From 552cb824889e146f4da0df682d96b1047b0e2402 Mon Sep 17 00:00:00 2001
From: Martin Czygan <martin.czygan@gmail.com>
Date: Thu, 15 Jul 2021 03:26:59 +0200
Subject: extra: cdx reshape

---
 extra/cdx/cdx_reshape.py | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)
 create mode 100644 extra/cdx/cdx_reshape.py

diff --git a/extra/cdx/cdx_reshape.py b/extra/cdx/cdx_reshape.py
new file mode 100644
index 0000000..4e96134
--- /dev/null
+++ b/extra/cdx/cdx_reshape.py
@@ -0,0 +1,19 @@
+
+# {"summary":{"last":"20131126234003","ok":"20131126234003","delta":0},"numRows":1,"line":"http://128.118.178.162/eps/pe/papers/0501/0501001.pdf"}
+
+import fileinput
+import json
+
+by_url = {}
+
+for line in fileinput.input():
+    line = line.strip()
+    doc = json.loads(line)
+    last = doc.get("summary", {}).get("last", "")
+    if not last:
+        continue
+    by_url[doc["line"]] = doc
+
+
+for k, v in sorted(by_url):
+    print("{}\t{}".format(k, v))
-- 
cgit v1.2.3