fix: Save history in main index (to mimic previous behaviour)

archivebox/core/models.py

@@ -47,6 +47,10 @@ class Snapshot(models.Model):
     def as_link(self) -> Link:
         return Link.from_json(self.as_json())
 
+    def as_link_with_details(self) -> Link:
+        from ..index import load_link_details
+        return load_link_details(self.as_link())
+
     @cached_property
     def bookmarked(self):
         return parse_date(self.timestamp)

archivebox/main.py

@@ -560,8 +560,8 @@ def add(urls: Union[str, List[str]],
         archive_links(new_links, overwrite=False, out_dir=out_dir)
     else:
         return all_links
-
-    write_static_index([link.as_link() for link in all_links], out_dir=out_dir)
+    
+    write_static_index([link.as_link_with_details() for link in all_links], out_dir=out_dir)
     return all_links
 
 @enforce_types
@@ -638,7 +638,7 @@ def remove(filter_str: Optional[str]=None,
 
     remove_from_sql_main_index(snapshots=snapshots, out_dir=out_dir)
     all_snapshots = load_main_index(out_dir=out_dir)
-    write_static_index([link.as_link() for link in all_snapshots], out_dir=out_dir)
+    write_static_index([link.as_link_with_details() for link in all_snapshots], out_dir=out_dir)
     log_removal_finished(all_snapshots.count(), to_remove)
     
     return all_snapshots
@@ -695,7 +695,7 @@ def update(resume: Optional[float]=None,
 
     # Step 4: Re-write links index with updated titles, icons, and resources
     all_links = load_main_index(out_dir=out_dir)
-    write_static_index([link.as_link() for link in all_links], out_dir=out_dir)
+    write_static_index([link.as_link_with_details() for link in all_links], out_dir=out_dir)
     return all_links
 
 @enforce_types

tests/test_add.py

@@ -63,3 +63,14 @@ def test_overwrite_flag_is_accepted(process, disable_extractors_dict):
     )
     assert 'unrecognized arguments: --overwrite' not in arg_process.stderr.decode("utf-8")
     assert 'favicon' in arg_process.stdout.decode('utf-8'), 'archive methods probably didnt run, did overwrite work?'
+
+def test_add_updates_history_json_index(tmp_path, process, disable_extractors_dict):
+    subprocess.run(
+        ["archivebox", "add", "--depth=0", "http://127.0.0.1:8080/static/example.com.html"],
+        capture_output=True,
+        env=disable_extractors_dict,
+    )
+
+    with open(tmp_path / "index.json", "r") as f:
+        output_json = json.load(f)
+    assert output_json["links"][0]["history"] != {}