#1750 shows a way to debloat an object by extracting its encoding / decoding logic into a serializer class.
In this case, the knowledge_plugins.functions.Function moved its serialize and parse_from_cmg methods.

There are several places along the angr codebase where this pattern could be applied:

• angr/knowledge_plugins/cfg/cfg_model.py
• `angr/knowledge_plugins/cfg/cfg

when browsing, i should be able to perf-check right from the browser and results should be siphoned back to the extension.

It would be nice to have some documentation on search operators and what are some of the common fields to search on (e.g. data_type, event_identifier, etc.). Also probably want to mention partial word matches requires globbing characters, etc.

I have dropped this feature when porting Thinlet UI to JavaFX (later, replaced by Swing.)

Thinlet Lule has "Tools > Export" menu. This wiki describes the feature.
https://github.com/DmitryKey/luke/wiki/Exporting-index-to-xml

I would like to re-think this feature (but cannot take enough time for this now.)
When (re)implementing this feature, file format selection (e.g. export to xml, json o

So instead of having to do rita import --config etc/rita.yaml ... we would do rita --config etc/rita.yaml import .... This makes it easier to modify most commands since --config is less likely to change than the rita subcommand. It also makes it possible to create aliases like:

• alias rita1=rita --config /etc/rita/config1.yaml
• alias rita2=rita --config /etc/rita/config2.yaml

Wher

enable cyclic-import check in pylint and fix the imports if there are problems

Currently, PySceneDetect uses the root logger with a basic configuration. It should be replaced with an explicitly named logger instead to avoid conflicts with other projects using scenedetect as a dependency.

Furthermore, some objects/functions take explicit logger objects as an argument. Modify them to use the named logger instead.

Lastly, ensure that when the package is imported, the