Releases: DocNow/hydrator
0.0.6
The v0.0.6 release of Hydrator includes no new functionality but it does include vastly improved handling of large tweet id datasets (e.g. millions of tweets or more). Previously Hydrator would search through the file of tweet ids for the next set of 100 ids to hydrate on each request to Twitter. This was highly inefficient both in terms of time as well as disk and CPU utilization.
Hydrator now opens the file and iteratively reads ids in chunks of 100 until it is told to stop (by the user) or encounters a rate limit (from Twitter). Previously a heartbeat was used to regulate the requests going to Twitter, but now the application will ask for as many as it can until it is rate limited. This also should mean shorter hydration times even for smaller datasets.
In addition Hydrator also will allow you to hydrate multiple datasets concurrently. This doesn't let you get around the API rate limits set by Twitter, but it will let you actively hydrate more than one dataset at a time.
0.0.5
0.0.4
This release includes no new functionality (and hopefully no new bugs) but it represents a much needed refresh to the code base.
Hydrator is now based on Electron v8, (which is a big step up from Electron 1.7) as well as React v16. A new Electron boilerplate is being used electron-react-redux-boilerplate which simplifies the build process without requiring webpack.
The addition of action-electron-builder lets us build and publish Linux, Mac and Windows installers here using GitHub Actions after just adding this little YAML file to the Hydrator GitHub repository.
