Handle list of tags in JSON, and be more clever about comma vs. space

archivebox/parsers/generic_json.py

@@ -66,9 +66,14 @@ def parse_generic_json_export(json_file: IO[str], **_kwargs) -> Iterable[Link]:
             elif link.get('name'):
                 title = link['name'].strip()
 
-            tags = ''
-            if link.get('tags'):
-                tags = link.get('tags').replace(' ',',')
+            # if we have a list, join it with commas
+            tags = link.get('tags')
+            if type(tags) == list:
+                tags = ','.join(tags)
+            elif type(tags) == str:
+                # if there's no comma, assume it was space-separated
+                if ',' not in tags:
+                    tags = tags.replace(' ', ',')
 
             yield Link(
                 url=htmldecode(url),

tests/mock_server/templates/example.json

@@ -1 +1,6 @@
-[{"href":"http://127.0.0.1:8080/static/example.com.html","description":"Example","extended":"","meta":"18a973f09c9cc0608c116967b64e0419","hash":"910293f019c2f4bb1a749fb937ba58e3","time":"2014-06-14T15:51:42Z","shared":"no","toread":"no","tags":"Tag1 Tag2","trap":"http://www.example.com/should-not-exist"}]
+[
+{"href":"http://127.0.0.1:8080/static/example.com.html","description":"Example","extended":"","meta":"18a973f09c9cc0608c116967b64e0419","hash":"910293f019c2f4bb1a749fb937ba58e3","time":"2014-06-14T15:51:42Z","shared":"no","toread":"no","tags":"Tag1 Tag2","trap":"http://www.example.com/should-not-exist"},
+{"href":"http://127.0.0.1:8080/static/iana.org.html","description":"Example 2","extended":"","meta":"18a973f09c9cc0608c116967b64e0419","hash":"910293f019c2f4bb1a749fb937ba58e3","time":"2014-06-14T15:51:43Z","shared":"no","toread":"no","tags":"Tag3,Tag4 with Space"},
+{"href":"http://127.0.0.1:8080/static/shift_jis.html","description":"Example 2","extended":"","meta":"18a973f09c9cc0608c116967b64e0419","hash":"910293f019c2f4bb1a749fb937ba58e3","time":"2014-06-14T15:51:44Z","shared":"no","toread":"no","tags":["Tag5","Tag6 with Space"]},
+{"href":"http://127.0.0.1:8080/static/title_og_with_html","description":"Example 2","extended":"","meta":"18a973f09c9cc0608c116967b64e0419","hash":"910293f019c2f4bb1a749fb937ba58e3","time":"2014-06-14T15:51:45Z","shared":"no","toread":"no"}
+]

tests/test_add.py

@@ -110,12 +110,19 @@ def test_json(tmp_path, process, disable_extractors_dict):
 
     urls = list(map(lambda x: x[0], urls))
     assert "http://127.0.0.1:8080/static/example.com.html" in urls
+    assert "http://127.0.0.1:8080/static/iana.org.html" in urls
+    assert "http://127.0.0.1:8080/static/shift_jis.html" in urls
+    assert "http://127.0.0.1:8080/static/title_og_with_html" in urls
     # if the following URL appears, we must have fallen back to another parser
     assert not "http://www.example.com/should-not-exist" in urls
 
     tags = list(map(lambda x: x[0], tags))
     assert "Tag1" in tags
     assert "Tag2" in tags
+    assert "Tag3" in tags
+    assert "Tag4 with Space" in tags
+    assert "Tag5" in tags
+    assert "Tag6 with Space" in tags
 
 def test_json_with_leading_garbage(tmp_path, process, disable_extractors_dict):
     with open('../../mock_server/templates/example.json.bad', 'r', encoding='utf-8') as f: