@tryformation/querylight-cli 0.2.4 → 0.2.5

package/README.md

@@ -55,6 +55,8 @@ Publish releases from semantic version tags such as `0.1.1`.
 
 The GitHub Actions publish workflow publishes `@tryformation/querylight-cli` to the public npm registry.
 
+The publish workflow builds the package and verifies that the built CLI JSON envelope reports the same version as `package.json` before it publishes.
+
 Configure npm trusted publishing for this repository before the first release. The publish workflow uses GitHub OIDC and does not use an `NPM_TOKEN` secret.
 
 ### Local Development with `npm link`
@@ -248,6 +250,15 @@ crawler:
 
 Set `crawl.maxConcurrentRequests` on a website or RSS source when one source needs a different limit.
 
+Control the default number of search results returned when `--top-k` is omitted:
+
+```yaml
+search:
+  defaultTopK: 50
+```
+
+For `qli search --source-type rss` with a time-window filter such as `--since`, `--until`, or `--publication-date-from`, `qli` uses `500` results when `--top-k` is omitted.
+
 ## Supported Sources
 
 Current source types:

package/dist/cli/main.js

@@ -16,7 +16,11 @@ import path from "path";
 import YAML from "yaml";
 
 // src/core/constants.ts
-var PACKAGE_VERSION = "0.2.3";
+import { createRequire } from "module";
+var require2 = createRequire(import.meta.url);
+var packageJson = require2("../../package.json");
+var PACKAGE_NAME = packageJson.name;
+var PACKAGE_VERSION = packageJson.version;
 var DEFAULT_WORKSPACE = ".kb";
 var DEFAULT_SHARED_MODEL_CACHE_DIR = "~/.qli/models/huggingface";
 var LEGACY_WORKSPACE_MODEL_CACHE_DIR = ".kb/models/huggingface";
@@ -49,6 +53,9 @@ var defaultConfig = () => ({
     maxContextChars: 12e3,
     citationStyle: "markdown"
   },
+  search: {
+    defaultTopK: 50
+  },
   retrieval: {
     defaultMode: "lexical",
     dense: {
@@ -70,12 +77,12 @@ var defaultConfig = () => ({
     }
   },
   crawler: {
-    defaultUserAgent: "querylight-cli/0.1",
+    defaultUserAgent: "querylight-cli",
     obeyRobotsTxt: true,
     rateLimitMs: 1e3,
     maxConcurrentRequests: 5,
     renderJs: false,
-    retentionDays: 365,
+    retentionDays: 30,
     fetchArticles: true
   },
   limits: {
@@ -119,6 +126,10 @@ async function loadConfig(workspacePath, configPath) {
       ...defaults.rag,
       ...parsed.rag ?? {}
     },
+    search: {
+      ...defaults.search,
+      ...parsed.search ?? {}
+    },
     retrieval: {
       ...defaults.retrieval,
       ...parsed.retrieval ?? {},
@@ -2052,7 +2063,7 @@ async function fetchUrlDocument({
   publicationDate
 }) {
   const headers = {
-    "user-agent": source.crawl?.userAgent ?? "querylight-cli/0.1"
+    "user-agent": source.crawl?.userAgent ?? "querylight-cli"
   };
   if (previous?.httpCache?.etag) {
     headers["if-none-match"] = previous.httpCache.etag;
@@ -2351,7 +2362,7 @@ async function purgeDocuments(workspacePath, documentIds, documents) {
 async function fetchFeedText(source) {
   const response2 = await fetch(source.uri, {
     headers: {
-      "user-agent": source.crawl?.userAgent ?? "querylight-cli/0.1"
+      "user-agent": source.crawl?.userAgent ?? "querylight-cli"
     }
   });
   if (!response2.ok) {
@@ -4316,6 +4327,17 @@ function searchDateRanges(options) {
   }
   return entries;
 }
+function resolveSearchTopK(optionsTopK, sourceTypes, dateRanges, defaultTopK) {
+  const explicitTopK = parseOptionalPositiveInteger(optionsTopK, "--top-k");
+  if (explicitTopK !== void 0) {
+    return explicitTopK;
+  }
+  const includesRss = (sourceTypes ?? []).includes("rss");
+  if (includesRss && dateRanges.length > 0) {
+    return 500;
+  }
+  return defaultTopK;
+}
 async function resolveWorkspace(options) {
   return path22.resolve(options.workspace ?? DEFAULT_WORKSPACE);
 }
@@ -4629,7 +4651,7 @@ Examples:
     progress?.("info", "Rebuild complete");
     emit(global.json, capture, response("rebuild", workspace, data), `Processed ${ingest.processedSources} sources, wrote ${chunk.chunksWritten} chunks`);
   });
-  program.command("search").description("Search the built index and return ranked matching documents or chunks. Use search-json for raw JSON DSL queries.").argument("[query]", "Text query. Omit it to list the latest matching documents.").option("--top-k <n>", "Maximum number of results to return.", "12").option("--source <sourceIds>", "Restrict results to one or more source ids. Use comma-separated values.").option("--source-name <names>", "Restrict results to one or more source names. Use comma-separated values.").option("--source-type <types>", `Restrict results to one or more source types. Use comma-separated values: ${SOURCE_TYPE_LIST.join(", ")}`).option("--uri-prefix <prefixes>", "Restrict results to one or more URI prefixes. Use comma-separated values.").option("--tag <tags>", "Restrict results to one or more source tags. Use comma-separated values.").option("--metadata <key=value...>", "Restrict results to sources with matching metadata.").option("--since <date>", "Shortcut for --publication-date-from.").option("--until <date>", "Shortcut for --publication-date-to.").option("--changed-since <date>", "Only include documents changed on or after this date.").option("--has-publication-date", "Only include documents with a publication date.").option("--publication-date-from <date>", "Only include documents published on or after this date.").option("--publication-date-to <date>", "Only include documents published on or before this date.").option("--first-seen-at-from <date>", "Only include documents first seen on or after this date.").option("--first-seen-at-to <date>", "Only include documents first seen on or before this date.").option("--last-seen-at-from <date>", "Only include documents last seen on or after this date.").option("--last-seen-at-to <date>", "Only include documents last seen on or before this date.").option("--last-changed-at-from <date>", "Only include documents changed on or after this date.").option("--last-changed-at-to <date>", "Only include documents changed on or before this date.").option("--crawled-at-from <date>", "Only include documents crawled on or after this date.").option("--crawled-at-to <date>", "Only include documents crawled on or before this date.").option("--retrieval <mode>", `Retrieval mode: ${RETRIEVAL_MODE_LIST.join(", ")}`).option("--show-chunks", "Return chunk-level matches when available.").addHelpText("after", `
+  program.command("search").description("Search the built index and return ranked matching documents or chunks. Use search-json for raw JSON DSL queries.").argument("[query]", "Text query. Omit it to list the latest matching documents.").option("--top-k <n>", "Maximum number of results to return. Defaults to search.defaultTopK in config.yaml. RSS searches with a time window use 500 when omitted.").option("--source <sourceIds>", "Restrict results to one or more source ids. Use comma-separated values.").option("--source-name <names>", "Restrict results to one or more source names. Use comma-separated values.").option("--source-type <types>", `Restrict results to one or more source types. Use comma-separated values: ${SOURCE_TYPE_LIST.join(", ")}`).option("--uri-prefix <prefixes>", "Restrict results to one or more URI prefixes. Use comma-separated values.").option("--tag <tags>", "Restrict results to one or more source tags. Use comma-separated values.").option("--metadata <key=value...>", "Restrict results to sources with matching metadata.").option("--since <date>", "Shortcut for --publication-date-from.").option("--until <date>", "Shortcut for --publication-date-to.").option("--changed-since <date>", "Only include documents changed on or after this date.").option("--has-publication-date", "Only include documents with a publication date.").option("--publication-date-from <date>", "Only include documents published on or after this date.").option("--publication-date-to <date>", "Only include documents published on or before this date.").option("--first-seen-at-from <date>", "Only include documents first seen on or after this date.").option("--first-seen-at-to <date>", "Only include documents first seen on or before this date.").option("--last-seen-at-from <date>", "Only include documents last seen on or after this date.").option("--last-seen-at-to <date>", "Only include documents last seen on or before this date.").option("--last-changed-at-from <date>", "Only include documents changed on or after this date.").option("--last-changed-at-to <date>", "Only include documents changed on or before this date.").option("--crawled-at-from <date>", "Only include documents crawled on or after this date.").option("--crawled-at-to <date>", "Only include documents crawled on or before this date.").option("--retrieval <mode>", `Retrieval mode: ${RETRIEVAL_MODE_LIST.join(", ")}`).option("--show-chunks", "Return chunk-level matches when available.").addHelpText("after", `
 Examples:
   qli search "pricing api limits"
   qli search "authentication" --top-k 20 --tag docs
@@ -4642,22 +4664,27 @@ Examples:
 Notes:
   lexical works without vector models.
   dense, sparse, and hybrid require the relevant index artifacts to exist.
+  When you omit --top-k, qli uses search.defaultTopK from config.yaml. The default workspace value is 50.
+  RSS searches with a time window default to 500 results when you omit --top-k.
   Use search-json when you want the raw Querylight 0.11 JSON DSL and hit format.
   When you omit the query, qli returns the latest matching documents sorted by publication date.`).action(async function command(query, options) {
     const global = this.optsWithGlobals();
     const workspace = await resolveWorkspace({ workspace: global.workspace });
+    const config = await loadConfig(workspace, global.config);
+    const sourceTypes = parseSourceTypes(options.sourceType);
+    const dateRanges = searchDateRanges(options);
     const result = await searchIndex({
       workspacePath: workspace,
       query: query ?? "",
-      topK: Number(options.topK),
+      topK: resolveSearchTopK(options.topK, sourceTypes, dateRanges, config.search.defaultTopK),
       sourceIds: parseCommaSeparatedList(options.source),
       sourceNames: parseCommaSeparatedList(options.sourceName),
-      sourceTypes: parseSourceTypes(options.sourceType),
+      sourceTypes,
       uriPrefixes: parseCommaSeparatedList(options.uriPrefix),
       hasPublicationDate: Boolean(options.hasPublicationDate),
       tags: parseCommaSeparatedList(options.tag),
       metadata: (options.metadata ?? []).map(parseKeyValue).map(([key, value]) => ({ key, value })),
-      dateRanges: searchDateRanges(options),
+      dateRanges,
       retrievalMode: parseRetrievalMode(options.retrieval),
       showChunks: Boolean(options.showChunks)
     });

package/dist/core/constants.d.ts

@@ -1,5 +1,5 @@
-export declare const PACKAGE_NAME = "@tryformation/querylight-cli";
-export declare const PACKAGE_VERSION = "0.2.3";
+export declare const PACKAGE_NAME: string;
+export declare const PACKAGE_VERSION: string;
 export declare const DEFAULT_WORKSPACE = ".kb";
 export declare const DEFAULT_SHARED_MODEL_CACHE_DIR = "~/.qli/models/huggingface";
 export declare const LEGACY_WORKSPACE_MODEL_CACHE_DIR = ".kb/models/huggingface";

package/dist/index.js

@@ -22,6 +22,11 @@ import path from "path";
 import YAML from "yaml";
 
 // src/core/constants.ts
+import { createRequire } from "module";
+var require2 = createRequire(import.meta.url);
+var packageJson = require2("../../package.json");
+var PACKAGE_NAME = packageJson.name;
+var PACKAGE_VERSION = packageJson.version;
 var DEFAULT_SHARED_MODEL_CACHE_DIR = "~/.qli/models/huggingface";
 var LEGACY_WORKSPACE_MODEL_CACHE_DIR = ".kb/models/huggingface";
 
@@ -53,6 +58,9 @@ var defaultConfig = () => ({
     maxContextChars: 12e3,
     citationStyle: "markdown"
   },
+  search: {
+    defaultTopK: 50
+  },
   retrieval: {
     defaultMode: "lexical",
     dense: {
@@ -74,12 +82,12 @@ var defaultConfig = () => ({
     }
   },
   crawler: {
-    defaultUserAgent: "querylight-cli/0.1",
+    defaultUserAgent: "querylight-cli",
     obeyRobotsTxt: true,
     rateLimitMs: 1e3,
     maxConcurrentRequests: 5,
     renderJs: false,
-    retentionDays: 365,
+    retentionDays: 30,
     fetchArticles: true
   },
   limits: {
@@ -123,6 +131,10 @@ async function loadConfig(workspacePath, configPath) {
       ...defaults.rag,
       ...parsed.rag ?? {}
     },
+    search: {
+      ...defaults.search,
+      ...parsed.search ?? {}
+    },
     retrieval: {
       ...defaults.retrieval,
       ...parsed.retrieval ?? {},
@@ -1069,7 +1081,7 @@ async function fetchUrlDocument({
   publicationDate
 }) {
   const headers = {
-    "user-agent": source.crawl?.userAgent ?? "querylight-cli/0.1"
+    "user-agent": source.crawl?.userAgent ?? "querylight-cli"
   };
   if (previous?.httpCache?.etag) {
     headers["if-none-match"] = previous.httpCache.etag;
@@ -1368,7 +1380,7 @@ async function purgeDocuments(workspacePath, documentIds, documents) {
 async function fetchFeedText(source) {
   const response = await fetch(source.uri, {
     headers: {
-      "user-agent": source.crawl?.userAgent ?? "querylight-cli/0.1"
+      "user-agent": source.crawl?.userAgent ?? "querylight-cli"
     }
   });
   if (!response.ok) {

package/dist/types/models.d.ts

@@ -173,6 +173,9 @@ export type WorkspaceConfig = {
         maxContextChars: number;
         citationStyle: "markdown";
     };
+    search: {
+        defaultTopK: number;
+    };
     retrieval: {
         defaultMode: RetrievalMode;
         dense: DenseVectorModelConfig;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@tryformation/querylight-cli",
-  "version": "0.2.4",
+  "version": "0.2.5",
   "description": "Querylight CLI for building and querying local knowledge bases.",
   "license": "MIT",
   "homepage": "https://github.com/formation-res/querylight-cli#readme",
@@ -36,7 +36,8 @@
     "test:watch": "vitest",
     "lint": "tsc --noEmit",
     "check": "npm run lint && npm test",
-    "prepublishOnly": "npm run check && npm run build"
+    "prepublishOnly": "npm run check && npm run build && npm run verify:release-version",
+    "verify:release-version": "node scripts/assert-release-version.mjs"
   },
   "dependencies": {
     "@huggingface/transformers": "^3.8.1",

package/scripts/assert-release-version.mjs

@@ -0,0 +1,48 @@
+import assert from "node:assert/strict";
+import { mkdtemp, rm } from "node:fs/promises";
+import os from "node:os";
+import path from "node:path";
+import { spawn } from "node:child_process";
+import packageJson from "../package.json" with { type: "json" };
+
+function run(command, args, options = {}) {
+  return new Promise((resolve, reject) => {
+    const child = spawn(command, args, {
+      stdio: ["ignore", "pipe", "pipe"],
+      ...options
+    });
+    let stdout = "";
+    let stderr = "";
+
+    child.stdout.on("data", (chunk) => {
+      stdout += String(chunk);
+    });
+    child.stderr.on("data", (chunk) => {
+      stderr += String(chunk);
+    });
+    child.on("error", reject);
+    child.on("close", (code) => {
+      if (code === 0) {
+        resolve({ stdout, stderr });
+        return;
+      }
+      reject(new Error(`${command} ${args.join(" ")} failed with exit code ${code}\n${stderr}`));
+    });
+  });
+}
+
+const workspaceRoot = await mkdtemp(path.join(os.tmpdir(), "qli-release-version-"));
+const workspacePath = path.join(workspaceRoot, ".kb");
+
+try {
+  const { stdout } = await run("node", ["dist/cli/main.js", "init", "--workspace", workspacePath, "--json"], {
+    cwd: new URL("..", import.meta.url)
+  });
+  const parsed = JSON.parse(stdout);
+
+  assert.equal(parsed.ok, true, "Expected qli init --json to succeed");
+  assert.equal(parsed.version, packageJson.version, `Built CLI reported version ${parsed.version}, expected ${packageJson.version}`);
+  process.stdout.write(`Verified built CLI version ${parsed.version}\n`);
+} finally {
+  await rm(workspaceRoot, { recursive: true, force: true });
+}