feat: Stream download, extract and parse datasets

src/services/archive/Archive.ts

@@ -0,0 +1,7 @@
+import { Duplex } from "node:stream";
+
+interface Archive {
+  extract(source: string): Duplex;
+}
+
+export default Archive;

src/services/archive/ArchiveFactory.ts

@@ -0,0 +1,12 @@
+import { Archive, ZipArchive, ArchiveType, GzipArchive } from "./";
+
+class ArchiveFactory {
+  static getArchive(archiveType: ArchiveType): Archive {
+    if (archiveType === ArchiveType.ZIP) return ZipArchive.instance;
+    if ([ArchiveType.GZIP, ArchiveType.GZ].includes(archiveType)) return GzipArchive.instance;
+
+    throw new Error("Unsupported archive type");
+  }
+}
+
+export default ArchiveFactory;

src/services/archive/ArchiveType.ts

@@ -0,0 +1,7 @@
+enum ArchiveType {
+  ZIP = ".zip",
+  GZIP = ".gzip",
+  GZ = ".gz",
+}
+
+export default ArchiveType;

src/services/archive/GzipArchive.ts

@@ -0,0 +1,12 @@
+import { createGunzip } from "node:zlib";
+import { Duplex } from "node:stream";
+import { Archive } from "./";
+
+class GzipArchive implements Archive {
+    public static instance: Archive = new GzipArchive();
+
+    public extract(source: string): Duplex {
+        return createGunzip();
+    }
+}
+export default GzipArchive;

src/services/archive/ZipArchive.ts

@@ -0,0 +1,15 @@
+import { Archive } from "./";
+import { Duplex } from "node:stream";
+import { ParseOne } from "unzipper";
+
+class ZipArchive implements Archive {
+  public static instance: Archive = new ZipArchive();
+
+  public extract(source: string): Duplex {
+    return ParseOne(new RegExp(source), {
+      forceStream: true,
+    });
+  }
+}
+
+export default ZipArchive;

src/services/archive/index.ts

@@ -0,0 +1,7 @@
+export { default as ArchiveType } from "./ArchiveType";
+
+export { default as ArchiveFactory } from "./ArchiveFactory";
+export { default as Archive } from "./Archive";
+
+export { default as ZipArchive } from "./ZipArchive";
+export { default as GzipArchive } from "./GzipArchive";