The APL Orchard

« first day (2444 days earlier) ← previous day next day → last day (341 days later) »

RubenVerg

7:44 AM

parsing APL, even a restricted subset, turns out to be much harder than expected

the parsing part of tinyapl might have to wait some more

xpqz

8:09 AM

@RubenVerg I recommend studying the Bunda-Gerth parser included in the dfns workspace (dfns.dyalog.com/n_parse.htm). It's an intuitive approach to parsing APL. A reasonably complete 'grammar' for Dyalog is here: dfns.dyalog.com/s_parse.htm and the orgiginal B-G paper here: dl.acm.org/doi/pdf/10.1145/384283.801081

If you're interested in implementing it it not-APL, I wrote a Python version here: github.com/xpqz/numpapl/blob/main/apl.py

ngn-apl takes a more traditional recursive decent approach (github.com/abrudz/ngn-apl) which is clean and also ok to follow.

ivy's parser (github.com/robpike/ivy) is also a good resource.

2 hours later…

RubenVerg

10:27 AM

@xpqz this doens't seem to exist

thanks for everythine else though! i better go study :)

xpqz

@RubenVerg Try now -- apparently, I'd left it private.

The parser should be complete, although the interpreter itself is only half-baked.

4 hours later…

Adám

2:27 PM

Announcement: BAA Vector Webinar in half an hour.

4 hours later…

Lapwing482

5:59 PM

Are there any resources on dealing with graphs in APL? Running through old AOC problems, anything with graph traversal totally stumps me. Even checking solutions on the wiki doesn't help.

1 hour later…

Adám

7:21 PM

@Lapwing482 Check out aplcart.info/?q=dfns%20graph but also Aaron Hsu's work, e.g.