Drasil Research Group Meeting, Monday, June 24, 3:30 pm, ITB/225

[smiths]

The time for this meeting may have to change. I might have another conflicting meeting. I will post a notification if the time changes.

The agenda is as follows:

1. @B-rando1 discussion - does @balacij have advice on tracking stuff down in drasil-code?
2. @NoahCardoso discussion - how best to proceed with mermaid for diagrams?
3. @BilalM04 discussion -cross-file references in mdBook
4. Next meeting Wednesday (July 3) at 11:00 am.

Building out the specifics of the agenda will depend on what is most pressing for the meeting attendants. @B-rando1, @NoahCardoso and @BilalM04, as we discussed please post comments below to provide an update on what you have been up to this week, highlighting any problems that are blocking your progress. If you could point to the specific issues and/or PRs you are working on, that would be great. The update can be done Thursday afternoon or on Friday.

[B-rando1]

For my weekly update:

• I got around to adding better support for 1-indexed lists (Add support for 1-indexed languages #3810). After some digging to understand the semantics of some list functions in GOOL, this went pretty smoothly.
• Yesterday I started working on a fix for list slicing with negative step values in certain languages (Negative Step in listSlice (GOOL) #3809). I made the necessary changes in Swift, and I think fixing it for other languages should be similar. I'm not perfectly happy with some of the code the fix generates, and I think if there's a way to tell at generation-time if an SValue is a litInt, I would be able to get rid of a lot of the non-idiomatic code that the fix currently generates. I don't currently know if there is a way to do this, so hopefully I can find something.
• The main thing I was working on this week was adding to Choices (Redesigning Choices to enable new input patterns #3781). This has been very slow going, as every change involves dozens of functions and it's not always apparent how they need to change.
  • I looked into interleaving. A couple of weeks ago already I had it working where if Unbundled Interleaved was selected, then parsing and validation would be done (not interleaved) in the same function. When I got back to it on Tuesday to try interleaving them, I ran into some issues. For one thing, I wasn't sure if constraints were storing all the values they depend on. Fortunately, @NoahCardoso found some functions in Extract.hs that should help with that. Now I just need to get all the inputs, derived inputs, and constraints organized in the right way so that I can do something similar to getExecOrder to put the code blocks in the right order. My attempts so far suggest that this will not be easy due to the complex structure of the code and data involved.
  • I also started extending the Bundled options. So far I have only added Interleaved, which is at the same stage as for Unbundled. This mostly went well, but I'm having an issue where the constructor thinks it needs an extra parameter if Interleaved is selected. I've traced the problem pretty far and seem to be getting close to discovering the reason, but I'm not quite there yet.

I don't have anything blocking my progress, but the complexity of tracking stuff down in drasil-code is slowing me down quite a bit. I might need help getting all the data formatted properly for making Interleaved work, but other than that I think I just need to keep digging.

[BilalM04]

Weekly Update

• Worked on Extract duplicate strings into InternalConcepts #3642.
  • @balacij assigned to me an old PR that had been stale for a while and no longer being maintained.
  • Brought up to date with main, resolved merge conflicts, and addressed the old comments and requested changes in the thread.
• Found underlying problem for Percent (%) symbol is being rendered in italics in the HTML SRS #3748.
  • Findings are detailed in Percent (%) symbol is being rendered in italics in the HTML SRS #3748 (comment) and Percent (%) symbol is being rendered in italics in the HTML SRS #3748 (comment).
• Worked on Experiment with generating an mdBook-based variant of the SRS #3722.
  • Cleaned up the code for pExpr to remove any duplication and fixed any issues in printing expressions.
  • Utilized suggestions from Rendering bibliography in Markdown #3787 to render the bibliography in pure Markdown rather than HTML.
  • Completed the rendering of all LayoutObjs into Markdown. All of the components are now rendered as desired, along with a clean and organized raw Markdown code.
  • Created a working prototype for splitting the SRS into separate files, consistent with the manually ported example.
  • Created a draft PR (Generating an mdBook-based variant of the SRS #3814) to begin receiving feedback on my implementation.

Next Steps / Agenda

The next steps for #3722 are outlined in #3814 (comment). I am currently working with @balacij to figure out how to deal with cross-file references (#3796 (reply in thread)).

Currently, the most pressing issue would be the cross-file references in mdBook. @balacij would know best whether it’s beneficial to discuss this issue in the meeting, or if it’s more suitable for us to continue discussing it individually as we have been doing.

[NoahCardoso]

For weekly update

• Closed Merge addRe and addI, and mulI and mulRe #3153. Jason gave me some feedback and I was able to finish the issue
• Reviewed some Mermaid Mermaid for Diagrams #3804
• Working on implementing Strict Data in Drasil, contributes to Add stan and weeder to our dev. workflow #3710
• I spent most of my week debugging Strict Data in Haskell Add Strict Data to Drasil #3813. Learned how to read stack trace and debug recursive loops in Haskell. One of the major issues is an infinite loop that occurs in some of the body.hs of examples that have a section in their symbMap. This loop will take more time to fix and is deeply rooted within the project. There is also some self-recursion created in glassbr.unitals which I am currently fixing. @balacij has been helping me with the debugging. I plan on fixing this Strict Data implementation through small pr's as Jason suggested.
• For next week I will be working more with Jason to make some of the examples in Mermaid Mermaid for Diagrams #3804.

[JacquesCarette]

Comments:

• @B-rando1 do add comments and make small refactors to drasil-code to help future people. Whatever information would have been useful to you, please write it down in comments (or the README or ...)
• @BilalM04 there may well be more strings that ought to get pulled out like that.
• @NoahCardoso yes please, tiny PRs, even if it takes multiple ones to get there, are appreciated.

You are all working on rather non-trivial issues and making good progress. This is wonderful.

[smiths]

Yes, great summaries. We should definitely continue doing this each week. I particularly like insights that help set the meeting agenda. It is really helpful to know what problems are blocking and what you plan on working on for the next week.

I've created specific agenda items for everyone. As always, feel free to add additional discussion points to the agenda.

I am going to have to move our meeting start time to 3:30 pm because of a conflict with the teaching-track search committee. I believe we will be able to still meet in our usual room. I'll let you know if there needs to be a change.

[smiths]

All agenda items were covered.