entityDecoder.startEntity(
|
||||
this.xmlMode
|
||||
? DecodingMode.Strict
|
||||
: this.baseState === State.Text ||
|
||||
this.baseState === State.InSpecialTag
|
||||
? DecodingMode.Legacy
|
||||
: DecodingMode.Attribute,
|
||||
);
|
||||
}
|
||||
|
||||
private stateInEntity(): void {
|
||||
const indexInBuffer = this.index - this.offset;
|
||||
const length = this.entityDecoder.write(this.buffer, indexInBuffer);
|
||||
|
||||
// If `length` is positive, we are done with the entity.
|
||||
if (length >= 0) {
|
||||
this.state = this.baseState;
|
||||
|
||||
if (length === 0) {
|
||||
this.index -= 1;
|
||||
}
|
||||
} else {
|
||||
if (
|
||||
indexInBuffer < this.buffer.length &&
|
||||
this.buffer.charCodeAt(indexInBuffer) === CharCodes.Amp
|
||||
) {
|
||||
this.state = this.baseState;
|
||||
this.index -= 1;
|
||||
return;
|
||||
}
|
||||
|
||||
// Mark buffer as consumed.
|
||||
this.index = this.offset + this.buffer.length - 1;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Remove data that has already been consumed from the buffer.
|
||||
*/
|
||||
private cleanup() {
|
||||
// If we are inside of text or attributes, emit what we already have.
|
||||
if (this.running && this.sectionStart !== this.index) {
|
||||
if (
|
||||
this.state === State.Text ||
|
||||
(this.state === State.InSpecialTag && this.sequenceIndex === 0)
|
||||
) {
|
||||
this.cbs.ontext(this.sectionStart, this.index);
|
||||
this.sectionStart = this.index;
|
||||
} else if (
|
||||
this.state === State.InAttributeValueDq ||
|
||||
this.state === State.InAttributeValueSq ||
|
||||
this.state === State.InAttributeValueNq
|
||||
) {
|
||||
this.cbs.onattribdata(this.sectionStart, this.index);
|
||||
this.sectionStart = this.index;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
private shouldContinue() {
|
||||
return this.index < this.buffer.length + this.offset && this.running;
|
||||
}
|
||||
|
||||
/**
|
||||
* Iterates through the buffer, calling the function corresponding to the current state.
|
||||
*
|
||||
* States that are more likely to be hit are higher up, as a performance improvement.
|
||||
*/
|
||||
private parse() {
|
||||
while (this.shouldContinue()) {
|
||||
const c = this.buffer.charCodeAt(this.index - this.offset);
|
||||
switch (this.state) {
|
||||
case State.Text: {
|
||||
this.stateText(c);
|
||||
break;
|
||||
}
|
||||
case State.SpecialStartSequence: {
|
||||
this.stateSpecialStartSequence(c);
|
||||
break;
|
||||
}
|
||||
case State.InSpecialTag: {
|
||||
this.stateInSpecialTag(c);
|
||||
break;
|
||||
}
|
||||
case State.CDATASequence: {
|
||||
this.stateCDATASequence(c);
|
||||
break;
|
||||
}
|
||||
case State.InAttributeValueDq: {
|
||||
this.stateInAttributeValueDoubleQuotes(c);
|
||||
break;
|
||||
}
|
||||
case State.InAttributeName: {
|
||||
this.stateInAttributeName(c);
|
||||
break;
|
||||
}
|
||||
case State.InCommentLike: {
|
||||
this.stateInCommentLike(c);
|
||||
break;
|
||||
}
|
||||
case State.InSpecialComment: {
|
||||
this.stateInSpecialComment(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeAttributeName: {
|
||||
this.stateBeforeAttributeName(c);
|
||||
break;
|
||||
}
|
||||
case State.InTagName: {
|
||||
this.stateInTagName(c);
|
||||
break;
|
||||
}
|
||||
case State.InClosingTagName: {
|
||||
this.stateInClosingTagName(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeTagName: {
|
||||
this.stateBeforeTagName(c);
|
||||
break;
|
||||
}
|
||||
case State.AfterAttributeName: {
|
||||
this.stateAfterAttributeName(c);
|
||||
break;
|
||||
}
|
||||
case State.InAttributeValueSq: {
|
||||
this.stateInAttributeValueSingleQuotes(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeAttributeValue: {
|
||||
this.stateBeforeAttributeValue(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeClosingTagName: {
|
||||
this.stateBeforeClosingTagName(c);
|
||||
break;
|
||||
}
|
||||
case State.AfterClosingTagName: {
|
||||
this.stateAfterClosingTagName(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeSpecialS: {
|
||||
this.stateBeforeSpecialS(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeSpecialT: {
|
||||
this.stateBeforeSpecialT(c);
|
||||
break;
|
||||
}
|
||||
case State.InAttributeValueNq: {
|
||||
this.stateInAttributeValueNoQuotes(c);
|
||||
break;
|
||||
}
|
||||
case State.InSelfClosingTag: {
|
||||
this.stateInSelfClosingTag(c);
|
||||
break;
|
||||
}
|
||||
case State.InDeclaration: {
|
||||
this.stateInDeclaration(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeDeclaration: {
|
||||
this.stateBeforeDeclaration(c);
|
||||
break;
|
||||
}
|
||||
case State.BeforeComment: {
|
||||
this.stateBeforeComment(c);
|
||||
break;
|
||||
}
|
||||
case State.InProcessingInstruction: {
|
||||
this.stateInProcessingInstruction(c);
|
||||
break;
|
||||
}
|
||||
case State.InEntity: {
|
||||
this.stateInEntity();
|
||||
break;
|
||||
}
|
||||
}
|
||||
this.index++;
|
||||
}
|
||||
this.cleanup();
|
||||
}
|
||||
|
||||
private finish() {
|
||||
if (this.state === State.InEntity) {
|
||||
this.entityDecoder.end();
|
||||
this.state = this.baseState;
|
||||
}
|
||||
|
||||
this.handleTrailingData();
|
||||
|
||||
this.cbs.onend();
|
||||
}
|
||||
|
||||
/** Handle any trailing data. */
|
||||
private handleTrailingData() {
|
||||
const endIndex = this.buffer.length + this.offset;
|
||||
|
||||
// If there is no remaining data, we are done.
|
||||
if (this.sectionStart >= endIndex) {
|
||||
return;
|
||||
}
|
||||
|
||||
if (this.state === State.InCommentLike) {
|
||||
if (this.currentSequence === Sequences.CdataEnd) {
|
||||
this.cbs.oncdata(this.sectionStart, endIndex, 0);
|
||||
} else {
|
||||
this.cbs.oncomment(this.sectionStart, endIndex, 0);
|
||||
}
|
||||
} else if (
|
||||
this.state === State.InTagName ||
|
||||
this.state === State.BeforeAttributeName ||
|
||||
this.state === State.BeforeAttributeValue ||
|
||||
this.state === State.AfterAttributeName ||
|
||||
this.state === State.InAttributeName ||
|
||||
this.state === State.InAttributeValueSq ||
|
||||
this.state === State.InAttributeValueDq ||
|
||||
this.state === State.InAttributeValueNq ||
|
||||
this.state === State.InClosingTagName
|
||||
) {
|
||||
/*
|
||||
* If we are currently in an opening or closing tag, us not calling the
|
||||
* respective callback signals that the tag should be ignored.
|
||||
*/
|
||||
} else {
|
||||
this.cbs.ontext(this.sectionStart, endIndex);
|
||||
}
|
||||
}
|
||||
|
||||
private emitCodePoint(cp: number, consumed: number): void {
|
||||
if (
|
||||
this.baseState !== State.Text &&
|
||||
this.baseState !== State.InSpecialTag
|
||||
) {
|
||||
if (this.sectionStart < this.entityStart) {
|
||||
this.cbs.onattribdata(this.sectionStart, this.entityStart);
|
||||
}
|
||||
this.sectionStart = this.entityStart + consumed;
|
||||
this.index = this.sectionStart - 1;
|
||||
|
||||
this.cbs.onattribentity(cp);
|
||||
} else {
|
||||
if (this.sectionStart < this.entityStart) {
|
||||
this.cbs.ontext(this.sectionStart, this.entityStart);
|
||||
}
|
||||
this.sectionStart = this.entityStart + consumed;
|
||||
this.index = this.sectionStart - 1;
|
||||
|
||||
this.cbs.ontextentity(cp, this.sectionStart);
|
||||
}
|
||||
}
|
||||
}
|
||||
43
node_modules/htmlparser2/src/WritableStream.ts
generated
vendored
Normal file
43
node_modules/htmlparser2/src/WritableStream.ts
generated
vendored
Normal file
@@ -0,0 +1,43 @@
|
||||
import { Parser, type Handler, type ParserOptions } from "./Parser.js";
|
||||
/*
|
||||
* NOTE: If either of these two imports produces a type error,
|
||||
* please update your @types/node dependency!
|
||||
*/
|
||||
import { Writable } from "node:stream";
|
||||
import { StringDecoder } from "node:string_decoder";
|
||||
|
||||
// Following the example in https://nodejs.org/api/stream.html#stream_decoding_buffers_in_a_writable_stream
|
||||
function isBuffer(_chunk: string | Buffer, encoding: string): _chunk is Buffer {
|
||||
return encoding === "buffer";
|
||||
}
|
||||
|
||||
/**
|
||||
* WritableStream makes the `Parser` interface available as a NodeJS stream.
|
||||
*
|
||||
* @see Parser
|
||||
*/
|
||||
export class WritableStream extends Writable {
|
||||
private readonly _parser: Parser;
|
||||
private readonly _decoder = new StringDecoder();
|
||||
|
||||
constructor(cbs: Partial<Handler>, options?: ParserOptions) {
|
||||
super({ decodeStrings: false });
|
||||
this._parser = new Parser(cbs, options);
|
||||
}
|
||||
|
||||
override _write(
|
||||
chunk: string | Buffer,
|
||||
encoding: string,
|
||||
callback: () => void,
|
||||
): void {
|
||||
this._parser.write(
|
||||
isBuffer(chunk, encoding) ? this._decoder.write(chunk) : chunk,
|
||||
);
|
||||
callback();
|
||||
}
|
||||
|
||||
override _final(callback: () => void): void {
|
||||
this._parser.end(this._decoder.end());
|
||||
callback();
|
||||
}
|
||||
}
|
||||
115
node_modules/htmlparser2/src/index.ts
generated
vendored
Normal file
115
node_modules/htmlparser2/src/index.ts
generated
vendored
Normal file
@@ -0,0 +1,115 @@
|
||||
import { Parser, type ParserOptions } from "./Parser.js";
|
||||
export type { Handler, ParserOptions } from "./Parser.js";
|
||||
export { Parser } from "./Parser.js";
|
||||
|
||||
import {
|
||||
DomHandler,
|
||||
type DomHandlerOptions,
|
||||
type ChildNode,
|
||||
type Element,
|
||||
type Document,
|
||||
} from "domhandler";
|
||||
|
||||
export {
|
||||
DomHandler,
|
||||
// Old name for DomHandler
|
||||
DomHandler as DefaultHandler,
|
||||
type DomHandlerOptions,
|
||||
} from "domhandler";
|
||||
|
||||
export type Options = ParserOptions & DomHandlerOptions;
|
||||
|
||||
// Helper methods
|
||||
|
||||
/**
|
||||
* Parses the data, returns the resulting document.
|
||||
*
|
||||
* @param data The data that should be parsed.
|
||||
* @param options Optional options for the parser and DOM handler.
|
||||
*/
|
||||
export function parseDocument(data: string, options?: Options): Document {
|
||||
const handler = new DomHandler(undefined, options);
|
||||
new Parser(handler, options).end(data);
|
||||
return handler.root;
|
||||
}
|
||||
/**
|
||||
* Parses data, returns an array of the root nodes.
|
||||
*
|
||||
* Note that the root nodes still have a `Document` node as their parent.
|
||||
* Use `parseDocument` to get the `Document` node instead.
|
||||
*
|
||||
* @param data The data that should be parsed.
|
||||
* @param options Optional options for the parser and DOM handler.
|
||||
* @deprecated Use `parseDocument` instead.
|
||||
*/
|
||||
export function parseDOM(data: string, options?: Options): ChildNode[] {
|
||||
return parseDocument(data, options).children;
|
||||
}
|
||||
/**
|
||||
* Creates a parser instance, with an attached DOM handler.
|
||||
*
|
||||
* @param callback A callback that will be called once parsing has been completed, with the resulting document.
|
||||
* @param options Optional options for the parser and DOM handler.
|
||||
* @param elementCallback An optional callback that will be called every time a tag has been completed inside of the DOM.
|
||||
*/
|
||||
export function createDocumentStream(
|
||||
callback: (error: Error | null, document: Document) => void,
|
||||
options?: Options,
|
||||
elementCallback?: (element: Element) => void,
|
||||
): Parser {
|
||||
const handler: DomHandler = new DomHandler(
|
||||
(error: Error | null) => callback(error, handler.root),
|
||||
options,
|
||||
elementCallback,
|
||||
);
|
||||
return new Parser(handler, options);
|
||||
}
|
||||
/**
|
||||
* Creates a parser instance, with an attached DOM handler.
|
||||
*
|
||||
* @param callback A callback that will be called once parsing has been completed, with an array of root nodes.
|
||||
* @param options Optional options for the parser and DOM handler.
|
||||
* @param elementCallback An optional callback that will be called every time a tag has been completed inside of the DOM.
|
||||
* @deprecated Use `createDocumentStream` instead.
|
||||
*/
|
||||
export function createDomStream(
|
||||
callback: (error: Error | null, dom: ChildNode[]) => void,
|
||||
options?: Options,
|
||||
elementCallback?: (element: Element) => void,
|
||||
): Parser {
|
||||
const handler = new DomHandler(callback, options, elementCallback);
|
||||
return new Parser(handler, options);
|
||||
}
|
||||
|
||||
export {
|
||||
default as Tokenizer,
|
||||
type Callbacks as TokenizerCallbacks,
|
||||
QuoteType,
|
||||
} from "./Tokenizer.js";
|
||||
|
||||
/*
|
||||
* All of the following exports exist for backwards-compatibility.
|
||||
* They should probably be removed eventually.
|
||||
*/
|
||||
export * as ElementType from "domelementtype";
|
||||
|
||||
import { getFeed, type Feed } from "domutils";
|
||||
|
||||
export { getFeed, type Feed } from "domutils";
|
||||
|
||||
const parseFeedDefaultOptions = { xmlMode: true };
|
||||
|
||||
/**
|
||||
* Parse a feed.
|
||||
*
|
||||
* @param feed The feed that should be parsed, as a string.
|
||||
* @param options Optionally, options for parsing. When using this, you should set `xmlMode` to `true`.
|
||||
*/
|
||||
export function parseFeed(
|
||||
feed: string,
|
||||
options: Options = parseFeedDefaultOptions,
|
||||
): Feed | null {
|
||||
return getFeed(parseDOM(feed, options));
|
||||
}
|
||||
|
||||
export * as DomUtils from "domutils";
|
||||
Reference in New Issue
Block a user