Skip to content

Meta: GFQL compiler/planner post-M4 execution DAG (active wave) #1259

Description

@lmeyerov

Summary

Post-M4 GFQL compiler/planner execution DAG meta-tracker.

Completed Active Lanes

Direct-Cypher Residual Cleanup (#1353)

Fresh #1353 snapshot after graphistry/tck-gfql#125, against pygraphistry 592e34a6 + tck-gfql 0fbaa42e:

  • tracked non-validation debt: 1
  • success_matches_expected: 1 (match5-8 bookkeeping)
  • success_wrong_rows: 0
  • unexpected_success_expected_error: 0
  • additional fastfail drift: with5-2 now returns expected rows while its direct-Cypher xfail contract still expects GFQLValidationError

Completed:

Remaining support:

Closed Large Deletion Wave

Post-#1419 concrete deletion follow-up:

Final deletion-wave receipt:

  • _execute_compiled_query_compat_non_union() is deleted.
  • Projection-level WITH/RETURN DISTINCT now plans natively through logical Distinct.
  • Remaining safe deletion surface did not approach the original >=2K LOC target; D8 source delta was net -24.
  • Coordinator decision: treat the original [META] GFQL deletion wave: remove 2K+ LOC of obsolete compatibility paths #1419 LOC target as revised/de-scoped. New deletion work needs a new concrete target.

Next Dispatch

Coordination Metas

Activity

  1. lmeyerov commented on May 5, 2026

    @lmeyerov
    ContributorAuthor

    Status update for active #1262 lane: T1 slice #1296 is implemented in PR #1298 (https://github.com/graphistry/pygraphistry/pull/1298).\n\nChecks are green (including tck-gfql/docs). Pending merge.

  2. lmeyerov commented on May 5, 2026

    @lmeyerov
    ContributorAuthor

    Execution lane update: #1296 is now merged via PR #1298\nhttps://github.com//pull/1298\n\nMerge commit: 559bdc7\n\nStatus: T1 schema-contract/canonical-naming lane complete; downstream lanes can build on the stabilized schema surface.

  3. lmeyerov commented on May 5, 2026

    @lmeyerov
    ContributorAuthor

    Lane update: #1302 (T2 under #1262) has landed via PR #1308.\n\nMerge receipt:\n- merged to master on 2026-05-05\n- merge commit: a23c34c\n- checks green at merge time (including tck-gfql + ReadTheDocs)\n\nActive type-system lane now shifts to #1300 (T3) under #1262.

  4. lmeyerov commented on May 6, 2026

    @lmeyerov
    ContributorAuthor

    Landed update: #1320 completed via #1321 (master cd1f3c8). Active children under #1262 should now be T4 (#1312) then T5 (#1311).

  5. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Coordination note: opened #1046 follow-on child slices for remaining public-surface work:\n- #1337 (public declarative schema model)\n- #1338 (schema inference + typed topology)\n- #1339 (public schema↔Arrow/plottable APIs)\n\nThese are scoped as post-#1262 lanes and can be scheduled in this DAG as independent execution slices.

  6. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Coordination update: PR #1340 was closed as no-op (empty net diff after removing a throwaway scoping file).\n\nNo code/docs contract changes landed from that PR. Execution planning for #1046 follow-on remains tracked in child issues:\n- #1337\n- #1338\n- #1339

  7. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Status update: #989 has landed via merged PR #1344 () on 2026-05-07.\n\nIn this execution DAG, can be moved from active lanes to completed/recently-landed.\n\nPR #1344 checks are green (including and docs/readthedocs).

  8. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Correction/update (clean formatting):

    Issue #989 has landed via merged PR #1344 (commit a8c9397) on 2026-05-07.

    In this execution DAG, #989 can move from active lanes to completed/recently-landed.

    PR #1344 checks are green, including tck-gfql and docs/readthedocs.

  9. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Execution DAG sync for #1046 follow-on work: added lane D #1345 (API contract alignment matrix + compatibility policy).\n\nScheduling intent:\n- A/B/C (#1337/#1338/#1339) are executable independently with local tests.\n- D is parallel/non-blocking and focuses on cross-lane contract drift + compatibility policy.

  10. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Coordination update: PR #1347 is green and lands the #1046 post-#1262 narrowing map (A/B/C + parallel non-blocking D #1345) in GFQL maintainer docs for execution clarity.

  11. lmeyerov commented on May 7, 2026

    @lmeyerov
    ContributorAuthor

    Status update: PR #1346 (child #1342 under umbrella #880) is green. Scope admits bounded ordered top-k MATCH re-entry (WITH ORDER BY ... LIMIT <literal>) with tests/changelog; guard remains for unbounded ORDER and parameterized LIMIT. No additional residual split needed from this slice.

  12. lmeyerov commented on May 11, 2026

    @lmeyerov
    ContributorAuthor

    Execution DAG update for #1412: PR #1425 merged on 2026-05-11 UTC with pygraphistry-side native GFQL row-shaping coverage for the IC8 recent-replies and IS7 message-replies reply-author shapes: #1425

    Leaving #1412 open in the active #880 lane until pyg-bench removes the adapter-local joins and a fresh residual report confirms these lanes no longer emit gfql partial under graphistry/pygraphistry#880.

  13. lmeyerov commented on May 11, 2026

    @lmeyerov
    ContributorAuthor

    Execution DAG update for #1416: PR #1423 merged on 2026-05-11 UTC with pygraphistry-side native GFQL regression coverage for the IS6 / message-forum recursive ancestor row-join lane:
    #1423

    Merge commit: fdd0f84bc8c0c03a59cb164514a96784cbe1c76d

    The covered case is the zero-hop Post input through REPLY_OF*0.., where the post ancestor must bind before joining forum/moderator rows.

    Leaving #1416 open in the active #880 lane until pyg-bench removes or bypasses the adapter-local recursive ancestor join fallback and a fresh residual report confirms this bucket no longer emits gfql partial under graphistry/pygraphistry#880.

  14. lmeyerov commented on May 11, 2026

    @lmeyerov
    ContributorAuthor

    #1412 pyg-bench closeout receipt

    Validation:

    Rerun attempt:

    • IC8 run id snb-interactive-0691677af8: adapter_missing, no probes; missing pinned upstream Cypher query path and SF1 dataset in this fresh pyg-bench checkout.
    • IS7 run id snb-interactive-9645ae5af8: adapter_missing, no probes; same missing upstream/SF1 dataset blocker.

    Remaining gate: merge graphistry/pyg-bench#9 and rerun residuals on a fully staged benchmark environment before closing #1412 or clearing the #880 umbrella.

  15. 21 remaining items

  16. lmeyerov commented on May 16, 2026

    @lmeyerov
    ContributorAuthor

    #1460 planner/execution-DAG lane landed via #1463

    Merge commit: 945777177ee502c67f839a89645092ea562210d0

    Compiler/planner impact:

    • LogicalPlanner now admits the simple input-free first-clause OPTIONAL MATCH shape.
    • Planned form is PatternMatch(optional=True, arm_id="top_level_optional_0").
    • Physical planning routes it through same_path; runtime uses the existing chain path and no longer uses generic compat for this shape.
    • _attach_logical_plan_route() keeps optional reentry deferred, so broader optional planning remains a future lane.

    Validation:

    • Local targeted + broad pandas GFQL passed.
    • DGX RAPIDS 25.02/26.02 full GFQL GPU/cuDF passed.
    • GitHub CI green, including tck-gfql.
    • Review skill loop converged with no carried findings.
  17. lmeyerov commented on May 16, 2026

    @lmeyerov
    ContributorAuthor

    #1461 execution-DAG closeout

    PR #1462 merged: #1462

    • Merge commit: 061326273ad2ad137273d0c006851984cf461649
    • Optional reentry native route is landed and covered.
    • Final PR head: ef8951a642563d7d8fcd2a594febd349108166c2
    • Validation: local focused optional-reentry suite green, DGX RAPIDS 25.02/26.02 exact cuDF amplification green, PR CI green including tck-gfql.

    Execution-DAG body updated:

  18. lmeyerov commented on May 16, 2026

    @lmeyerov
    ContributorAuthor

    Coordinator dispatch update: filed #1466 as D8 for #1419 final deletion/shrink closeout.

    The execution-DAG/deletion priority is now #1466: re-audit _execute_compiled_query_compat_non_union() reachability after #1448/#1458/#1463/#1462, then delete/shrink or revise the 2K LOC target. Treat typed schema (#1337/#1338/#1339/#1465/#1464) as a separate PR stack unless explicitly reprioritized.

  19. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Execution-DAG sync after #1466 / PR #1467 merged:

    Validation:

    • Local targeted GFQL planner/runtime: 38 passed
    • Local broad non-cuDF GFQL selection: 983 passed, 12 skipped, 112 deselected
    • DGX RAPIDS 25.02 GFQL/cuDF: 2377 passed, 22 skipped, 15 xfailed
    • DGX RAPIDS 26.02 GFQL/cuDF: 2377 passed, 22 skipped, 15 xfailed
    • GitHub CI green, including both tck-gfql runs
    • Review skill converged after 2 clean waves

    Execution-DAG implication:

  20. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Execution-DAG catch-up after #1467:

  21. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Deletion-wave closeout update:

    #1419 is closed/de-scoped after #1466/#1467. The compat-executor wrapper deletion is done; future code-deletion workers should target new concrete surfaces, not #1419/#1466.

  22. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    New deletion-focused pygraphistry dispatch set:

    These are concrete successor surfaces after #1419/#1466 closure. They are not replacements for #1457/#1337, which is already active in /home/lmeyerov/Work/pygraphistry5.

  23. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Meta sync for #1471:

  24. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Fresh #1353 sync:

    • pygraphistry origin/master: 592e34a66e40b79f79029d1620911f63158c99ed
    • tck-gfql main: 0fbaa42eb36c047370b6d8dd0337c8a5996c6f32
    • python -m tests.cypher_tck.report: tracked non-validation debt 1; success_matches_expected 1 (match5-8); success_wrong_rows 0; unexpected_success_expected_error 0
    • focused report/contract tests: 13 passed in 2.55s
    • broader direct-Cypher xfail contract found with5-2 now returns expected rows while contract expects GFQLValidationError

    No pygraphistry wrong-row runtime bug found. #1353 remains open pending graphistry/tck-gfql#129, the narrow tck-gfql promotion/accounting child for match5-8 + with5-2.

  25. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Execution DAG sync for #1478 / PR #1487:

    Tracker implication: #1486 is now blocked by the remaining #1495 path-alias carry residual and any #1469 decision; #1478 needs no further worker.

  26. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    tck-gfql#129 follow-up is open and green:

    Scope/result:

    • Promoted match5-8 and with5-2 from direct-Cypher success_matches_expected non-validation debt into row support bookkeeping.
    • Moved both keys into promoted-from-xfail history.
    • Removed with5-2 from the row-pipeline tranche-2 xfail lane.
    • No pygraphistry runtime changes were needed.
    • No pygraphistry#1490 blocker was found.

    PR-head receipts:

    • promoted-only direct-Cypher: 928 (rows 789, errors 139)
    • direct-Cypher total: 2888 / 3627
    • status accounting: supported 2905, xfail 722
    • tracked non-validation debt: 0
    • non-validation sweep: shown: 0 / 0
    • local full CI: 4011 passed, 722 xfailed

    Because the tck-gfql change is not merged yet, leaving this metaissue open for now. After PR #134 lands, pygraphistry#1353 should be closable from the direct-Cypher non-validation/wrong-row tracking perspective.

  27. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    tck-gfql#129 has landed via graphistry/tck-gfql#134.

    Final result:

    • Promoted match5-8 and with5-2 from direct-Cypher success_matches_expected non-validation debt into row support bookkeeping.
    • Direct-Cypher non-validation debt is now zero: report tracked debt 0; sweep shown: 0 / 0.
    • No pygraphistry runtime change was needed.
    • No pygraphistry#1490 blocker was found.
    • pygraphistry#1353 was closed from the direct-Cypher non-validation/wrong-row tracking perspective after the tck-gfql PR landed.
  28. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    #1490 / #1504 has landed, with paired TCK accounting merged in graphistry/tck-gfql#135.

    Coordinator note for this metaissue: the direct-Cypher hardcoded-node-identity blocker is no longer open. Future direct-Cypher cleanup should use the post-merge pygraphistry master + tck-gfql main snapshot, because #129 promotions (match5-8, with5-2) and #1490 residual drift are both now represented in TCK.

  29. lmeyerov commented on May 17, 2026

    @lmeyerov
    ContributorAuthor

    Closeout update for IC4 direct-Cypher residual: pygraphistry#1502 merged at 3c5af8e and closed #1496.

    Validation is complete: local targeted/regression subset, ruff/typecheck/diff-check, DGX RAPIDS 25.02 + 26.02 targeted cuDF regression, and full PR CI green. tck-gfql was reconciled via graphistry/tck-gfql#135; duplicate/conflicting graphistry/tck-gfql#136 was closed.

    Next action remains benchmark-side only: graphistry/pyg-bench#4 rerun against latest pygraphistry master to prove IC4/new-topics direct-Cypher conformance clears.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions