You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
489 lines
16 KiB
489 lines
16 KiB
2 years ago
|
import Tokenizer, { QuoteType } from "./Tokenizer.js";
|
||
|
import { fromCodePoint } from "entities/lib/decode.js";
|
||
|
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([
|
||
|
["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",
|
||
|
]);
|
||
|
const reNameEnd = /\s|\//;
|
||
|
export class Parser {
|
||
|
constructor(cbs, options = {}) {
|
||
|
var _a, _b, _c, _d, _e;
|
||
|
this.options = options;
|
||
|
/** The start index of the last event. */
|
||
|
this.startIndex = 0;
|
||
|
/** The end index of the last event. */
|
||
|
this.endIndex = 0;
|
||
|
/**
|
||
|
* Store the start index of the current open tag,
|
||
|
* so we can update the start index for attributes.
|
||
|
*/
|
||
|
this.openTagStart = 0;
|
||
|
this.tagname = "";
|
||
|
this.attribname = "";
|
||
|
this.attribvalue = "";
|
||
|
this.attribs = null;
|
||
|
this.stack = [];
|
||
|
this.foreignContext = [];
|
||
|
this.buffers = [];
|
||
|
this.bufferOffset = 0;
|
||
|
/** The index of the last written buffer. Used when resuming after a `pause()`. */
|
||
|
this.writeIndex = 0;
|
||
|
/** Indicates whether the parser has finished running / `.end` has been called. */
|
||
|
this.ended = false;
|
||
|
this.cbs = cbs !== null && cbs !== void 0 ? cbs : {};
|
||
|
this.lowerCaseTagNames = (_a = options.lowerCaseTags) !== null && _a !== void 0 ? _a : !options.xmlMode;
|
||
|
this.lowerCaseAttributeNames =
|
||
|
(_b = options.lowerCaseAttributeNames) !== null && _b !== void 0 ? _b : !options.xmlMode;
|
||
|
this.tokenizer = new ((_c = options.Tokenizer) !== null && _c !== void 0 ? _c : Tokenizer)(this.options, this);
|
||
|
(_e = (_d = this.cbs).onparserinit) === null || _e === void 0 ? void 0 : _e.call(_d, this);
|
||
|
}
|
||
|
// Tokenizer event handlers
|
||
|
/** @internal */
|
||
|
ontext(start, endIndex) {
|
||
|
var _a, _b;
|
||
|
const data = this.getSlice(start, endIndex);
|
||
|
this.endIndex = endIndex - 1;
|
||
|
(_b = (_a = this.cbs).ontext) === null || _b === void 0 ? void 0 : _b.call(_a, data);
|
||
|
this.startIndex = endIndex;
|
||
|
}
|
||
|
/** @internal */
|
||
|
ontextentity(cp) {
|
||
|
var _a, _b;
|
||
|
/*
|
||
|
* Entities can be emitted on the character, or directly after.
|
||
|
* We use the section start here to get accurate indices.
|
||
|
*/
|
||
|
const idx = this.tokenizer.getSectionStart();
|
||
|
this.endIndex = idx - 1;
|
||
|
(_b = (_a = this.cbs).ontext) === null || _b === void 0 ? void 0 : _b.call(_a, fromCodePoint(cp));
|
||
|
this.startIndex = idx;
|
||
|
}
|
||
|
isVoidElement(name) {
|
||
|
return !this.options.xmlMode && voidElements.has(name);
|
||
|
}
|
||
|
/** @internal */
|
||
|
onopentagname(start, endIndex) {
|
||
|
this.endIndex = endIndex;
|
||
|
let name = this.getSlice(start, endIndex);
|
||
|
if (this.lowerCaseTagNames) {
|
||
|
name = name.toLowerCase();
|
||
|
}
|
||
|
this.emitOpenTag(name);
|
||
|
}
|
||
|
emitOpenTag(name) {
|
||
|
var _a, _b, _c, _d;
|
||
|
this.openTagStart = this.startIndex;
|
||
|
this.tagname = name;
|
||
|
const impliesClose = !this.options.xmlMode && openImpliesClose.get(name);
|
||
|
if (impliesClose) {
|
||
|
while (this.stack.length > 0 &&
|
||
|
impliesClose.has(this.stack[this.stack.length - 1])) {
|
||
|
const el = this.stack.pop();
|
||
|
(_b = (_a = this.cbs).onclosetag) === null || _b === void 0 ? void 0 : _b.call(_a, el, true);
|
||
|
}
|
||
|
}
|
||
|
if (!this.isVoidElement(name)) {
|
||
|
this.stack.push(name);
|
||
|
if (foreignContextElements.has(name)) {
|
||
|
this.foreignContext.push(true);
|
||
|
}
|
||
|
else if (htmlIntegrationElements.has(name)) {
|
||
|
this.foreignContext.push(false);
|
||
|
}
|
||
|
}
|
||
|
(_d = (_c = this.cbs).onopentagname) === null || _d === void 0 ? void 0 : _d.call(_c, name);
|
||
|
if (this.cbs.onopentag)
|
||
|
this.attribs = {};
|
||
|
}
|
||
|
endOpenTag(isImplied) {
|
||
|
var _a, _b;
|
||
|
this.startIndex = this.openTagStart;
|
||
|
if (this.attribs) {
|
||
|
(_b = (_a = this.cbs).onopentag) === null || _b === void 0 ? void 0 : _b.call(_a, 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) {
|
||
|
this.endIndex = endIndex;
|
||
|
this.endOpenTag(false);
|
||
|
// Set `startIndex` for next node
|
||
|
this.startIndex = endIndex + 1;
|
||
|
}
|
||
|
/** @internal */
|
||
|
onclosetag(start, endIndex) {
|
||
|
var _a, _b, _c, _d, _e, _f;
|
||
|
this.endIndex = endIndex;
|
||
|
let name = this.getSlice(start, endIndex);
|
||
|
if (this.lowerCaseTagNames) {
|
||
|
name = name.toLowerCase();
|
||
|
}
|
||
|
if (foreignContextElements.has(name) ||
|
||
|
htmlIntegrationElements.has(name)) {
|
||
|
this.foreignContext.pop();
|
||
|
}
|
||
|
if (!this.isVoidElement(name)) {
|
||
|
const pos = this.stack.lastIndexOf(name);
|
||
|
if (pos !== -1) {
|
||
|
if (this.cbs.onclosetag) {
|
||
|
let count = this.stack.length - pos;
|
||
|
while (count--) {
|
||
|
// We know the stack has sufficient elements.
|
||
|
this.cbs.onclosetag(this.stack.pop(), count !== 0);
|
||
|
}
|
||
|
}
|
||
|
else
|
||
|
this.stack.length = pos;
|
||
|
}
|
||
|
else if (!this.options.xmlMode && name === "p") {
|
||
|
// Implicit open before close
|
||
|
this.emitOpenTag("p");
|
||
|
this.closeCurrentTag(true);
|
||
|
}
|
||
|
}
|
||
|
else if (!this.options.xmlMode && name === "br") {
|
||
|
// We can't use `emitOpenTag` for implicit open, as `br` would be implicitly closed.
|
||
|
(_b = (_a = this.cbs).onopentagname) === null || _b === void 0 ? void 0 : _b.call(_a, "br");
|
||
|
(_d = (_c = this.cbs).onopentag) === null || _d === void 0 ? void 0 : _d.call(_c, "br", {}, true);
|
||
|
(_f = (_e = this.cbs).onclosetag) === null || _f === void 0 ? void 0 : _f.call(_e, "br", false);
|
||
|
}
|
||
|
// Set `startIndex` for next node
|
||
|
this.startIndex = endIndex + 1;
|
||
|
}
|
||
|
/** @internal */
|
||
|
onselfclosingtag(endIndex) {
|
||
|
this.endIndex = endIndex;
|
||
|
if (this.options.xmlMode ||
|
||
|
this.options.recognizeSelfClosing ||
|
||
|
this.foreignContext[this.foreignContext.length - 1]) {
|
||
|
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);
|
||
|
}
|
||
|
}
|
||
|
closeCurrentTag(isOpenImplied) {
|
||
|
var _a, _b;
|
||
|
const name = this.tagname;
|
||
|
this.endOpenTag(isOpenImplied);
|
||
|
// Self-closing tags will be on the top of the stack
|
||
|
if (this.stack[this.stack.length - 1] === name) {
|
||
|
// If the opening tag isn't implied, the closing tag has to be implied.
|
||
|
(_b = (_a = this.cbs).onclosetag) === null || _b === void 0 ? void 0 : _b.call(_a, name, !isOpenImplied);
|
||
|
this.stack.pop();
|
||
|
}
|
||
|
}
|
||
|
/** @internal */
|
||
|
onattribname(start, endIndex) {
|
||
|
this.startIndex = start;
|
||
|
const name = this.getSlice(start, endIndex);
|
||
|
this.attribname = this.lowerCaseAttributeNames
|
||
|
? name.toLowerCase()
|
||
|
: name;
|
||
|
}
|
||
|
/** @internal */
|
||
|
onattribdata(start, endIndex) {
|
||
|
this.attribvalue += this.getSlice(start, endIndex);
|
||
|
}
|
||
|
/** @internal */
|
||
|
onattribentity(cp) {
|
||
|
this.attribvalue += fromCodePoint(cp);
|
||
|
}
|
||
|
/** @internal */
|
||
|
onattribend(quote, endIndex) {
|
||
|
var _a, _b;
|
||
|
this.endIndex = endIndex;
|
||
|
(_b = (_a = this.cbs).onattribute) === null || _b === void 0 ? void 0 : _b.call(_a, 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 = "";
|
||
|
}
|
||
|
getInstructionName(value) {
|
||
|
const idx = value.search(reNameEnd);
|
||
|
let name = idx < 0 ? value : value.substr(0, idx);
|
||
|
if (this.lowerCaseTagNames) {
|
||
|
name = name.toLowerCase();
|
||
|
}
|
||
|
return name;
|
||
|
}
|
||
|
/** @internal */
|
||
|
ondeclaration(start, endIndex) {
|
||
|
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, endIndex) {
|
||
|
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, endIndex, offset) {
|
||
|
var _a, _b, _c, _d;
|
||
|
this.endIndex = endIndex;
|
||
|
(_b = (_a = this.cbs).oncomment) === null || _b === void 0 ? void 0 : _b.call(_a, this.getSlice(start, endIndex - offset));
|
||
|
(_d = (_c = this.cbs).oncommentend) === null || _d === void 0 ? void 0 : _d.call(_c);
|
||
|
// Set `startIndex` for next node
|
||
|
this.startIndex = endIndex + 1;
|
||
|
}
|
||
|
/** @internal */
|
||
|
oncdata(start, endIndex, offset) {
|
||
|
var _a, _b, _c, _d, _e, _f, _g, _h, _j, _k;
|
||
|
this.endIndex = endIndex;
|
||
|
const value = this.getSlice(start, endIndex - offset);
|
||
|
if (this.options.xmlMode || this.options.recognizeCDATA) {
|
||
|
(_b = (_a = this.cbs).oncdatastart) === null || _b === void 0 ? void 0 : _b.call(_a);
|
||
|
(_d = (_c = this.cbs).ontext) === null || _d === void 0 ? void 0 : _d.call(_c, value);
|
||
|
(_f = (_e = this.cbs).oncdataend) === null || _f === void 0 ? void 0 : _f.call(_e);
|
||
|
}
|
||
|
else {
|
||
|
(_h = (_g = this.cbs).oncomment) === null || _h === void 0 ? void 0 : _h.call(_g, `[CDATA[${value}]]`);
|
||
|
(_k = (_j = this.cbs).oncommentend) === null || _k === void 0 ? void 0 : _k.call(_j);
|
||
|
}
|
||
|
// Set `startIndex` for next node
|
||
|
this.startIndex = endIndex + 1;
|
||
|
}
|
||
|
/** @internal */
|
||
|
onend() {
|
||
|
var _a, _b;
|
||
|
if (this.cbs.onclosetag) {
|
||
|
// Set the end index for all remaining tags
|
||
|
this.endIndex = this.startIndex;
|
||
|
for (let i = this.stack.length; i > 0; this.cbs.onclosetag(this.stack[--i], true))
|
||
|
;
|
||
|
}
|
||
|
(_b = (_a = this.cbs).onend) === null || _b === void 0 ? void 0 : _b.call(_a);
|
||
|
}
|
||
|
/**
|
||
|
* Resets the parser to a blank state, ready to parse a new HTML document
|
||
|
*/
|
||
|
reset() {
|
||
|
var _a, _b, _c, _d;
|
||
|
(_b = (_a = this.cbs).onreset) === null || _b === void 0 ? void 0 : _b.call(_a);
|
||
|
this.tokenizer.reset();
|
||
|
this.tagname = "";
|
||
|
this.attribname = "";
|
||
|
this.attribs = null;
|
||
|
this.stack.length = 0;
|
||
|
this.startIndex = 0;
|
||
|
this.endIndex = 0;
|
||
|
(_d = (_c = this.cbs).onparserinit) === null || _d === void 0 ? void 0 : _d.call(_c, this);
|
||
|
this.buffers.length = 0;
|
||
|
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.
|
||
|
*/
|
||
|
parseComplete(data) {
|
||
|
this.reset();
|
||
|
this.end(data);
|
||
|
}
|
||
|
getSlice(start, end) {
|
||
|
while (start - this.bufferOffset >= this.buffers[0].length) {
|
||
|
this.shiftBuffer();
|
||
|
}
|
||
|
let str = this.buffers[0].slice(start - this.bufferOffset, end - this.bufferOffset);
|
||
|
while (end - this.bufferOffset > this.buffers[0].length) {
|
||
|
this.shiftBuffer();
|
||
|
str += this.buffers[0].slice(0, end - this.bufferOffset);
|
||
|
}
|
||
|
return str;
|
||
|
}
|
||
|
shiftBuffer() {
|
||
|
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.
|
||
|
*/
|
||
|
write(chunk) {
|
||
|
var _a, _b;
|
||
|
if (this.ended) {
|
||
|
(_b = (_a = this.cbs).onerror) === null || _b === void 0 ? void 0 : _b.call(_a, 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.
|
||
|
*/
|
||
|
end(chunk) {
|
||
|
var _a, _b;
|
||
|
if (this.ended) {
|
||
|
(_b = (_a = this.cbs).onerror) === null || _b === void 0 ? void 0 : _b.call(_a, 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.
|
||
|
*/
|
||
|
pause() {
|
||
|
this.tokenizer.pause();
|
||
|
}
|
||
|
/**
|
||
|
* Resumes parsing after `pause` was called.
|
||
|
*/
|
||
|
resume() {
|
||
|
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
|
||
|
*/
|
||
|
parseChunk(chunk) {
|
||
|
this.write(chunk);
|
||
|
}
|
||
|
/**
|
||
|
* Alias of `end`, for backwards compatibility.
|
||
|
*
|
||
|
* @param chunk Optional final chunk to parse.
|
||
|
* @deprecated
|
||
|
*/
|
||
|
done(chunk) {
|
||
|
this.end(chunk);
|
||
|
}
|
||
|
}
|
||
|
//# sourceMappingURL=Parser.js.map
|