@qaecy/cue-cli 0.0.17 → 0.0.19

package/main.js

@@ -93,7 +93,7 @@ var init_worker_pool = __esm({
 
 // apps/desktop/cue-cli/src/main.ts
 var import_commander = require("commander");
-var import_fs7 = require("fs");
+var import_fs8 = require("fs");
 var import_path5 = require("path");
 
 // apps/desktop/cue-cli/src/variables.ts
@@ -6065,13 +6065,97 @@ async function syncHandler(options) {
   }
 }
 
+// apps/desktop/cue-cli/src/cue-cli-util-remove-rdf-star.ts
+var import_fs7 = require("fs");
+var import_zlib2 = require("zlib");
+
+// libs/js/rdf-tools/src/lib/nq-to-nt.ts
+var import_n35 = require("n3");
+var { quad, defaultGraph } = import_n35.DataFactory;
+
+// libs/js/rdf-tools/src/lib/remove-rdf-star.ts
+var import_n36 = require("n3");
+var import_stream2 = require("stream");
+function isRDFStarTerm(term) {
+  return term.termType === "Quad";
+}
+function removeRDFStar(inputStream, starCount) {
+  const parser = new import_n36.Parser({ format: "N-Quads*" });
+  const outputStream = new import_stream2.PassThrough();
+  const writer = new import_n36.StreamWriter({ format: "N-Quads" });
+  writer.pipe(outputStream);
+  let count = 0;
+  parser.parse(inputStream, (error, quad2, prefixes3) => {
+    if (error) {
+      outputStream.emit("error", error);
+      writer.end();
+      return;
+    }
+    if (quad2) {
+      if (!isRDFStarTerm(quad2.subject) && !isRDFStarTerm(quad2.object)) {
+        writer.write(quad2);
+      } else if (starCount) {
+        count++;
+        starCount(count);
+      }
+    } else {
+      console.log("Finished processing input stream of RDF*, ending writer.");
+      writer.end();
+    }
+  });
+  return outputStream;
+}
+
+// apps/desktop/cue-cli/src/cue-cli-util-remove-rdf-star.ts
+async function utilRemoveRdfStarHandler(options) {
+  const { input, output, verbose } = options;
+  const isGzipped = input.endsWith(".gz");
+  if (verbose)
+    console.info(`Input: ${input} (${isGzipped ? "gzipped" : "plain"})`);
+  if (verbose)
+    console.info(`Output: ${output}`);
+  const fileStream = (0, import_fs7.createReadStream)(input);
+  const inputStream = isGzipped ? fileStream.pipe((0, import_zlib2.createGunzip)()) : fileStream;
+  let removed = 0;
+  const cleanStream = removeRDFStar(inputStream, (count) => {
+    removed = count;
+    if (verbose && count % 1e3 === 0) {
+      console.info(`Removed RDF-star triples so far: ${count}`);
+    }
+  });
+  const writeStream = (0, import_fs7.createWriteStream)(output);
+  if (isGzipped) {
+    const gzip = (0, import_zlib2.createGzip)();
+    gzip.pipe(writeStream);
+    for await (const chunk of cleanStream) {
+      gzip.write(chunk);
+    }
+    gzip.end();
+    await new Promise((resolve2, reject) => {
+      writeStream.on("finish", resolve2);
+      writeStream.on("error", reject);
+      gzip.on("error", reject);
+    });
+  } else {
+    for await (const chunk of cleanStream) {
+      writeStream.write(chunk);
+    }
+    writeStream.end();
+    await new Promise((resolve2, reject) => {
+      writeStream.on("finish", resolve2);
+      writeStream.on("error", reject);
+    });
+  }
+  console.info(`Done. Removed ${removed} RDF-star triple(s). Output written to: ${output}`);
+}
+
 // apps/desktop/cue-cli/src/main.ts
 var packageJson;
 try {
-  packageJson = JSON.parse((0, import_fs7.readFileSync)((0, import_path5.join)(__dirname, "package.json"), "utf8"));
+  packageJson = JSON.parse((0, import_fs8.readFileSync)((0, import_path5.join)(__dirname, "package.json"), "utf8"));
 } catch {
   try {
-    packageJson = JSON.parse((0, import_fs7.readFileSync)((0, import_path5.join)(__dirname, "../package.json"), "utf8"));
+    packageJson = JSON.parse((0, import_fs8.readFileSync)((0, import_path5.join)(__dirname, "../package.json"), "utf8"));
   } catch {
     packageJson = { version: "0.0.0" };
     console.warn("Could not find package.json, using fallback version");
@@ -6084,4 +6168,5 @@ program.command("dump").description("Dump Cue Knowledge Graph data to file\n
 program.command("compare").description("Compares folder content to files already updated to Cue").requiredOption("-s, --space <id>", "Specify the space ID (required)").requiredOption("-p, --path <id>", "Specify the folder path (required)").option("-k, --key <api-key>", "Specify the API key (or set CUE_API_KEY env variable)").option("--provider <provider ID>", "Specify the provider ID (eg. sharepoint, drive, dropbox) or leave empty for default provider", "").option("-v, --verbose", "Enable verbose output", false).option("-e, --emulators", "Uses emulators for sync", false).option("-z, --zip", "Include zipped content (will temporarily unzip files with same logic as when syncing and delete them again after the comparison)", false).action(compareHandler);
 program.command("dump-processed").description("Dump processed files to local folder").requiredOption("-s, --space <id>", "Specify the space ID (required)").requiredOption("-p, --processor <id>", "Id of the processor to dump processed files from (required) [eg. writers-blob, processors-cad-files]").option("-k, --key <api-key>", "Specify the API key (or set CUE_API_KEY env variable)").option("-v, --verbose", "Enable verbose output", false).option("-e, --emulators", "Uses emulators for sync", false).action(dumpProcessedHandler);
 program.command("repair-ttl").description("Repair TTL files in the specified space").requiredOption("-s, --space <id>", "Specify the space ID (required)").requiredOption("-p, --processor <id>", "Id of the processor to repair TTL files (required) [eg. writers-blob, processors-cad-files]").requiredOption("-f, --from <subString>", "Substring (Regex) to replace in the TTL file (required)").requiredOption("-t, --to <substituteString>", "Substring to replace in the TTL file (required)").option("-k, --key <api-key>", "Specify the API key (or set CUE_API_KEY env variable)").option("-v, --verbose", "Enable verbose output", false).option("-e, --emulators", "Uses emulators for sync", false).action(repairTtlHandler);
+program.command("util-remove-rdf-star").description("Remove RDF-star (quoted) triples from an NQuads file. Supports .nq and .nq.gz input/output.").requiredOption("-i, --input <path>", "Input file path (.nq or .nq.gz)").requiredOption("-o, --output <path>", "Output file path (.nq or .nq.gz)").option("-v, --verbose", "Enable verbose output", false).action(utilRemoveRdfStarHandler);
 program.parse(process.argv);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@qaecy/cue-cli",
-  "version": "0.0.17",
+  "version": "0.0.19",
   "description": "Cue CLI for QAECY platform",
   "main": "main.js",
   "bin": {

package/readme.md

@@ -32,3 +32,17 @@ To sync the current dir to the space with id `<space-id>` under the provider id
 | `-v, --verbose`       | Enable verbose output                                                                                | `false`      |
 | `-e, --emulators`     | Use emulators for sync                                                                               | `false`      |
 | `-z, --zip`           | Include zipped content. Will be unzipped to `<zip_path>_unzipped`. Max uncompressed size: 500 MB, max recursion depth: 3. Cleans up unzipped files after sync. | `false`      |
+
+### util-remove-rdf-star
+
+Remove RDF-star (quoted triple) statements from an NQuads file. Supports both plain `.nq` and gzipped `.nq.gz` files. If the input is gzipped the output will also be gzipped.
+
+`npx @qaecy/cue-cli util-remove-rdf-star -i <input.nq.gz> -o <output.nq.gz> -v`
+
+#### Options
+
+| Option               | Description                             | Default |
+|----------------------|-----------------------------------------|---------|
+| `-i, --input <path>` | Input file path (`.nq` or `.nq.gz`) (required) | N/A |
+| `-o, --output <path>`| Output file path (`.nq` or `.nq.gz`) (required) | N/A |
+| `-v, --verbose`      | Enable verbose output                   | `false` |