WIP: Main loop api and singlelogparser class

[simonbowly]

This adds:

• SingleLogParser class to parse a full log by delegating to the section handlers
• Main loop with both parse("*.log").summary() and get_dataframe(["*.log"]) alternatives for compatibility with the old API

@maliheha there is still a bit to do here but could you take a look at single_log_parser.py and api.py?

Some other changes:

• Removed feather files, regression tests now run directly against the old code still in the repo
• Updated grblogtools.py to v1.3.1 (merged in master branch) to test against this code directly

---

• To see the specific tasks where the Asana app for GitHub is being used, see below:
  • https://app.asana.com/0/0/1201705063151537

[simonbowly]

Also - integrating the termination parser with the different termination messages from barrier and nodelogs is looking tricky. Switching the termination parser settings on the fly once we determine the model type doesn't seem to fit very well. What do you think of the following?

1. Include "Optimal objective" in the barrier parser
2. Include "Best objective/best bound/solution count" as well as "Optimal solution found" in the nodelog parser, since they are specific to models with a node log
3. Instead of the termination parser, have a status message parser which picks up "Time limit reached", "Interrupted" and so on (without interrupting other parsers).

In the case of a MIP where barrier solves the relaxation but the node processing times out, when merging all the log info, Status="TIME_LIMIT" from the status parser would then taken precedence over Status="OPTIMAL" coming from the barrier parser, and we would get the right result.

[maliheha]

Thanks for all the work! It looks very great!

I looked at the two modules api.py and single_log_parser.py and wrote three comments. In one of the comments, you will find a possible idea for addressing the termination parser.

There are a few TODOs in these tow files:

• TODO fill_default_parameters could be much cleaner now, without column regexes
• TODO extend this to also check multiple logs from one file.
• TODO extend this with other args from the old api
• Termination parser

I can take care of some of these. For example, if you agree with the termination idea, I can give it a try. I can also work on extending the logic to multiple logs in one file. We can probably turn the remaining TODOs into issues and work on them.

There are some inconsistency in the docstrings formats that we can take care of it at the very end.