refactor(cpp): ♻️ extract the row hash table from OperatorIndex - #303
Open
robertodr wants to merge 1 commit into
Open
refactor(cpp): ♻️ extract the row hash table from OperatorIndex#303robertodr wants to merge 1 commit into
robertodr wants to merge 1 commit into
Conversation
robertodr
requested review from
diagonal-hamiltonian,
fpietra and
ludmilaasb
as code owners
August 29, 2026 14:31
5 tasks
|
Docs preview: https://pr-303.monoprop-docs.pages.dev |
robertodr
force-pushed
the
split/02-row-hash-table
branch
from
August 29, 2026 14:35
403debb to
8ded3bd
Compare
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## split/01-baseline-tool #303 +/- ##
=======================================================
Coverage 97.70% 97.70%
=======================================================
Files 14 14
Lines 742 742
Branches 98 98
=======================================================
Hits 725 725
Misses 12 12
Partials 5 5
Flags with carried forward coverage won't be shown. Click here to find out more. |
OperatorIndex was two things: a packed position-list row representation and a keyless open-addressing index over those rows. RowHashTable is the second half, lifted out whole -- power-of-2 slots, linear probing, load factor 0.7, a 32-bit folded hash per slot used only as an equality pre-filter. Keyless is what makes the split possible: the table never stores or compares a key, so the hash and the equality test arrive as callables and the row representation stays entirely on the caller's side. find_batch keeps its pipeline by taking the row prefetch as a third callable -- deferring confirmation past the probe is the whole reason that prefetch has somewhere to go. The table's slot layout fixes for_each_slot's iteration order, which is the order of a propagator's evolved-term list and therefore its floating-point accumulation order, so this had to come out inert. Checked byte-wise against the previous implementation over 4000 random monomials: identical iteration order, find and find_batch results, clone order and memory_bytes. Assisted-by: ClaudeCode:claude-opus-5
robertodr
force-pushed
the
split/02-row-hash-table
branch
from
August 29, 2026 14:59
8ded3bd to
e4b1217
Compare
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
🤖 AI text below 🤖
Stacked on #302 — review that first; this PR's diff against it is
RowHashTable.hplus theOperatorIndexrewrite.Summary
Second of five PRs carved out of #226.
OperatorIndexcarries two things that have nothing to do with each other: a keyless open-addressing hash table, and the packed row storage the table indexes into. A second row backend is coming, and it needs the table and not the rows, so the table moves out first, on its own, with no behaviour change.RowHashTableis the table alone — power-of-two slots, linear probing, load factor 0.7, a 32-bit folded hash kept beside each slot as an equality pre-filter. It stores no keys: hashing, equality and prefetch arrive as callables from the owner, which is what lets a caller key its rows in whatever form it already has them in.OperatorIndexkeeps its own API and forwards to it.This is deliberately inert. The table's slot layout fixes
for_each_slotorder, which fixes evolved-term order, which fixes floating-point accumulation order — so a reordering here would move energies. Verified against main's implementation over 4000 random monomials: identical iteration order,find/find_batchresults, clone order andmemory_bytes.Changes
cpp/monoprop/detail/operator/RowHashTable.h: new. The table,TermIndexCeilingReached,kIndexCeiling,fold,find/find_batch/emplace/insert_distinct/insert_distinct_range,for_each_slot,reserve,slot_bytes.cpp/monoprop/detail/operator/OperatorIndex.h: keeps the rows, delegates the table.-195lines.cpp/monoprop/detail/operator/CMakeLists.txt: list the new header.Checklist
docs/,CONTRIBUTING.md) if neededCHANGELOG/ release notes updated if applicableCovered by the existing
operator_index_tests.cpp; the gate for a change of this kind isjust diff-baseline(#302), which must come out byte-identical.AI/LLM disclosure