venkatbala / lmptools Goto Github PK

• Create a changelog file and a workflow for automating package version increments
• Setup a version workflow

• In order to properly assign each atom to its respective molecule from the get go, while parsing a snapshot build the molecule linked list with its atoms
• This should aid when persisting the snapshot to database if requested

To ensure code quality, add workflow to run a pre-commit CI action on each pull request and push

Enable a feature to allow users to filter atoms based on their id's positions and other possible metrics to enable custom parsing of the dumpfile

• Provide options to either write all snapshots to a single file, or one file per selected snapshot
• Provide method to filter out atoms based on type and region

To allow converting the dumpfile snapshots to different file formats, enable custom writers that users can simply use while parsing the dump file

• After tests pass on develop, push the latest develop to release branch

Add support to enable chaining multiple writers one after the other. I.e enable persistence of more than one kind to be invoked one after the other while parsing the dump file

• To allow processing multiple snapshots at once or to enable computation that involve multiple snapshots create view to hold multiple snapshots in memory at the same time

Define an abstract template + with a implementation of a dump file parser.
The purpose of this should be to allow users to write their own parser and to primarily define what is absolutely required to parse a LAMMPS dump file properly

• Refactor the lmptools source files into meaninful sub-directories
• Update imports in various files
• Fix pre-commit ci bot

• to leverage multiple cores speed up sequential processing of a dump file by breaking the file into several disjoint chunks and parse each on independently
• Producer/Consumer pattern