dan/node-html-parser
1
0
Fork 0
mirror of https://github.com/danbulant/node-html-parser synced 2026-05-19 20:38:46 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

🐛 issue #35

Browse source
This commit is contained in:
taoqf 2020-03-17 15:58:05 +08:00
parent 9a48e032f1
commit 30a5701544
3 changed files with 44517 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
src/nodes/html.ts
View file

@ -286,11 +286,16 @@ export default class HTMLElement extends Node {
}
matcher = new Matcher(selector);
}
const stack = [] as { 0: Node; 1: 0 | 1; 2: boolean }[];
interface IStack {
0: Node; // node
1: number; // children
2: boolean; // found flag
}
const stack = [] as IStack[];
return this.childNodes.reduce((res, cur) => {
stack.push([cur, 0, false]);
while (stack.length) {
const state = arr_back(stack);
const state = arr_back(stack); // get last element
const el = state[0];
if (state[1] === 0) {
// Seen for first time.
@ -298,10 +303,12 @@ export default class HTMLElement extends Node {
stack.pop();
continue;
}
state[2] = matcher.advance(el as HTMLElement);
const html_el = el as HTMLElement;
state[2] = matcher.advance(html_el);
if (state[2]) {
if (matcher.matched) {
res.push(el as HTMLElement);
res.push(html_el);
res.push(...(html_el.querySelectorAll(selector)));
// no need to go further.
matcher.rewind();
stack.pop();

7
test/html.js
View file

@ -207,6 +207,13 @@ describe('HTML Parser', function () {
const root = parseHTML(fs.readFileSync(__dirname + '/html/incomplete-script').toString());
});
it('should parse talble currect', function () {
const root = parseHTML(fs.readFileSync(__dirname + '/html/tables.html').toString(), {
script: true
});
const tables = root.querySelectorAll('table');
tables.length.should.eql(3176);
});
});
describe('parseWithValidation', function () {

44499
test/html/tables.html Normal file
View file

File diff suppressed because it is too large Load diff