Adding some tests for Arabic characters (have not ran these yet)

client/test/journal_client_test.ts

@@ -141,6 +141,24 @@ snapshotTest("JournalClient.read content test", async ({ assertSnapshot }) => {
   assertSnapshot(filtered_results, masks);
 });
 
+snapshotTest("JournalClient.read Arabic content test", async ({ assertSnapshot }) => {
+  const client = new JournalClient(BASE_URL, await makeJwt({}));
+
+  const req = {
+    journal: "acmeCo/arabic-greetings/pivot=00",
+    offset: "10",
+    endOffset: "1024",
+  };
+  const stream = (await client.read(req)).unwrap();
+  const docStream = parseJournalDocuments(stream!);
+  const results = await readStreamToEnd(docStream);
+
+  let filtered_results = results.filter(r=>!(r._meta as any).ack).slice(0,5)
+
+  const masks = ["/*/_meta/uuid", "/*/ts"];
+  assertSnapshot(filtered_results, masks);
+});
+
 Deno.test("JournalClient.read unauthorized prefix", async () => {
   const client = new JournalClient(BASE_URL, await makeJwt({}));
 

test.sh

@@ -63,6 +63,7 @@ export GATEWAY_PORT=28318
 
 export BUILD_ID=test-build-id
 export CATALOG_SOURCE="test/acmeCo/source-hello-world.flow.yaml"
+export CATALOG_SOURCE_ARABIC="test/acmeCo/arabic-source-hello-world.flow.yaml"
 
 # This is needed in order for docker run commands to work on ARM macs.
 export DOCKER_DEFAULT_PLATFORM="linux/amd64"
@@ -117,6 +118,12 @@ ${FLOW_BIN} api build \
   --build-id=${BUILD_ID} \
   --source=${CATALOG_SOURCE} || bail "Build failed."
 
+# Build the catalog for some minor multi language testing
+${FLOW_BIN} api build \
+  --build-db=${TESTDIR}/builds/${BUILD_ID} \
+  --build-id=${BUILD_ID} \
+  --source=${CATALOG_SOURCE_ARABIC} || bail "Build failed."
+
 log "Build finished"
 
 # Activate the catalog.

test/acmeCo/arabic-source-hello-world.flow.yaml

@@ -0,0 +1,20 @@
+storageMappings:
+  "":
+    stores:
+      - provider: GCS
+        bucket: example
+collections:
+  acmeCo/arabic-greetings:
+    schema: greetings.schema.yaml
+    key: [/ts]
+captures:
+  acmeCo/arabic-source-hello-world:
+    endpoint:
+      connector:
+        image: ghcr.io/estuary/source-hello-world:dev
+        config: source-hello-world.config.yaml
+    bindings:
+      - resource:
+          name: greetings
+          prefix: 'مرحبا #{}'
+        target: acmeCo/arabic-greetings