feat: add import data from gql script

.gitignore

@@ -14,6 +14,9 @@
 /vendor
 /wallets
 test-results.xml
+/scripts/import-data/bundles
+/scripts/import-data/transactions
+/scripts/import-data/parquet
 
 # Generated docs
 /docs/sqlite/bundles

docker-compose.yaml

@@ -108,6 +108,7 @@ services:
       - ENABLE_BACKGROUND_DATA_VERIFICATION=${ENABLE_BACKGROUND_DATA_VERIFICATION:-}
       - BACKGROUND_DATA_VERIFICATION_INTERVAL_SECONDS=${BACKGROUND_DATA_VERIFICATION_INTERVAL_SECONDS:-}
       - CLICKHOUSE_URL=${CLICKHOUSE_URL:-}
+      - BUNDLE_DATA_IMPORTER_QUEUE_SIZE=${BUNDLE_DATA_IMPORTER_QUEUE_SIZE:-}
     networks:
       - ar-io-network
     depends_on:

scripts/import-data/count-fetched-ids.ts

@@ -0,0 +1,134 @@
+/**
+ * AR.IO Gateway
+ * Copyright (C) 2022-2023 Permanent Data Solutions, Inc. All Rights Reserved.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+import * as fs from 'node:fs/promises';
+import path from 'node:path';
+import { fileURLToPath } from 'node:url';
+import { getFilesInRange } from './utils.js';
+const args = process.argv.slice(2);
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+
+let TRANSACTIONS_DIR = path.join(__dirname, 'transactions');
+let BUNDLES_DIR = path.join(__dirname, 'bundles');
+let MIN_BLOCK_HEIGHT = 0;
+let MAX_BLOCK_HEIGHT = Infinity;
+
+args.forEach((arg, index) => {
+  switch (arg) {
+    case '--transactionsDir':
+      if (args[index + 1]) {
+        TRANSACTIONS_DIR = args[index + 1];
+      } else {
+        console.error('Missing value for --transactionsDir');
+        process.exit(1);
+      }
+      break;
+    case '--bundlesDir':
+      if (args[index + 1]) {
+        BUNDLES_DIR = args[index + 1];
+      } else {
+        console.error('Missing value for --bundlesDir');
+        process.exit(1);
+      }
+      break;
+    case '--minHeight':
+      if (args[index + 1]) {
+        MIN_BLOCK_HEIGHT = parseInt(args[index + 1], 10);
+      } else {
+        console.error('Missing value for --minHeight');
+        process.exit(1);
+      }
+      break;
+    case '--maxHeight':
+      if (args[index + 1]) {
+        MAX_BLOCK_HEIGHT = parseInt(args[index + 1], 10);
+      } else {
+        console.error('Missing value for --maxHeight');
+        process.exit(1);
+      }
+      break;
+    default:
+      break;
+  }
+});
+
+const countIds = async ({
+  folder,
+  files,
+}: {
+  folder: string;
+  files: string[];
+}) => {
+  let counter = 0;
+  for (const file of files) {
+    const filePath = path.join(folder, file);
+    const ids = JSON.parse(await fs.readFile(filePath, 'utf-8')) as string[];
+    counter += ids.length;
+  }
+  return counter;
+};
+
+(async () => {
+  const transactionFiles = await getFilesInRange({
+    folder: TRANSACTIONS_DIR,
+    min: MIN_BLOCK_HEIGHT,
+    max: MAX_BLOCK_HEIGHT,
+  });
+  const bundleFiles = await getFilesInRange({
+    folder: BUNDLES_DIR,
+    min: MIN_BLOCK_HEIGHT,
+    max: MAX_BLOCK_HEIGHT,
+  });
+
+  console.log({ transactionFiles, bundleFiles });
+
+  if (transactionFiles.length > 0) {
+    const firstTransactionHeight = parseInt(
+      transactionFiles[0].split('.')[0],
+      10,
+    );
+    const lastTransactionHeight = parseInt(
+      transactionFiles[transactionFiles.length - 1].split('.')[0],
+      10,
+    );
+    const transactionCount = await countIds({
+      folder: TRANSACTIONS_DIR,
+      files: transactionFiles,
+    });
+
+    console.log(
+      `Total transactions from ${firstTransactionHeight} to ${lastTransactionHeight}: ${transactionCount}`,
+    );
+  }
+
+  if (bundleFiles.length > 0) {
+    const firstBundleHeight = parseInt(bundleFiles[0].split('.')[0], 10);
+    const lastBundleHeight = parseInt(
+      bundleFiles[bundleFiles.length - 1].split('.')[0],
+      10,
+    );
+    const bundleCount = await countIds({
+      folder: BUNDLES_DIR,
+      files: bundleFiles,
+    });
+
+    console.log(
+      `Total bundles from ${firstBundleHeight} to ${lastBundleHeight}: ${bundleCount}`,
+    );
+  }
+})();

scripts/import-data/export-parquet.ts

@@ -0,0 +1,138 @@
+/**
+ * AR.IO Gateway
+ * Copyright (C) 2022-2023 Permanent Data Solutions, Inc. All Rights Reserved.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+import path from 'node:path';
+import { fileURLToPath } from 'node:url';
+import { fetchLatestBlockHeight, fetchWithRetry } from './utils.js';
+const args = process.argv.slice(2);
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+
+let ARIO_ENDPOINT = 'http://localhost:4000';
+let ADMIN_KEY: string | undefined;
+let OUTPUT_DIR = path.join(__dirname, 'parquet');
+let MAX_FILE_ROWS = 1_000_000;
+let MIN_BLOCK_HEIGHT = 0;
+let MAX_BLOCK_HEIGHT: number | undefined;
+
+args.forEach((arg, index) => {
+  switch (arg) {
+    case '--adminKey':
+      if (args[index + 1]) {
+        ADMIN_KEY = args[index + 1];
+      } else {
+        console.error('Missing value for --adminKey');
+        process.exit(1);
+      }
+      break;
+    case '--arioNode':
+      if (args[index + 1]) {
+        ARIO_ENDPOINT = args[index + 1];
+      } else {
+        console.error('Missing value for --arioNode');
+        process.exit(1);
+      }
+      break;
+    case '--outputDir':
+      if (args[index + 1]) {
+        OUTPUT_DIR = args[index + 1];
+      } else {
+        console.error('Missing value for --outputDir');
+        process.exit(1);
+      }
+      break;
+    case '--minHeight':
+      if (args[index + 1]) {
+        MIN_BLOCK_HEIGHT = parseInt(args[index + 1], 10);
+      } else {
+        console.error('Missing value for --minHeight');
+        process.exit(1);
+      }
+      break;
+    case '--maxHeight':
+      if (args[index + 1]) {
+        MAX_BLOCK_HEIGHT = parseInt(args[index + 1], 10);
+      } else {
+        console.error('Missing value for --maxHeight');
+        process.exit(1);
+      }
+      break;
+
+    case '--maxFileRows':
+      if (args[index + 1]) {
+        MAX_FILE_ROWS = parseInt(args[index + 1], 10);
+      } else {
+        console.error('Missing value for --maxFileRows');
+        process.exit(1);
+      }
+      break;
+    default:
+      break;
+  }
+});
+
+(async () => {
+  if (ADMIN_KEY === undefined) {
+    throw new Error('Missing admin key');
+  }
+
+  if (MAX_BLOCK_HEIGHT === undefined) {
+    MAX_BLOCK_HEIGHT = await fetchLatestBlockHeight();
+  }
+
+  await fetchWithRetry(`${ARIO_ENDPOINT}/ar-io/admin/export-parquet`, {
+    method: 'POST',
+    headers: {
+      'Content-Type': 'application/json',
+      Authorization: `Bearer ${ADMIN_KEY}`,
+    },
+    body: JSON.stringify({
+      outputDir: OUTPUT_DIR,
+      startHeight: MIN_BLOCK_HEIGHT,
+      endHeight: MAX_BLOCK_HEIGHT,
+      maxFileRows: MAX_FILE_ROWS,
+    }),
+  });
+
+  console.log(
+    `Parquet export started from block ${MIN_BLOCK_HEIGHT} to ${MAX_BLOCK_HEIGHT}`,
+  );
+
+  let isComplete = false;
+
+  while (!isComplete) {
+    const response = await fetchWithRetry(
+      `${ARIO_ENDPOINT}/ar-io/admin/export-parquet/status`,
+      {
+        method: 'GET',
+        headers: {
+          Authorization: `Bearer ${ADMIN_KEY}`,
+        },
+      },
+    );
+
+    const data = await response.json();
+    isComplete = data.status === 'completed';
+
+    if (isComplete) {
+      console.log('Parque export finished!');
+      console.log(data);
+    } else {
+      await new Promise((resolve) => setTimeout(resolve, 5000));
+    }
+  }
+})();