Update example

example

@@ -1,54 +1,173 @@
-{
-    "_id": "6a30f84db2bafb037face9c6",
-    "hasMore": true,
-    "limitReached": false,
-    "pageSize": 20,
-    "requestID": "6a30f84db2bafb037face9c6",
-    "requestType": "find",
-    "response": [
-        {
-            "_id": {
-                "$oid": "6a30f849eb71684d09ee22ae"
-            },
-            "message": "Success",
-            "data": {
-                "name": "SHAMEELA QAYOOM",
-                "fatherName": "ABDUL QAYYUM SHAH",
-                "motherName": "ABDUL QAYYUM SHAH",
-                "address1": "POST OFFICE KHAS SAMAL TEHSIL GOPAS DISTRICT GHAZAR",
-                "address2": "POST OFFICE KHAS SAMAL TEHSIL GOPAS DISTRICT GHAZAR",
-                "placeOfBirth": "GOPAS GHAZER",
-                "cnic": "7140132783264",
-                "aksaTransId": "8723608"
-            },
-            "LOG_CODE": "3",
-            "LOG_DESC": "service_name: custDataTranToEnglish, type: success",
-            "LOG_DATE": {
-                "$date": {
-                    "$numberLong": "1781594185116"
+import json
+import csv
+from pathlib import Path
+
+
+# ==========================
+# CONFIGURATION
+# ==========================
+
+INPUT_JSON = "input.json"          # Change to your JSON filename
+OUTPUT_CSV = "hbl_pipe_dataset.csv"
+
+# Fields to extract in order
+FIELDS = [
+    "name",
+    "fatherName",
+    "motherName",
+    "address1",
+    "address2",
+    "placeOfBirth",
+    "cnic"
+]
+
+
+# ==========================
+# HELPER FUNCTION
+# ==========================
+
+def clean_value(value):
+    """
+    Clean extracted values.
+    Keeps empty fields instead of removing them.
+    """
+
+    if value is None:
+        return ""
+
+    # Convert everything to string
+    return str(value).strip()
+
+
+def join_fields(data):
+    """
+    Convert dictionary fields into pipe-separated string.
+    """
+
+    return "|".join(
+        clean_value(data.get(field, ""))
+        for field in FIELDS
+    )
+
+
+# ==========================
+# MAIN PROCESSING
+# ==========================
+
+def process_json():
+
+    input_path = Path(INPUT_JSON)
+
+    if not input_path.exists():
+        raise FileNotFoundError(
+            f"JSON file not found: {INPUT_JSON}"
+        )
+
+    print("Loading JSON...")
+
+    with open(
+        INPUT_JSON,
+        "r",
+        encoding="utf-8"
+    ) as f:
+        json_data = json.load(f)
+
+
+    rows = []
+
+    total_records = 0
+    skipped_records = 0
+
+
+    # Option A:
+    # Outer object contains response array
+
+    response_list = json_data.get("response", [])
+
+
+    for item in response_list:
+
+        total_records += 1
+
+        try:
+
+            # Roman Urdu
+            roman_data = item.get(
+                "data",
+                {}
+            )
+
+            # Urdu
+            urdu_data = item.get(
+                "REQUEST_BODY",
+                {} 
+            )
+
+
+            if not urdu_data or not roman_data:
+                skipped_records += 1
+                continue
+
+
+            urdu_text = join_fields(
+                urdu_data
+            )
+
+            roman_text = join_fields(
+                roman_data
+            )
+
+
+            rows.append(
+                {
+                    "urdu": urdu_text,
+                    "roman": roman_text
                 }
-            },
-            "x-req-id": "10009706040",
-            "statusCode": {
-                "$numberInt": "200"
-            },
-            "statusMessage": "OK",
-            "CITIZEN_NUMBER": "7140132783264",
-            "REQUEST_BODY": {
-                "name": "شمیلہ قیوم",
-                "fatherName": "عبدالقیوم شاہ",
-                "motherName": "عبدالقیوم شاہ",
-                "address1": "ڈاک خانہ خاص،سمال،تحصیل گوپس،ضلع غذر",
-                "address2": "ڈاک خانہ خاص،سمال،تحصیل گوپس،ضلع غذر",
-                "placeOfBirth": "گوپس,غذر",
-                "cnic": "7140132783264"
-            },
-            "URL": "URL: https://apigateway.hbl.com:8343/api/v1/urdu/translate/english",
-            "REQUEST_HEADERS": {
-                "Content-Type": "application/json",
-                "x-channel-id": "HBL",
-                "x-req-id": "10009706040",
-                "x-sub-channel-id": "HBL",
-                "x-country-code": "PK"
-            }
-        },
+            )
+
+
+        except Exception as e:
+
+            print(
+                "Skipping record due to error:",
+                e
+            )
+
+            skipped_records += 1
+
+
+
+    print(f"Total records: {total_records}")
+    print(f"Created pairs: {len(rows)}")
+    print(f"Skipped: {skipped_records}")
+
+
+    # Save CSV
+
+    with open(
+        OUTPUT_CSV,
+        "w",
+        encoding="utf-8",
+        newline=""
+    ) as f:
+
+        writer = csv.DictWriter(
+            f,
+            fieldnames=[
+                "urdu",
+                "roman"
+            ]
+        )
+
+        writer.writeheader()
+
+        writer.writerows(rows)
+
+
+    print(
+        f"Saved dataset: {OUTPUT_CSV}"
+    )
+
+
+if __name__ == "__main__":
+    process_json()