initial Api extracotr

.gitignore

@@ -0,0 +1 @@
+.idea

main.py

@@ -0,0 +1,115 @@
+import csv, json, os, time, random, requests
+
+API_KEY="579b464db66ec23bdd000001acb35eb12c5943f068a59a162fc0d8fa"
+RID="5c2f62fe-5afa-4119-a499-fec9d604d5bd"
+BASE=f"https://api.data.gov.in/resource/{RID}"
+
+OUT="india_pincode_city_state.csv"
+STATE="state.json"
+PAGE=1000
+SAVE_EVERY_OFFSETS = 10000
+FLUSH_EVERY_ROWS = 5000
+
+def pad6(x):
+    s = str(x or "")
+    return ("000000" + s)[-6:]
+
+def load_offset():
+    if os.path.exists(STATE):
+        with open(STATE, "r", encoding="utf-8") as f:
+            return int(json.load(f).get("offset", 0))
+    return 0
+
+def save_offset(offset):
+    tmp = STATE + ".tmp"
+    with open(tmp, "w", encoding="utf-8") as f:
+        json.dump({"offset": offset}, f)
+    os.replace(tmp, STATE)
+
+offset = load_offset()
+
+new_file = not (os.path.exists(OUT) and os.path.getsize(OUT) > 0)
+f = open(OUT, "a", newline="", encoding="utf-8")
+w = csv.writer(f)
+if new_file:
+    w.writerow(["pincode", "locality", "city", "state"])
+    f.flush()
+
+s = requests.Session()
+s.headers.update({
+    "User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 Chrome/120 Safari/537.36",
+    "Accept": "application/json",
+    "Accept-Language": "en-US,en;q=0.9",
+    "Connection": "keep-alive",
+})
+
+backoff = 1.0
+rows = 0
+
+while True:
+    url = f"{BASE}?api-key={API_KEY}&format=json&limit={PAGE}&offset={offset}"
+
+    try:
+        r = s.get(url, timeout=(5, 45))
+
+        if r.status_code == 429:
+            ra = r.headers.get("Retry-After")
+            if ra and ra.isdigit():
+                sleep_s = int(ra)
+            else:
+                sleep_s = max(15, backoff) + random.random() * 2
+
+            print("429 at", offset, "sleep", round(sleep_s, 2), flush=True)
+            time.sleep(sleep_s)
+            backoff = min(backoff * 1.8, 300)  # cap 5 min
+            continue
+
+        r.raise_for_status()
+        data = r.json()
+
+    except Exception as e:
+        sleep_s = backoff + random.random() * 0.5
+        print("err at", offset, repr(e), "sleep", round(sleep_s, 2), flush=True)
+        time.sleep(sleep_s)
+        backoff = min(backoff * 1.7, 120)
+        continue
+
+    # sometimes API returns {"error": "..."} with 200
+    if isinstance(data, dict) and data.get("error") and not data.get("records"):
+        sleep_s = backoff + random.random() * 0.5
+        print("api error at", offset, data["error"], "sleep", round(sleep_s, 2), flush=True)
+        time.sleep(sleep_s)
+        backoff = min(backoff * 1.7, 120)
+        continue
+
+    recs = (data.get("records") or []) if isinstance(data, dict) else []
+    got = len(recs)
+    if not recs:
+        print("done at", offset, flush=True)
+        break
+
+    backoff = 1.0  # reset on success
+
+    for x in recs:
+        w.writerow([
+            pad6(x.get("pincode")),
+            (x.get("officename") or "").strip(),
+            (x.get("district") or "").strip(),
+            (x.get("statename") or "").strip(),
+        ])
+
+    rows += got
+    offset += got  # <-- replace offset += PAGE
+
+    if rows % FLUSH_EVERY_ROWS == 0:
+        f.flush()
+    if offset % SAVE_EVERY_OFFSETS == 0:
+        save_offset(offset)
+
+    if rows % 1000 == 0:
+        print("asked", PAGE, "got", len(recs), "api_limit", data.get("limit"), "api_count", data.get("count"))
+
+    # gentle pacing to avoid 429 storms (tune 0.2-1.0)
+   # time.sleep(0.25)
+f.close()
+print("final rows written (this run):", rows, flush=True)