Press n or j to go to the next uncovered block, b, p or k for the previous block.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 | 1x 9x 9x 9x 9x 9x 5x 5x 4x 4x 4x 4x 4x 1x 1x 1x 1x 1x 1x 3x 3x 4x 4x 3x 17x 17x 35x 35x 2x 2x 33x 5x 5x 28x 11x 17x 12x 12x 5x 5x 2x 3x 3x 1x 1x 2x 2x 2x 1x 1x 1x 2x 17x 17x 17x 2x 15x 15x 15x 3x 12x 12x 12x 12x 4x 1x 3x 3x 1x 2x 2x 2x 2x 1x 1x 1x 1x 1x 1x 1x 1x | /**
* @module scripts/postprocess-translated-brief
* @description
* Post-processing pass for the news-translate workflow. After Pass-1
* translation, each `executive-brief_<lang>.md` file's H1 may still
* carry locale-specific boilerplate (`Exekutiv sammanfattning — `,
* `Zusammenfassung — `, `执行摘要：…`, …) or a trailing date suffix
* that the EN-only renderer regex doesn't catch. This helper re-applies
* the same `cleanArticleTitle` pipeline the renderer uses, so the H1
* on disk matches what the SERP eventually surfaces — preventing
* cleanup-only drift PRs and ensuring per-language pages render
* the same scrubbed title the contract guarantees.
*
* Usage:
* npx tsx scripts/postprocess-translated-brief.ts \
* analysis/daily/2026-05-15/propositions/executive-brief_sv.md
*
* Multiple paths may be passed on the command line; each is processed
* independently. The script exits 0 on success (even when no rewrite
* is needed) and exits 1 only on an unrecoverable error (file I/O,
* malformed path).
*
* Library exports are pure for unit-testability.
*
* @author Hack23 AB (Infrastructure Team)
* @license Apache-2.0
*/
import { readFileSync, writeFileSync, existsSync } from 'node:fs';
import { fileURLToPath } from 'node:url';
import { sep as PATH_SEP } from 'node:path';
import { cleanArticleTitle } from './render-lib/aggregator/seo/title.js';
import type { Language } from './types/language.js';
const LANG_CODES: ReadonlySet<Language> = new Set<Language>([
'en', 'sv', 'da', 'no', 'fi', 'de', 'fr', 'es', 'nl',
'ar', 'he', 'ja', 'ko', 'zh',
]);
/**
* Parse the language and `daily/<date>/<subfolder>` location from a
* `analysis/daily/<date>/<subfolder>/executive-brief_<lang>.md` path.
*
* Returns `null` for any path that doesn't fit the canonical layout,
* so the caller can skip non-brief files without crashing.
*/
export function parseBriefPath(filepath: string): { lang: Language; subfolder: string } | null {
const parts = filepath.split(/[\\/]/).filter(Boolean);
const filename = parts[parts.length - 1];
Iif (!filename) return null;
const m = filename.match(/^executive-brief_([a-z]{2})\.md$/);
if (!m) return null;
const lang = m[1] as Language;
if (!LANG_CODES.has(lang)) return null;
// Find the `daily` segment in the path. The subfolder is everything
// between `daily/<date>/` and the filename — joined with `/`.
const dailyIdx = parts.lastIndexOf('daily');
Iif (dailyIdx < 0 || dailyIdx + 2 >= parts.length - 1) return null;
const subfolderParts = parts.slice(dailyIdx + 2, parts.length - 1);
Iif (subfolderParts.length === 0) return null;
return { lang, subfolder: subfolderParts.join('/') };
}
/**
* Apply the post-processing transformation to a brief's markdown body
* and return both the (possibly rewritten) markdown and the before/after
* H1 strings. Pure function — no I/O.
*
* The transformation only rewrites the H1 line; every other line is
* passed through verbatim. When `cleanArticleTitle` returns `null`
* (cleaned title too short, or equal to the subfolder fallback), the
* original H1 is preserved — the translator's intent wins over the
* scrubber's veto, since the workflow won't synthesise a BLUF-based
* replacement post-hoc.
*/
// ─────────────────────────────────────────────────────────────────────────
// H1 discovery
// ─────────────────────────────────────────────────────────────────────────
/**
* Lines that wrap the title without being the title themselves.
* Translator output for RTL languages frequently leads with
* `<div dir="rtl">` so the rendered HTML aligns correctly; centered or
* attributed brief layouts use `<center>` / `<section>` / `<header>`;
* some templates emit a `<figure>` with a hero image before the title.
* These wrapper tags must be skipped so the scanner can reach the real
* heading instead of bailing out on the first non-`#` line.
*/
const HTML_WRAPPER_TAG_RE =
/^\s*<\/?(?:div|center|section|article|main|header|figure|figcaption|p|aside|nav)\b[^>]*\/?>\s*$/i;
/** Standalone image lines — HTML `<img>` or Markdown `` form. */
const IMAGE_LINE_RE = /^\s*(?:<img\b[^>]*\/?>|!\[[^\]]*\]\([^)]+\)(?:\s*\{[^}]*\})?)\s*$/i;
/** Markdown H1 (`# Title`). */
const MARKDOWN_H1_RE = /^#\s+/;
/** Same-line HTML H1: `<h1 ...>Title</h1>`. */
const HTML_H1_INLINE_RE = /^\s*<h1\b[^>]*>([\s\S]*?)<\/h1>\s*$/i;
/** Opening `<h1>` tag at line start (multi-line H1). */
const HTML_H1_OPEN_RE = /^\s*<h1\b[^>]*>(.*)$/i;
/** Closing `</h1>` tag (multi-line H1). */
const HTML_H1_CLOSE_RE = /^(.*?)<\/h1>\s*$/i;
interface H1Match {
/** First line of the heading block (inclusive). */
readonly startLine: number;
/** Last line of the heading block (inclusive). */
readonly endLine: number;
/** Heading kind: markdown `#` form or HTML `<h1>` form. */
readonly kind: 'markdown' | 'html';
/** Heading inner text with any nested HTML tags stripped. */
readonly text: string;
}
/** Strip nested HTML tags from a fragment captured inside an `<h1>`. */
function stripInnerTags(text: string): string {
// Apply repeatedly to prevent incomplete sanitization (CWE-20/CWE-80):
// nested or malformed sequences like `<<b>script>` could survive a
// single pass.
let prev: string;
let result = text;
do {
prev = result;
result = result.replace(/<[^>]*>/g, '');
} while (result !== prev);
return result.trim();
}
/**
* Locate the body H1 (markdown `#` form OR HTML `<h1>` form), skipping
* YAML frontmatter, HTML comments, blank lines, leading wrapper tags
* (`<div dir="rtl">`, `<center>`, `<section>`, …) and standalone
* image lines. Returns `null` when no H1 is present in the document.
*/
function findBodyH1(lines: readonly string[]): H1Match | null {
let inFrontmatter = false;
for (let i = 0; i < lines.length; i++) {
const ln = lines[i];
if (i === 0 && ln.trim() === '---') {
inFrontmatter = true;
continue;
}
if (inFrontmatter) {
if (ln.trim() === '---') inFrontmatter = false;
continue;
}
// Skip blank lines, HTML comments, wrapper tags, and standalone images.
if (
ln.trim() === '' ||
/^\s*<!--/.test(ln) ||
HTML_WRAPPER_TAG_RE.test(ln) ||
IMAGE_LINE_RE.test(ln)
) {
continue;
}
// Markdown H1
if (MARKDOWN_H1_RE.test(ln)) {
const text = ln.replace(MARKDOWN_H1_RE, '').trim();
return { startLine: i, endLine: i, kind: 'markdown', text };
}
// Inline HTML H1 — `<h1 ...>Title</h1>` on one line.
const inline = HTML_H1_INLINE_RE.exec(ln);
if (inline) {
return { startLine: i, endLine: i, kind: 'html', text: stripInnerTags(inline[1]) };
}
// Multi-line HTML H1 — opening `<h1>` here, closing `</h1>` later.
const open = HTML_H1_OPEN_RE.exec(ln);
if (open) {
// Gather inner text across subsequent lines until the closing tag.
const parts: string[] = [open[1]];
for (let j = i + 1; j < lines.length; j++) {
const inner = lines[j];
const close = HTML_H1_CLOSE_RE.exec(inner);
if (close) {
parts.push(close[1]);
return {
startLine: i,
endLine: j,
kind: 'html',
text: stripInnerTags(parts.join(' ')),
};
}
parts.push(inner);
}
// Unterminated `<h1>` — treat as no-match and let downstream
// markdown validation surface the malformed input.
return null;
}
// First content line is neither a wrapper, image, nor a heading —
// the document has no H1 we can safely rewrite.
return null;
}
return null;
}
export function postprocessBriefMarkdown(
markdown: string,
lang: Language,
subfolder: string,
): { markdown: string; changed: boolean; originalH1: string | null; cleanedH1: string | null } {
const lines = markdown.split('\n');
const match = findBodyH1(lines);
if (!match) {
return { markdown, changed: false, originalH1: null, cleanedH1: null };
}
const originalH1Text = match.text;
const cleaned = cleanArticleTitle(originalH1Text, subfolder, lang);
// When cleanArticleTitle returns null OR a string identical to the
// original (case-sensitive), no rewrite needed.
if (!cleaned || cleaned === originalH1Text) {
return { markdown, changed: false, originalH1: originalH1Text, cleanedH1: cleaned };
}
// Rewrite the heading block in place. For markdown form we replace a
// single line; for HTML form we collapse the whole `<h1>...</h1>`
// block into a normalised single-line markdown `# ` heading so the
// downstream renderer's title-extraction logic (which expects
// markdown H1s) keeps working uniformly across languages.
const before = lines.slice(0, match.startLine);
const after = lines.slice(match.endLine + 1);
const newLines = [...before, `# ${cleaned}`, ...after];
return {
markdown: newLines.join('\n'),
changed: true,
originalH1: originalH1Text,
cleanedH1: cleaned,
};
}
/**
* Read a brief file, apply post-processing, and write it back when
* the H1 changed. Returns the diff result for the caller (CLI or
* orchestrator) to log.
*/
export function postprocessBriefFile(filepath: string): {
filepath: string;
status: 'rewrote' | 'unchanged' | 'skipped' | 'error';
reason?: string;
originalH1?: string | null;
cleanedH1?: string | null;
} {
if (!existsSync(filepath)) {
return { filepath, status: 'error', reason: 'file not found' };
}
const parsed = parseBriefPath(filepath);
if (!parsed) {
return { filepath, status: 'skipped', reason: 'not a canonical executive-brief_<lang>.md path' };
}
let raw: string;
try {
raw = readFileSync(filepath, 'utf8');
} catch (e) {
return { filepath, status: 'error', reason: (e as Error).message };
}
const { markdown, changed, originalH1, cleanedH1 } = postprocessBriefMarkdown(
raw,
parsed.lang,
parsed.subfolder,
);
if (!changed) {
return { filepath, status: 'unchanged', originalH1, cleanedH1 };
}
try {
writeFileSync(filepath, markdown);
} catch (e) {
return { filepath, status: 'error', reason: (e as Error).message };
}
return { filepath, status: 'rewrote', originalH1, cleanedH1 };
}
function isMainModule(): boolean {
// ES-module entry-point check that doesn't rely on `require.main`.
Iif (typeof process === 'undefined' || !process.argv[1]) return false;
try {
return fileURLToPath(import.meta.url).split(PATH_SEP).join('/') ===
process.argv[1].split(PATH_SEP).join('/');
} catch {
return false;
}
}
function main(): void {
const args = process.argv.slice(2);
if (args.length === 0) {
console.error('usage: postprocess-translated-brief.ts <file> [<file> ...]');
process.exit(2);
}
let rewroteAny = false;
let hadError = false;
for (const filepath of args) {
const result = postprocessBriefFile(filepath);
if (result.status === 'rewrote') {
rewroteAny = true;
console.log(
`✏️ ${result.filepath}\n H1 was: ${result.originalH1}\n H1 now: ${result.cleanedH1}`,
);
} else if (result.status === 'unchanged') {
console.log(`✓ ${result.filepath} (H1 already clean)`);
} else if (result.status === 'skipped') {
console.log(`⊘ ${result.filepath} (${result.reason})`);
} else {
hadError = true;
console.error(`✗ ${result.filepath}: ${result.reason}`);
}
}
if (hadError) process.exit(1);
if (!rewroteAny) console.log('\nAll briefs already had a scrubbed H1 — nothing to rewrite.');
}
Iif (isMainModule()) {
main();
}
|