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 | 1x 89x 89x 11x 6x 3x 3x 3x 3x 3x 8x 3x 5x 21x 25x 25x 25x 25x 25x 31x 31x 31x 31x 31x 23x 23x 31x 8x 8x 25x 66x 66x 66x 96x 66x 25x 25x 25x 25x 25x 25x 25x 10x 10x 10x 9x 10x 10x 25x 75x 24x 24x 62x 24x 24x 24x 24x 3x 24x 24x 26x 26x 26x 11x 10x 10x 25x 25x 25x 25x 25x 25x 25x 25x 25x 25x 25x 25x 25x 25x 25x 1x | /**
* @module Infrastructure/BackfillLib/HtmlInspector
* @category Intelligence Operations / Supporting Infrastructure
* @name HTML metadata extractor for existing `news/*.html` articles
*
* @description
* Parses an existing `news/*.html` file and returns the ten metadata
* surfaces named in the PR 2 issue without triggering the full render
* pipeline. This is the read-only counterpart to
* `scripts/rewrite-article-metadata.ts`; it never mutates the file.
*
* Parsing is regex-based rather than DOM-based for three reasons:
*
* 1. **Speed.** Inspecting thousands of articles in dry-run mode must
* complete in a few seconds on CI, not minutes.
* 2. **Byte safety.** The existing rewriter already uses the same
* regex approach so output is consistent between the two tools.
* 3. **No extra dependency.** `jsdom` / `cheerio` would pull a few MB
* of node_modules for a task the built-in `RegExp` handles.
*
* @author Hack23 AB (Infrastructure Team)
* @license Apache-2.0
*/
import fs from 'fs';
/** Structured metadata extracted from a single article HTML file. */
export interface ArticleMetadata {
/** Absolute path of the inspected file. */
readonly filePath: string;
/** `<html lang="…">` value (e.g. `en`, `sv`, `ar`). Empty string when
* the attribute is missing — callers fall back to the filename code
* via `meta.lang || fp.lang`. */
readonly lang: string;
/** `<title>…</title>` content (HTML-decoded, no trimming beyond trim). */
readonly title: string;
/** `<meta name="description" content="…">`. */
readonly metaDescription: string;
/** `<meta property="og:title" content="…">`. */
readonly ogTitle: string;
/** `<meta property="og:description" content="…">`. */
readonly ogDescription: string;
/** `<meta name="twitter:title" content="…">`. */
readonly twitterTitle: string;
/** `<meta name="twitter:description" content="…">`. */
readonly twitterDescription: string;
/** JSON-LD `headline` field (first `<script type="application/ld+json">`). */
readonly jsonLdHeadline: string;
/** JSON-LD `alternativeHeadline` field. */
readonly jsonLdAlternativeHeadline: string;
/** JSON-LD `description` field. */
readonly jsonLdDescription: string;
/** First ~2KB of the `<article>` plain text (tags stripped, whitespace
* collapsed). Used by PRs 3/4 as a diagnostic preview — never
* validated against the contract directly. */
readonly bodyPlainText: string;
}
const REGEXES = {
htmlTag: /<html\b[^>]*>/i,
title: /<title[^>]*>([\s\S]*?)<\/title>/i,
article: /<article\b[^>]*>([\s\S]*?)<\/article>/i,
} as const;
/** Decode the small subset of HTML entities used by the rendered articles.
* Single-pass so `&quot;` survives as the literal text `"`
* rather than being double-unescaped to `"`. (CodeQL js/double-escaping.) */
function htmlDecode(s: string): string {
const named: Record<string, string> = {
amp: '&',
quot: '"',
apos: "'",
lt: '<',
gt: '>',
nbsp: ' ',
};
return s.replace(/&(?:#(\d+)|#x([0-9a-fA-F]+)|([a-zA-Z][a-zA-Z0-9]+)|#39);?/g,
(full, dec?: string, hex?: string, name?: string) => {
if (dec) return decodeCodePoint(Number(dec), full);
if (hex) return decodeCodePoint(parseInt(hex, 16), full);
Iif (full === ''') return "'";
Eif (name) {
const lower = name.toLowerCase();
Eif (Object.prototype.hasOwnProperty.call(named, lower)) {
return named[lower] as string;
}
}
return full;
});
}
function decodeCodePoint(value: number, fallback: string): string {
if (
value < 0 ||
value > 0x10ffff ||
(value >= 0xd800 && value <= 0xdfff)
) {
return fallback;
}
return String.fromCodePoint(value);
}
/** Strip inline HTML tags and collapse whitespace to produce plain prose.
* Script / style end tags tolerate whitespace and malformed trailing
* tokens before `>` so an odd `</script\t\n bar>` cannot leak body text. */
function stripTags(fragment: string): string {
return fragment
.replace(/<script\b[^>]*>[\s\S]*?<\/script\b[^>]*>/gi, ' ')
.replace(/<style\b[^>]*>[\s\S]*?<\/style\b[^>]*>/gi, ' ')
.replace(/<br\s*\/?>/gi, ' ')
.replace(/<[^>]+>/g, ' ')
.replace(/\s+/g, ' ')
.trim();
}
/** Extract a captured group or return `''`. */
function match1(html: string, re: RegExp): string {
const m = html.match(re);
return m ? htmlDecode(m[1] ?? '').trim() : '';
}
function extractMetaContent(
html: string,
selectorAttr: 'name' | 'property',
selectorValue: string,
): string {
const map = parseAllMetaTags(html);
return map.get(`${selectorAttr}:${selectorValue.toLowerCase()}`) ?? '';
}
/** Walk every `<meta …>` once and build a lookup keyed by
* `${name|property}:${value}` (lowercased). The CLI scans thousands of
* files, so collapsing five sequential regex passes into one keeps the
* hot path small. */
function parseAllMetaTags(html: string): Map<string, string> {
const map = new Map<string, string>();
const metaTagRe = /<meta\b[^>]*>/gi;
let m: RegExpExecArray | null;
while ((m = metaTagRe.exec(html)) !== null) {
const attrs = parseAttributes(m[0]);
const content = htmlDecode(attrs.content ?? '').trim();
const name = attrs.name?.toLowerCase();
const property = attrs.property?.toLowerCase();
if (name) {
const key = `name:${name}`;
Eif (!map.has(key)) map.set(key, content);
}
if (property) {
const key = `property:${property}`;
Eif (!map.has(key)) map.set(key, content);
}
}
return map;
}
function parseAttributes(tag: string): Record<string, string> {
const attrs: Record<string, string> = {};
const attrRe = /([^\s=/>]+)\s*=\s*(?:"([^"]*)"|'([^']*)'|([^\s"'=<>`]+))/g;
let m: RegExpExecArray | null;
while ((m = attrRe.exec(tag)) !== null) {
attrs[m[1]!.toLowerCase()] = m[2] ?? m[3] ?? m[4] ?? '';
}
return attrs;
}
/** Extract the `lang` attribute from the document's `<html>` tag,
* tolerating single-quoted, double-quoted, and unquoted attribute
* values plus any preceding attributes. Returns `''` when missing so
* callers can apply their own fallback (e.g. filename language). */
function extractHtmlLang(html: string): string {
const m = html.match(REGEXES.htmlTag);
Iif (!m) return '';
const attrs = parseAttributes(m[0]);
return htmlDecode(attrs.lang ?? '').trim();
}
/** Walk every `<script>` once, keep the JSON-LD ones, and return their
* parsed (or `null` for malformed) bodies. The script tag matcher
* tolerates additional attributes, alternate orderings, and
* single-quoted `type` values. */
function parseAllJsonLdBlocks(html: string): readonly unknown[] {
const scriptRe = /<script\b([^>]*)>([\s\S]*?)<\/script\b[^>]*>/gi;
const blocks: unknown[] = [];
let m: RegExpExecArray | null;
while ((m = scriptRe.exec(html)) !== null) {
const attrs = parseAttributes(`<script ${m[1] ?? ''}>`);
const type = (attrs.type ?? '').toLowerCase();
if (type !== 'application/ld+json') continue;
const body = m[2] ?? '';
try {
blocks.push(JSON.parse(body));
} catch {
// Skip malformed blocks — never fatal.
}
}
return blocks;
}
function readJsonLdFieldFromBlocks(
blocks: readonly unknown[],
field: 'headline' | 'alternativeHeadline' | 'description',
): string {
for (const parsed of blocks) {
const value = readJsonLdField(parsed, field);
if (value) return value;
}
return '';
}
function readJsonLdField(
parsed: unknown,
field: 'headline' | 'alternativeHeadline' | 'description',
): string {
Iif (!parsed || typeof parsed !== 'object') return '';
// JSON-LD can be a single object or a `@graph` array.
const candidates: unknown[] = [];
const record = parsed as Record<string, unknown>;
if (Array.isArray(record['@graph'])) {
candidates.push(...(record['@graph'] as unknown[]));
}
candidates.push(record);
for (const c of candidates) {
Eif (c && typeof c === 'object') {
const v = (c as Record<string, unknown>)[field];
if (typeof v === 'string' && v.trim().length > 0) return v.trim();
}
}
return '';
}
/**
* Inspect a single `news/*.html` file and return its full metadata
* surface. The file is read from disk; callers who already have the
* contents in memory should use {@link inspectHtmlContent} instead.
*/
export function inspectHtmlFile(filePath: string): ArticleMetadata {
const html = fs.readFileSync(filePath, 'utf8');
return inspectHtmlContent(html, filePath);
}
/**
* Inspect an already-loaded HTML string and return its metadata. Split
* from {@link inspectHtmlFile} so tests can run against literal HTML
* without writing to disk.
*/
export function inspectHtmlContent(html: string, filePath: string = ''): ArticleMetadata {
const lang = extractHtmlLang(html);
const title = match1(html, REGEXES.title);
// Parse `<meta>` tags and JSON-LD blocks once per file rather than
// scanning the document for every requested field. The CLI walks
// thousands of articles, so collapsing 5 + 3 sequential passes into
// one each keeps the hot path tight.
const metaMap = parseAllMetaTags(html);
const jsonLdBlocks = parseAllJsonLdBlocks(html);
const metaDescription = metaMap.get('name:description') ?? '';
const ogTitle = metaMap.get('property:og:title') ?? '';
const ogDescription = metaMap.get('property:og:description') ?? '';
const twitterTitle = metaMap.get('name:twitter:title') ?? '';
const twitterDescription = metaMap.get('name:twitter:description') ?? '';
const jsonLdHeadline = readJsonLdFieldFromBlocks(jsonLdBlocks, 'headline');
const jsonLdAlternativeHeadline = readJsonLdFieldFromBlocks(jsonLdBlocks, 'alternativeHeadline');
const jsonLdDescription = readJsonLdFieldFromBlocks(jsonLdBlocks, 'description');
const articleMatch = html.match(REGEXES.article);
const bodyPlainText = articleMatch
? stripTags(articleMatch[1] ?? '').slice(0, 2048)
: '';
return {
filePath,
lang,
title,
metaDescription,
ogTitle,
ogDescription,
twitterTitle,
twitterDescription,
jsonLdHeadline,
jsonLdAlternativeHeadline,
jsonLdDescription,
bodyPlainText,
};
}
export const __test__ = {
htmlDecode,
stripTags,
extractMetaContent,
parseAttributes,
REGEXES,
};
|