This gist contains a summary of the progress from my GSoC 2018 Project: UD Annotatrix.
- complete refactor of the source code
- migrate Python/Django server to Node/Express in order to better integrate with the JavaScript front end
- migrate from global browser-native vanilla JavaScript to ES2015+ modules with
gulp
/babel
/webpack
pipeline - migrate from vanilla HTML to
EJS
-generated HTML (withgulp
and command line scripts to manage this)
- implementation of the
notatrix
inter-format tool- support for detecting, tokenizing, parsing, and generating multiple dependency formats
-
Brackets
-
CG3
-
CoNLL-U
-
Plain text
-
SDParse
- parameter-list/serial
-
- extensible architecture for adding new formats (e.g.
apertium stream
) - corpus import/export in/out of any of these formats
- robust error handling for invalid data
- command line interface for loading/exporting files
- native management of
- multiword tokens
- ambiguous analyses
- splitting of tokens (multiword or normal)
- combining tokens (into a multiword token)
- merging tokens (into a single token)
- enhanced dependencies
- sentence labels
- filter sentences by label
- add/remove/edit label names & colors
- native management of sentence IDs with searching capabilities
- support for detecting, tokenizing, parsing, and generating multiple dependency formats
- implementation of a complete
mocha
testing suite - real-time collaboration support
- sentence editing
- chat
- current-sentence tracking
- mouse tracking
- locking of nodes during edit
- GitHub account integration
- login to account
- edit document from git url
-
github.com
scheme -
raw.githubusercontent.com
scheme
-
- fork repo using GitHub API
- make pull requests using GitHub API
Links