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 | 48x 8x 8x 8x 8x 8x 8x 8x 8x 8x 8x 8x 8x 7x 8x 5x 8x 7x 8x 7x 8x 7x 8x 7x 8x 7x 8x 56x 8x 56x 56x 48x 48x 48x 1x 8x 8x 8x 7x 7x 7x 17x 16x 16x 16x 48x 17x 12x 12x | /**
* @module pre-article-analysis/data-downloader
* @description Downloads all relevant parliamentary documents from riksdag-regering-mcp
* for the current Swedish parliamentary session (riksmöte). Returns typed `RawDocument[]`
* collections plus a manifest of which MCP tools returned successful results and how many
* documents each returned.
*
* Note: The download methods fetch session-wide latest documents (bounded by `limit`
* and `rm`). Date-specific filtering should be applied by the caller after download
* (e.g., filtering by the `datum` field on each `RawDocument`).
*
* This module is intentionally side-effect-free with respect to the filesystem;
* callers are responsible for writing any output.
*
* @author Hack23 AB
* @license Apache-2.0
*/
import type { RawDocument } from '../data-transformers/types.js';
import type { MCPClient } from '../mcp-client/client.js';
// ---------------------------------------------------------------------------
// Types
// ---------------------------------------------------------------------------
export interface DownloadedData {
/** Government propositions */
propositions: RawDocument[];
/** Parliamentary motions */
motions: RawDocument[];
/** Committee reports */
committeeReports: RawDocument[];
/** Parliamentary votes */
votes: RawDocument[];
/** Parliamentary speeches (anföranden) */
speeches: RawDocument[];
/** Written questions */
questions: RawDocument[];
/** Interpellations */
interpellations: RawDocument[];
}
export interface DownloadManifest {
/** Names of MCP tools with successful fetch + post-processing during download */
dataSources: string[];
/** Document counts per type */
docCounts: Record<string, number>;
/** Total download duration in milliseconds */
durationMs: number;
}
export interface DownloadResult {
data: DownloadedData;
manifest: DownloadManifest;
}
// ---------------------------------------------------------------------------
// Helpers
// ---------------------------------------------------------------------------
function normalise(raw: unknown[]): RawDocument[] {
return (raw as RawDocument[]).filter(Boolean);
}
/**
* Returns the current Swedish parliamentary session (riksmöte) in `YYYY/YY` format.
* The Swedish parliamentary year runs from October to September:
* - October–December of year N: session is `N/N+1`
* - January–September of year N: session is `N-1/N`
*
* Examples: 2025-11 → "2025/26", 2026-03 → "2025/26", 2026-10 → "2026/27"
*/
function currentRm(): string {
const now = new Date();
const year = now.getUTCFullYear();
const month = now.getUTCMonth() + 1; // 1-based
// Swedish parliamentary year runs roughly October–September
Iif (month >= 10) {
return `${year}/${String(year + 1).slice(-2)}`;
}
return `${year - 1}/${String(year).slice(-2)}`;
}
// ---------------------------------------------------------------------------
// Download orchestration
// ---------------------------------------------------------------------------
/**
* Download all relevant documents from riksdag-regering-mcp.
*
* Each document type is fetched with a configurable limit (default 20).
* Failures in individual calls are caught so a partial result is returned
* rather than failing the entire pipeline.
*
* @param client - MCPClient instance (caller-supplied for testability)
* @param options - Optional overrides for limits and riksmöte
*/
export async function downloadAllDocuments(
client: MCPClient,
options: { limit?: number; rm?: string } = {},
): Promise<DownloadResult> {
const start = Date.now();
const limit = options.limit ?? 20;
const rm = options.rm ?? currentRm();
const dataSources: string[] = [];
const data: DownloadedData = {
propositions: [],
motions: [],
committeeReports: [],
votes: [],
speeches: [],
questions: [],
interpellations: [],
};
// Run independent MCP fetches in parallel to reduce total latency while
// still collecting partial results on failure.
const fetchTasks = [
{
name: 'fetchPropositions',
source: 'get_propositioner',
fetch: () => client.fetchPropositions(limit, rm),
assign: (raw: RawDocument[]) => { data.propositions = raw; },
},
{
name: 'fetchMotions',
source: 'get_motioner',
fetch: () => client.fetchMotions(limit, rm),
assign: (raw: RawDocument[]) => { data.motions = raw; },
},
{
name: 'fetchCommitteeReports',
source: 'get_betankanden',
fetch: () => client.fetchCommitteeReports(limit, rm),
assign: (raw: RawDocument[]) => { data.committeeReports = raw; },
},
{
name: 'fetchVotingRecords',
source: 'search_voteringar',
fetch: () => client.fetchVotingRecords({ limit, rm }),
assign: (raw: RawDocument[]) => { data.votes = raw; },
},
{
name: 'searchSpeeches',
source: 'search_anforanden',
fetch: () => client.searchSpeeches({ limit, rm }),
assign: (raw: RawDocument[]) => { data.speeches = raw; },
},
{
name: 'fetchWrittenQuestions',
source: 'get_fragor',
fetch: () => client.fetchWrittenQuestions({ limit, rm }),
assign: (raw: RawDocument[]) => { data.questions = raw; },
},
{
name: 'fetchInterpellations',
source: 'get_interpellationer',
fetch: () => client.fetchInterpellations({ limit, rm }),
assign: (raw: RawDocument[]) => { data.interpellations = raw; },
},
] as const;
const fetchResults = await Promise.allSettled(
fetchTasks.map(task => task.fetch()),
);
fetchResults.forEach((result, index) => {
const task = fetchTasks[index]!;
if (result.status === 'fulfilled') {
try {
task.assign(normalise(result.value as unknown[]));
dataSources.push(task.source);
} catch (err) {
console.warn(
`[pre-analysis] ${task.name} post-processing failed:`,
err instanceof Error ? err.message : String(err),
);
}
} else {
console.warn(
`[pre-analysis] ${task.name} failed:`,
result.reason instanceof Error ? result.reason.message : String(result.reason),
);
}
});
const docCounts: Record<string, number> = {
propositions: data.propositions.length,
motions: data.motions.length,
committeeReports: data.committeeReports.length,
votes: data.votes.length,
speeches: data.speeches.length,
questions: data.questions.length,
interpellations: data.interpellations.length,
};
return {
data,
manifest: {
dataSources,
docCounts,
durationMs: Date.now() - start,
},
};
}
/**
* Flatten all downloaded document collections into a single array,
* deduplicated by a best-effort document identifier.
*
* The primary deduplication key is `dok_id` when present. If `dok_id` is
* missing or empty, the function falls back in order to:
* `dokument_id`, `id`, `dok_url`, `url`, `rel_dok_id`, `titel`, and `title`.
* The first non-empty string among these fields is used as the dedup key,
* and documents resolving to the same key are treated as duplicates.
*/
export function flattenDocuments(data: DownloadedData): RawDocument[] {
const all: RawDocument[] = [
...data.propositions,
...data.motions,
...data.committeeReports,
...data.votes,
...data.speeches,
...data.questions,
...data.interpellations,
];
const seen = new Set<string>();
return all.filter(doc => {
if (!doc) return false;
const record = doc as Record<string, unknown>;
const idCandidates = [
record['dok_id'],
record['dokument_id'],
record['id'],
record['dok_url'],
record['url'],
record['rel_dok_id'],
record['titel'],
record['title'],
];
const id = idCandidates.find((candidate): candidate is string =>
typeof candidate === 'string' && candidate.trim().length > 0,
)?.trim() ?? '';
if (!id || seen.has(id)) return false;
seen.add(id);
return true;
});
}
|