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 | 1x 1x 1x 1x 182x 182x 14x 14x 182x 182x 170x 170x 18x 18x 20x 20x 20x 20x 20x 3x 3x 1x 20x 20x 20x 14x 14x 20x 20x 20x 20x 19x 1x 1x 1x 1x 1x 14x 19x 14x 14x 2x 14x 8x 5x 1x 1x 14x 22x 22x 22x 22x 29x 29x 4x 25x 20x 22x 1x | #!/usr/bin/env tsx
/**
* @module catalog-downloaded-data
* @description Scans `analysis/data/` and produces a JSON catalog of all
* downloaded MCP data files with their metadata. This catalog is consumed
* by the AI agent during agentic workflows so it can perform per-file
* political intelligence analysis.
*
* The output is a JSON object ("data catalog") with overall metadata and an
* `entries` array describing each data file.
*
* Top-level catalog fields:
* - `generatedAt` – ISO 8601 timestamp when the catalog was generated
* - `dataRoot` – root directory that was scanned (e.g. "analysis/data")
* - `totalFiles` – total number of discovered data files
* - `pendingAnalysis` – number of files without analysis
* - `completedAnalysis` – number of files with existing analysis
* - `entries` – array of per-file catalog entries
*
* Each item in `entries` has:
* - `id` – document / record identifier (filename without `.json`)
* - `type` – persistence document type (e.g. "propositions", "mps")
* - `path` – path to the data file relative to repo root
* - `analysisPath` – expected path for the per-file analysis markdown
* - `hasAnalysis` – whether the analysis markdown already exists
* - `sizeBytes` – file size in bytes
*
* Usage:
* npx tsx scripts/catalog-downloaded-data.ts [--data-root <path>] [--type <type>] [--pending-only]
*
* Options:
* --data-root <path> Override analysis/data root (default: analysis/data)
* --type <type> Filter to a specific document type
* --pending-only Only list files that do NOT yet have analysis
*
* @author Hack23 AB
* @license Apache-2.0
*/
import * as fs from 'node:fs';
import * as path from 'node:path';
import { fileURLToPath } from 'node:url';
/** Stable repo root derived from script location (matches pre-article-analysis.ts). */
const __dirname = path.dirname(fileURLToPath(import.meta.url));
const REPO_ROOT = path.resolve(__dirname, '..');
/* ------------------------------------------------------------------ */
/* Types */
/* ------------------------------------------------------------------ */
/** A single entry in the data catalog. */
export interface CatalogEntry {
id: string;
type: string;
path: string;
analysisPath: string;
hasAnalysis: boolean;
sizeBytes: number;
meta: Record<string, unknown> | null;
}
/** Full catalog output. */
export interface DataCatalog {
generatedAt: string;
dataRoot: string;
totalFiles: number;
pendingAnalysis: number;
completedAnalysis: number;
entries: CatalogEntry[];
}
/* ------------------------------------------------------------------ */
/* Catalog builder (exported for testing) */
/* ------------------------------------------------------------------ */
const DATA_SUBDIRS = [
'documents/propositions',
'documents/motions',
'documents/committeeReports',
'documents/speeches',
'documents/questions',
'documents/interpellations',
'documents/votes',
'votes',
'events',
'mps',
'worldbank',
'scb',
'mcp-responses',
] as const;
/**
* Derive a human-friendly `type` from the subdirectory path.
* e.g. "documents/propositions" → "propositions", "mps" → "mps"
*/
function typeFromSubdir(subdir: string): string {
const parts = subdir.split('/');
return parts[parts.length - 1];
}
/**
* Build the catalog by scanning `dataRoot`.
*
* @param dataRoot - absolute or relative path to `analysis/data`
* @param filterType - optional type filter
* @param pendingOnly - if true, only include files without analysis
*/
export function buildCatalog(
dataRoot: string,
filterType?: string,
pendingOnly = false,
): DataCatalog {
const allEntries: CatalogEntry[] = [];
for (const subdir of DATA_SUBDIRS) {
const docType = typeFromSubdir(subdir);
if (filterType && docType !== filterType) continue;
const dirPath = path.join(dataRoot, subdir);
if (!fs.existsSync(dirPath)) continue;
// The directory may contain date subdirectories (e.g. votes/2026-03-28/)
// or direct JSON files. We scan recursively but only pick up *.json
// files that are NOT .meta.json.
const jsonFiles = collectJsonFiles(dirPath);
for (const filePath of jsonFiles) {
// Use path relative to the type directory (sans .json) as id.
// For flat dirs: "P1". For nested dirs: "ind1/SE".
const id = path.relative(dirPath, filePath).replace(/\.json$/, '').split(path.sep).join('/');
const metaPath = filePath.replace(/\.json$/, '.meta.json');
const analysisPath = filePath.replace(/\.json$/, '.analysis.md');
let meta: Record<string, unknown> | null = null;
if (fs.existsSync(metaPath)) {
try {
meta = JSON.parse(fs.readFileSync(metaPath, 'utf-8'));
} catch {
meta = null;
}
}
const hasAnalysis = fs.existsSync(analysisPath);
const stat = fs.statSync(filePath);
allEntries.push({
id: id,
type: docType,
path: path.relative(REPO_ROOT, filePath).split(path.sep).join('/'),
analysisPath: path.relative(REPO_ROOT, analysisPath).split(path.sep).join('/'),
hasAnalysis,
sizeBytes: stat.size,
meta,
});
}
}
// De-duplicate vote entries: when the same vote file appears in both
// documents/votes and votes/YYYY-MM-DD, prefer the date-stamped path.
// Only votes are scoped for dedup (the only type with two scan dirs).
// Vote ids use the basename portion for matching since documents/votes/
// stores files flat while votes/YYYY-MM-DD/ nests under date dirs.
const bestByKey = new Map<string, (typeof allEntries)[number]>();
for (const e of allEntries) {
// For votes, use basename as dedup key (ignores date-dir nesting)
const idPart = e.type === 'votes' ? e.id.split('/').pop()! : e.id;
const key = `${e.type}::${idPart}`;
const existing = bestByKey.get(key);
if (!existing) {
bestByKey.set(key, e);
} else if (Ee.type === 'votes') {
// Prefer the entry whose path contains a date directory (votes/YYYY-MM-DD)
const existingHasDate = /votes\/\d{4}-\d{2}-\d{2}\//.test(existing.path);
const currentHasDate = /votes\/\d{4}-\d{2}-\d{2}\//.test(e.path);
Eif (currentHasDate && !existingHasDate) {
bestByKey.set(key, e);
}
// Otherwise keep existing (first-seen or already date-stamped)
}
// Non-vote duplicates: keep first-seen (shouldn't occur with current DATA_SUBDIRS)
}
const dedupedEntries = [...bestByKey.values()];
// Compute totals from the full scan (before pendingOnly filter)
const totalCompleted = dedupedEntries.filter((e) => e.hasAnalysis).length;
const totalPending = dedupedEntries.length - totalCompleted;
// Apply pendingOnly filter after computing totals
const entries = pendingOnly
? dedupedEntries.filter((e) => !e.hasAnalysis)
: dedupedEntries;
// Ensure deterministic ordering across platforms/filesystems.
// Use simple < / > string compare (locale-independent) for stable collation.
entries.sort((a, b) => {
if (a.type < b.type) return -1;
if (a.type > b.type) return 1;
Iif (a.id < b.id) return -1;
Eif (a.id > b.id) return 1;
return 0;
});
return {
generatedAt: new Date().toISOString(),
dataRoot: path.relative(REPO_ROOT, dataRoot).split(path.sep).join('/'),
totalFiles: dedupedEntries.length,
pendingAnalysis: totalPending,
completedAnalysis: totalCompleted,
entries,
};
}
/* ------------------------------------------------------------------ */
/* Helpers */
/* ------------------------------------------------------------------ */
/** Recursively collect *.json files, excluding *.meta.json. */
function collectJsonFiles(dir: string): string[] {
const results: string[] = [];
let dirEntries: fs.Dirent[];
try {
dirEntries = fs.readdirSync(dir, { withFileTypes: true });
} catch {
return results;
}
for (const entry of dirEntries) {
const full = path.join(dir, entry.name);
if (entry.isDirectory()) {
results.push(...collectJsonFiles(full));
} else if (
entry.isFile() &&
entry.name.endsWith('.json') &&
!entry.name.endsWith('.meta.json')
) {
results.push(full);
}
}
return results;
}
/* ------------------------------------------------------------------ */
/* CLI entry point */
/* ------------------------------------------------------------------ */
function parseArgs(argv: string[]) {
let dataRoot = path.join(REPO_ROOT, 'analysis/data');
let filterType: string | undefined;
let pendingOnly = false;
for (let i = 2; i < argv.length; i++) {
if (argv[i] === '--data-root' && argv[i + 1]) {
dataRoot = path.resolve(argv[++i]);
} else if (argv[i] === '--type' && argv[i + 1]) {
filterType = argv[++i];
} else if (argv[i] === '--pending-only') {
pendingOnly = true;
}
}
return { dataRoot, filterType, pendingOnly };
}
/* istanbul ignore next -- CLI wrapper */
function main() {
const { dataRoot, filterType, pendingOnly } = parseArgs(process.argv);
const catalog = buildCatalog(dataRoot, filterType, pendingOnly);
console.error(
`╔══════════════════════════════════════════════════════════════╗`,
);
console.error(
`║ 📋 Analysis Data Catalog ║`,
);
console.error(
`╚══════════════════════════════════════════════════════════════╝`,
);
console.error(` 📂 Data root: ${catalog.dataRoot}`);
console.error(` 📄 Total files: ${catalog.totalFiles}`);
console.error(` ✅ Analyzed: ${catalog.completedAnalysis}`);
console.error(` ⏳ Pending: ${catalog.pendingAnalysis}`);
console.error();
// Write catalog JSON to stdout for piping
console.log(JSON.stringify(catalog, null, 2));
}
// Run CLI when invoked directly (matches repo convention in pre-article-analysis.ts)
Iif (path.resolve(fileURLToPath(import.meta.url)) === path.resolve(process.argv[1] ?? '')) {
main();
}
|