Post B34oC1PR4vQBeUSmIa by aras@mastodon.gamedev.place
 (DIR) More posts by aras@mastodon.gamedev.place
 (DIR) Post #B34oByjX1kiJLwqHho by lritter@mastodon.gamedev.place
       2025-11-29T11:23:15Z
       
       0 likes, 0 repeats
       
       today: more debugging.#devlog #LRDL
       
 (DIR) Post #B34oC0CHaLJ1tPIl9s by dougbinks@mastodon.gamedev.place
       2025-11-29T11:44:50Z
       
       0 likes, 0 repeats
       
       @lritter It's only debugging if it's from the municipality of Buggingen, DE. Otherwise it's just sparkling printf's.
       
 (DIR) Post #B34oC1PR4vQBeUSmIa by aras@mastodon.gamedev.place
       2025-11-29T18:44:32Z
       
       0 likes, 0 repeats
       
       @dougbinks @lritter it’s only profiling if filings are done by the pros. Otherwise it’s just sprinkling QueryPerformanceCounters
       
 (DIR) Post #B34oC2VUzrrj3aJ8O8 by dougbinks@mastodon.gamedev.place
       2025-11-30T10:32:16Z
       
       1 likes, 0 repeats
       
       @aras @lritter This explains the Quake story - Carmack goes into a nail salon to get some pro filing and comes out with well optimized software *and* nine inch nails.
       
 (DIR) Post #B34oC6NUbZq13uanSq by lritter@mastodon.gamedev.place
       2025-11-29T12:11:02Z
       
       0 likes, 0 repeats
       
       done debugging. the conclusion is that i finally have to implement the SCC -> DAG unlooper that I had put off for later, making it a tomorrow-me problem.well guess what, i'm tomorrow-me and later is now. #devlog #LRDL
       
 (DIR) Post #B34oCDnH27cU3thz7Y by lritter@mastodon.gamedev.place
       2025-11-29T15:43:23Z
       
       0 likes, 0 repeats
       
       after 5 hours of experimenting i have to conclude that i don't exactly understand yet how i want to control subloops.#devlog #LRDL
       
 (DIR) Post #B34oCKmp2CPxkWYDtA by lritter@mastodon.gamedev.place
       2025-12-01T23:24:42Z
       
       0 likes, 0 repeats
       
       still unclear. it's stressing me out.wondering if i shouldn't schedule more like NOIR. #devlog #LRDL
       
 (DIR) Post #B34oCRd9aXqKyZ56I4 by lritter@mastodon.gamedev.place
       2025-12-02T12:48:05Z
       
       0 likes, 0 repeats
       
       think i have something now that allows to structure recursive actions better, and also permits rule-local aggregates.i like it because the top level is no longer special. every rule can open and factorize a new top-level context.this is still distinct from `merge` because merges continue factorization (exporting variables), whereas fold folds it (exporting relations), hence the name.also note the low editing distance between opt 1 and 2 (+2 move, +2 kw, +2 indent)#devlog #LRDL
       
 (DIR) Post #B34oCZ1W6MUTu9X9jk by lritter@mastodon.gamedev.place
       2025-12-02T14:33:17Z
       
       0 likes, 0 repeats
       
       step 1: refactored all toplevel handlers into handlers that operate on fold context. the module top level is then also a fold context.step 2: parser understands and builds fold expressions. #devlog #LRDL
       
 (DIR) Post #B34oCgknMPrtsblume by lritter@mastodon.gamedev.place
       2025-12-03T10:37:04Z
       
       0 likes, 0 repeats
       
       now come the harder parts. on the agenda today:step 3: understand and typecheck free variables imported from the parent ✅ step 4: export namespace from folds to remainder of rule ✅step 5: properly generate C++ code for folds ✅#devlog #LRDL
       
 (DIR) Post #B34oCox8sOUxJDbt8i by lritter@mastodon.gamedev.place
       2025-12-03T14:20:51Z
       
       0 likes, 0 repeats
       
       wee, this test runs now. we have support for aggregates.#devlog #LRDL