Issue 6685 - Offer incremental filter list downloads

abp/filters/renderer.py

 # This file is part of Adblock Plus <https://adblockplus.org/>,
 # Copyright (C) 2006-present eyeo GmbH
 #
 # Adblock Plus is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version 3 as
 # published by the Free Software Foundation.
 #
 # Adblock Plus is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
@@ skipping 166 matching lines @@
     _logger.info('Rendering: %s', name)
     lines, default_source = _get_and_parse_fragment(name, sources, top_source)
     lines = _process_includes(sources, default_source, [name], lines)
     for proc in [_process_timestamps, _insert_version, _remove_duplicates,
                  _validate]:
         lines = proc(lines)
     return lines
 
 
 def _split_list_for_diff(list_in):
-    filterlist, metadata, keys = set(), set(), set()
+    """Split a filter list into metadata, keys, and rules."""

[Vasily Kuznetsov, 2018/08/31 12:11:02]

Nit: strictly speaking this function now returns metadata and rules (the keys
are just part of metadata). The docstring could be updated.

[rhowell, 2018/08/31 21:21:02]

Ah, good point! I'll update that.

+    metadata = {}
+    rules = set()
     for line in parse_filterlist(list_in):
         if line.type == 'metadata':
-            metadata.add(line.to_string())
-            keys.add(line.key.lower())
+            metadata[line.key.lower()] = line
         elif line.type == 'filter':
-            filterlist.add(line.to_string())
-    return filterlist, metadata, keys
+            rules.add(line.to_string())
+    return metadata, rules
 
 
 def render_diff(base, latest):
     """Return a diff between two filter lists.
 
     Parameters
     ----------
     base : iterator of str
         The base (old) list that we want to update to latest.
     lastest : iterator  of str
         The latest (most recent) list that we want to update to.
 
     Returns
     -------
     iterable of str
         A diff between two lists (https://issues.adblockplus.org/ticket/6685)
 
     """
-    latest_fl, latest_md, latest_keys = _split_list_for_diff(latest)
-    base_fl, base_md, base_keys = _split_list_for_diff(base)
-
-    new_md = latest_md - base_md
-    removed_keys = base_keys - latest_keys
-    add_fl = latest_fl - base_fl
-    remove_fl = base_fl - latest_fl
+    latest_metadata, latest_rules = _split_list_for_diff(latest)
+    base_metadata, base_rules = _split_list_for_diff(base)
 
     yield '[Adblock Plus Diff]'
-    for item in new_md:
-        yield item
-    for key in removed_keys:
-        # If a special comment has been removed, enter it with a blank value
-        # so the client will set it back to the default value
-        yield '! {}:'.format(key)
-    for item in remove_fl:
-        yield '- {}'.format(item)
-    for item in add_fl:
-        yield '+ {}'.format(item)
+    for key, latest in latest_metadata.items():
+        base = base_metadata.get(key)
+        if not base or base.value != latest.value:
+            yield latest.to_string()
+    for key in set(base_metadata) - set(latest_metadata):
+        yield '! {}:'.format(base_metadata[key].key)
+    for rule in base_rules - latest_rules:
+        yield '- {}'.format(rule)
+    for rule in latest_rules - base_rules:
+        yield '+ {}'.format(rule)