Issues

• HermitCrab: one partial rule switches off the final-template prune for the whole grammar; Mbugwe chaanjiyɛ never finishes

  sillsdev/machine#643 · Filter by author johnml1135 opened on Nov 12, 2026

• HermitCrab analysis: store folded equivalent analyses as differences, not whole Words (per-word memory)

  Hermit Crabperformance
  sillsdev/machine#640 · Filter by author johnml1135 opened on Nov 12, 2026

• HermitCrab analysis: finish each stratum as it completes instead of keeping every state until synthesis (per-word memory)

  Hermit Crabperformance
  sillsdev/machine#639 · Filter by author johnml1135 opened on Nov 12, 2026

• Fst.ResultCompare: nondeterministic branch can order results inconsistently when Priorities arrays differ in length

  sillsdev/machine#638 · Filter by author johnml1135 opened on Nov 12, 2026

• HermitCrab: affix-process matching enumerates every choice over runs of optional segments

  Hermit Crabperformance
  sillsdev/machine#635 · Filter by author johnml1135 opened on Nov 11, 2026
  2

• HermitCrab: copy-agreement prune keeps matches with an empty copy, re-running the whole search

  Hermit Crabperformance
  sillsdev/machine#634 · Filter by author johnml1135 opened on Nov 11, 2026
  2

• HermitCrab allocation: FeatureStruct clone dominates; jemalloc not applicable

  sillsdev/machine#632 · Filter by author johnml1135 opened on Nov 11, 2026

• HermitCrab: allomorph environments are tested at each piece of a root split by an infix

  sillsdev/machine#630 · Filter by author johnml1135 opened on Nov 10, 2026
  1

• C# HermitCrab synthesis generates iiiaiii from aaaa for a simultaneous aa → iii rule, but its own analysis rejects that surface form

  sillsdev/machine#628 · Filter by author johnml1135 opened on Nov 10, 2026

• HermitCrab analysis can spin on simultaneous rewrites with incompatible environments

  sillsdev/machine#623 · Filter by author johnml1135 opened on Nov 7, 2026

• HermitCrab: analysis metathesis drops the middle context that synthesis keeps

  sillsdev/machine#619 · Filter by author johnml1135 opened on Nov 6, 2026

• HermitCrab: Stratum nulls Stratum on Remove, so FieldWorks named strata moves leave items with Stratum == null

  Hermit Crab
  sillsdev/machine#618 · Filter by author johnml1135 opened on Nov 3, 2026

• HermitCrab: no word reaches the MaxApplicationCount gate, and the mutant stack-overflow path has no end-to-end test

  Hermit Crab
  sillsdev/machine#616 · Filter by author johnml1135 opened on Nov 2, 2026

• PlaceMarkersUsfmUpdateBlockHandler gets confused with table cells

  Bug

  sillsdev/machine#614 · Filter by author pmachapman opened on Nov 1, 2026

• HermitCrab analysis: combined stack cut work 14-64% on four real grammars, but R2 is closed; re-measure without it (unmeasured)

  sillsdev/machine#611 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab: a restricted clitic adds 17.5% work steps, mostly host search that repeats per MSA (measured on the port)

  sillsdev/machine#610 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab: do literal-identity and natural-class restoration on hc.dll match the port's fixes? (unmeasured)

  sillsdev/machine#609 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab: #546 follow-ons (compiled slot-rule sharing, ParallelApplySlots) are unbuilt and unmeasured in work steps

  sillsdev/machine#608 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab synthesis: 98% of Mbugwe synthesis work goes to candidates whose root lookup hits but never match the surface

  sillsdev/machine#607 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab: 86 of 1,645 Mbugwe words hit the 2M step cap and carry about two-thirds of its work; causes not examined

  sillsdev/machine#606 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab analysis: dead-internal work is material only on Aweti (39.8%) and v1 (12.2%); shadow-mode summary not run (research lead)

  sillsdev/machine#605 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab analysis: one-template battery is 45% of work on a private grammar; FST-derived precheck ceiling unmeasured

  sillsdev/machine#604 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab synthesis: 25.6% of synthesis steps end in SurfaceFormMismatch; early-reject check not built (research lead)

  sillsdev/machine#603 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab analysis: excluded co-occurrence check is sound but cuts 0% of work on real grammars (parked)

  sillsdev/machine#602 · Filter by author johnml1135 opened on Oct 31, 2026

• HermitCrab: dead ends and measured-and-rejected speedups (record, from PR #490)

  sillsdev/machine#601 · Filter by author johnml1135 opened on Oct 31, 2026
  1