chore: Review fix

This commit is contained in:
iamsivin
2026-04-29 18:28:34 +05:30
parent 6630eb5c14
commit 122295b5f4
2 changed files with 49 additions and 12 deletions
@@ -126,11 +126,12 @@ export class EmailQuoteExtractor {
return this.findBlocksContainingText(root, HEADER_PATTERNS);
}
// ---------- 5. Plain-text RFC `>` tail ----------
// For text/plain emails (after sanitizeTextForRender converts \n → <br>),
// find the earliest top-level text node whose visible lines all begin with
// `>` and strip from there to the end (collapsing leading <br>/whitespace
// separators back into the tail).
// ---------- 5. Top-level quote tail ----------
// For replies that arrive as text + <br> with no block wrapper (text/plain
// bodies after sanitizeTextForRender). Find the earliest top-level text
// node that begins a quote tail — either every visible line starts with `>`
// (RFC quote prefix) or the text contains a header marker — and strip from
// there, collapsing leading <br>/whitespace separators into the tail.
static removePlainTextTail(root) {
const start = this.findPlainTextTailStart(root);
@@ -142,7 +143,7 @@ export class EmailQuoteExtractor {
static findPlainTextTailStart(root) {
const children = Array.from(root.childNodes);
const tailIdx = children.findIndex(node =>
this.isQuotePrefixedTextNode(node)
this.isQuoteTailStartTextNode(node)
);
if (tailIdx === -1) return -1;
let start = tailIdx;
@@ -152,14 +153,18 @@ export class EmailQuoteExtractor {
return start;
}
static isQuotePrefixedTextNode(node) {
static isQuoteTailStartTextNode(node) {
if (node.nodeType !== Node.TEXT_NODE) return false;
const text = node.textContent;
if (!text.trim()) return false;
return text
.split('\n')
.filter(line => line.trim() !== '')
.every(line => line.trim().startsWith('>'));
const lines = text.split('\n').filter(line => line.trim() !== '');
if (lines.length > 0 && lines.every(l => l.trim().startsWith('>'))) {
return true;
}
return (
HEADER_PATTERNS.some(p => p.test(text)) ||
HARD_HEADER_PATTERNS.some(p => p.test(text))
);
}
static isNeutralNode(node) {
@@ -1,4 +1,4 @@
import { describe, it, expect } from 'vitest';
import { describe, expect, it } from 'vitest';
import { EmailQuoteExtractor } from '../emailQuoteExtractor.js';
const SAMPLE_EMAIL_HTML = `
@@ -237,6 +237,38 @@ describe('EmailQuoteExtractor', () => {
expect(cleaned).toContain('Reply text');
expect(cleaned).not.toContain('From: Sam');
});
// Header markers at the TOP LEVEL — text + <br> shape with no block
// wrapper. The marker's nearest block ancestor is root itself.
it('detects top-level "On … wrote:" header (no wrapper)', () => {
const html =
'Reply text<br><br>On Tue, Pat wrote:<br>Original line 1<br>Original line 2';
expect(EmailQuoteExtractor.hasQuotes(html)).toBe(true);
const c = document.createElement('div');
c.innerHTML = EmailQuoteExtractor.extractQuotes(html);
expect(c.textContent).toContain('Reply text');
expect(c.textContent).not.toContain('On Tue, Pat wrote');
expect(c.textContent).not.toContain('Original line 1');
});
it('detects top-level "From:/Sent:" header (no wrapper)', () => {
const html =
'Reply text<br>From: Sam<br>Sent: Wednesday<br>Original body';
expect(EmailQuoteExtractor.hasQuotes(html)).toBe(true);
const c = document.createElement('div');
c.innerHTML = EmailQuoteExtractor.extractQuotes(html);
expect(c.textContent).toContain('Reply text');
expect(c.textContent).not.toContain('From: Sam');
});
it('detects top-level "-----Original Message-----" (no wrapper)', () => {
const html = 'Reply<br>-----Original Message-----<br>From: Sam<br>Body';
expect(EmailQuoteExtractor.hasQuotes(html)).toBe(true);
const c = document.createElement('div');
c.innerHTML = EmailQuoteExtractor.extractQuotes(html);
expect(c.textContent).toContain('Reply');
expect(c.textContent).not.toContain('Original Message');
});
});
it('strips RFC-style `>` quoted lines from a plain-text only body (iPhone Mail)', () => {