From f72f9059849a5c1524e6ed3d28657e1a1d4eb64d Mon Sep 17 00:00:00 2001
From: Sebastian Nagel <sebastian@commoncrawl.org>
Date: Thu, 16 Mar 2023 13:44:44 +0100
Subject: [PATCH] Drop support for Python 2.7, fixes #40 - update README - drop
 support for Python 2.x module urlparse (replaced by   urllib.parse)

---
 README.md            | 2 +-
 server_ip_address.py | 7 +------
 wat_extract_links.py | 7 +------
 3 files changed, 3 insertions(+), 13 deletions(-)

diff --git a/README.md b/README.md
index 5d329e7..e48cc09 100644
--- a/README.md
+++ b/README.md
@@ -39,7 +39,7 @@ pip install -r requirements.txt
 
 ## Compatibility and Requirements
 
-Tested with Spark 2.1.0 – 2.4.6 in combination with Python 2.7 or 3.5, 3.6, 3.7, and with Spark 3.0.0 - 3.2.1 in combination with Python 3.7, 3.8 and 3.9.
+Tested with with Spark 3.2.3 and 3.3.2 in combination with Python 3.8, 3.9 and 3.10. See the branch [python-2.7](/commoncrawl/cc-pyspark/tree/python-2.7) if you want to run the job on Python 2.7 and older Spark versions.
 
 
 ## Get Sample Data
diff --git a/server_ip_address.py b/server_ip_address.py
index 673792c..41f19cb 100644
--- a/server_ip_address.py
+++ b/server_ip_address.py
@@ -1,11 +1,6 @@
 import ujson as json
 
-try:
-    # Python2
-    from urlparse import urlparse
-except ImportError:
-    # Python3
-    from urllib.parse import urlparse
+from urllib.parse import urlparse
 
 from pyspark.sql.types import StructType, StructField, StringType, LongType
 
diff --git a/wat_extract_links.py b/wat_extract_links.py
index 9d7b3de..ad1e74d 100644
--- a/wat_extract_links.py
+++ b/wat_extract_links.py
@@ -4,12 +4,7 @@
 
 import ujson as json
 
-try:
-    # Python2
-    from urlparse import urljoin, urlparse
-except ImportError:
-    # Python3
-    from urllib.parse import urljoin, urlparse
+from urllib.parse import urljoin, urlparse
 
 from pyspark.sql.types import StructType, StructField, StringType