collapse hashtags with different case

DocNow · Oct 10, 2022 · 41ccc04 · 41ccc04
1 parent 296ac4c
commit 41ccc04
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 4 deletions.
diff --git a/setup.py b/setup.py
@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="twarc-network",
-    version="0.1.1",
+    version="0.2.0",
     url="https://github.com/docnow/twarc-network",
     author="Ed Summers",
     author_email="ehs@pobox.com",

diff --git a/test_twarc_network.py b/test_twarc_network.py
@@ -91,7 +91,7 @@ def test_hashtags():
     )
     assert result.exit_code == 0
     graph = json.loads(result.output)
-    assert len(graph["nodes"]) == 383
+    assert len(graph["nodes"]) == 352
 
 
 def test_edges():

diff --git a/twarc_network/__init__.py b/twarc_network/__init__.py
@@ -167,8 +167,8 @@ def get_graph(infile, nodes_type, edge_types, id_as_label):
                 for ht1, ht2 in hashtag_pairs:
                     add_hashtag_edge(
                         g,
-                        "#" + ht1,
-                        "#" + ht2,
+                        "#" + ht1.lower(),
+                        "#" + ht2.lower(),
                         created_at_date,
                     )