#121: prune compiler Snowflake dialect support by wjduenow · Pull Request #127 · wjduenow/SignalForge

wjduenow · 2026-05-26T15:16:52Z

Summary

Make the prune compiler emit valid Snowflake SQL purely from the Dialect value object — no branching on dialect name, no warehouse SDK import under signalforge/prune/. Part of epic #118; depends on the #119 skeleton.

Closes #121.

Changes

Dialect extended with five declarative SQL-fragment fields (all BigQuery-defaulted so BigQuery output is byte-identical): sample_row_hash_expr, timestamp_literal_template, date_literal_template, quote_qualified_per_component, sample_cte_alias. SNOWFLAKE_DIALECT sets the Snowflake forms.
Compiler consumes them — fold-then-quote per identifier_case (graduated from declared-but-unused → load-bearing; Snowflake folds to UPPER), per-component qualified-name quoting ("DB"."SCH"."T"), HASH(*) sample predicate, ::TIMESTAMP/::DATE literal casts.
sample_cte_alias — Snowflake uses the quoted "sample" because SAMPLE is a reserved keyword (WITH sample AS is a syntax error there). This bug was caught in the Quality Gate by a gated sqlglot parse-guard, not by snapshots.
QUALIFY deliberately NOT wired — unique keeps the dialect-portable GROUP BY … HAVING; supports_qualify stays forward-compat metadata.
11 byte-exact Snowflake snapshot fixtures + snapshot tests; AST import-confinement guard for prune/; gated @pytest.mark.snowflake fakesnow execution + sqlglot parse validation.

Testing

Default suite: 2245 passed (BigQuery snapshots byte-identical — the load-bearing invariant).
Gated: uv run pytest -m snowflake --no-cov → 19 passed (fakesnow rule-semantic execution of the 4 built-ins + sqlglot parse of all 11 fixtures).
ruff + ruff format + pyright clean. CI green across 3.11 / 3.12 / 3.13.
4 code-review passes (no blocking bugs beyond the QG-fixed reserved-word issue) + fixes.
Real-Snowflake HASH(*) semantics / case-folding / sampling deferred to test+docs: fakesnow harness + FakeSnowflakeClient + gated live e2e (TPCH_SF1) + ops docs #124's live harness (DEC-005).

Compounding Update

.claude/rules/prune-engine.md — full Dialect-field list the compiler reads; identifier_case graduation; the snapshot-pins-shape-not-validity lesson; Snowflake subsection.
docs/prune-ops.md — Snowflake compiler dialect section (case-folding + residual, HASH() cross-version reproducibility caveat, validation tiers).
CLAUDE.md — public-API surface for the new Dialect fields + compiler Snowflake path.
.claude/rules/warehouse-adapters.md — Dialect now carries prune-compiler SQL-fragment templates.

🤖 Generated with Claude Code

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

coderabbitai · 2026-05-26T15:17:01Z

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

🗂️ Base branches to auto review (4)

feature/.*
bug/.*
hotfix/.*
feat/.*

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 9a4bdd36-06c2-48a3-bca5-873ae4d679c3

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

🔍 Trigger review

📝 Walkthrough

Walkthrough

This PR implements complete Snowflake SQL code generation for the prune compiler by extending the Dialect data model with SQL-fragment templates and refactoring the compiler to be fully dialect-driven, eliminating BigQuery-specific hardcoding while preserving byte-identical BigQuery output via defaults.

Changes

Snowflake Dialect Support for Prune Compiler

Layer / File(s)	Summary
Dialect Model Extension `src/signalforge/warehouse/models.py`, `tests/warehouse/test_models.py`	`Dialect` class adds five SQL-fragment template fields (`sample_row_hash_expr`, `timestamp_literal_template`, `date_literal_template`, `quote_qualified_per_component`, `sample_cte_alias`) with BigQuery defaults; `SNOWFLAKE_DIALECT` overrides these with Snowflake-specific expressions and quoting behavior; three new unit tests validate the new fields on both dialects and verify backward-compatible construction.
Compiler Rendering Refactor - Dialect-Driven `src/signalforge/prune/compiler.py` (lines 134–265, 277–326)	Core rendering functions (`_fold_identifier`, `_quote`, `_qualified_table_name`, `_render_partition_filter`, `_render_sample_cte`, `_wrap_with_sample_or_partition`) now accept `dialect` parameter and drive all identifier folding, quoting, and literal template rendering from dialect fields instead of hardcoded BigQuery logic; per-component qualification handling added for Snowflake.
Built-in Test Compilers - Dialect Integration `src/signalforge/prune/compiler.py` (lines 341–719, 854–902)	Five test compilers (`_compile_not_null`, `_compile_unique`, `_compile_accepted_values`, `_compile_relationships`, `_compile_custom_sql`) and `_compile_test` dispatch refactored to accept and thread `dialect` parameter through identifier quoting, table qualification, and sample/partition wrapping; replaces prior `quote_char` plumbing.
Snowflake SQL Fixtures & Snapshot Tests `tests/fixtures/prune/compiled_sql/snowflake/`, `tests/prune/test_compiler.py` (lines 1055–1399)	Twelve Snowflake SQL fixture files (not_null, unique, accepted_values, relationships, custom_sql variants in full and sample modes) using UPPER-cased identifiers, double-quote marks, hash-based sample predicates, and Snowflake cast literals; snapshot tests (US-003) assert byte-exact output; guard tests enforce fixture conventions (double-quotes, HASH predicate).
Unit Tests - Dialect-Specific Behavior `tests/prune/test_compiler.py` (lines 1055–1173)	New US-002 test block validates `_quote` folding/quoting for Snowflake vs BigQuery, `_qualified_table_name` per-component vs whole-path quoting, Snowflake hash-based sample predicates, Snowflake datetime/date partition filter cast forms, and relationships SQL quoting expectations.
Fakesnow Semantic Validation `tests/prune/test_compiler_fakesnow.py`	New gated (`pytest.mark.snowflake`) test module with eight execution tests validating failure-count semantics (not_null, unique, accepted_values, relationships) against fakesnow in-memory Snowflake, plus a fixture parsing gate that validates all Snowflake SQL syntax via sqlglot Snowflake dialect.
Import Guard & Test Configuration `tests/prune/test_compiler_import_guard.py`, `pyproject.toml`	New test module enforces zero `snowflake` or `google.cloud` imports under `src/signalforge/prune/` via AST-based detector; `pyproject.toml` adds `fakesnow>=0.9` dependency and configures `snowflake` pytest marker as gated/maintainer-only category (excluded by default).
Documentation & Planning `.claude/rules/prune-engine.md`, `.claude/rules/warehouse-adapters.md`, `CLAUDE.md`, `docs/prune-ops.md`, `plans/super/121-prune-snowflake-dialect.md`	Clarifies dialect-driven compiler contract; documents Snowflake identifier folding, hash determinism caveats, and validation tiers (byte-exact snapshots + fakesnow + sqlglot parsing); updates public API surface for Snowflake support; comprehensive super-plan for issue `#121` with architecture, refinements, and detailed work items.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

Possibly related issues

wjduenow/SignalForge#121: This PR fully implements the Snowflake compiler support described in issue #121, extending Dialect and making the compiler dialect-driven.

Possibly related PRs

wjduenow/SignalForge#125: Both PRs modify the warehouse-level Dialect and SNOWFLAKE_DIALECT definitions; this PR extends them for compiler SQL generation while the other introduces the SnowflakeAdapter skeleton.
wjduenow/SignalForge#117: Both PRs refactor src/signalforge/prune/compiler.py—the other PR modifies test compilation paths, while this PR comprehensively refactors all rendering to be dialect-driven.

Poem

A rabbit hops through warehouses wide,
From BigQuery plains to Snowflake's side—
No branching chains, just dialect's grace,
Templates fold and identifiers embrace,
The compiler now dances with every dialect face! 🐰❄️

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	Docstring coverage is 80.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly identifies the main change: adding Snowflake dialect support to the prune compiler, which is the central objective of this PR.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

codecov-commenter · 2026-05-26T15:18:32Z

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…ields Add four BigQuery-defaulted declarative fields to the frozen Dialect dataclass (DEC-001 of #121): sample_row_hash_expr, timestamp_literal_template, date_literal_template, and quote_qualified_per_component. Defaults reproduce BigQuery SQL byte-for-byte so every existing construction site stays valid unedited. Set SNOWFLAKE_DIALECT's four values (DEC-002): ABS(HASH(*)), '{value}'::TIMESTAMP, '{value}'::DATE, per-component quoting. Leave POSTGRES_DIALECT at the BigQuery defaults with a docstring note (DEC-007) that they are corrected when the Postgres adapter's warehouse ops land. Refresh the Dialect class docstring to describe what the prune compiler reads each new field for. Add unit tests pinning the SNOWFLAKE_DIALECT values, the BIGQUERY_DIALECT defaults, and that constructing a Dialect with only the five original fields still succeeds. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…-fragment fields

Make signalforge.prune.compiler emit warehouse-correct SQL purely from the Dialect value object — no branching on dialect name, no warehouse-SDK import under prune/. - _quote/_qualified_table_name now take the Dialect: fold identifiers per identifier_case (upper/lower/preserve) before quoting (DEC-003); branch on quote_qualified_per_component (Snowflake "DB"."SCH"."T" vs BigQuery whole- path) (DEC-002). - _render_sample_cte renders the hash-mod predicate from dialect.sample_row_hash_expr (Snowflake ABS(HASH(*)) vs BigQuery FARM_FINGERPRINT). - _render_partition_filter renders datetime/date via the dialect's literal templates (Snowflake '...'::TIMESTAMP vs BigQuery TIMESTAMP('...')). - Thread dialect (not bare quote_char) through _wrap_with_sample_or_partition and every _compile_* helper. BigQuery's identifier_case="preserve" + quote_qualified_per_component=False make the refactor a no-op for BigQuery: all 11 compiled_sql/*.sql snapshots stay byte-identical (zero fixture edits) — the regression gate (DEC-001). Adds tests/prune/test_compiler_import_guard.py (DEC-008): AST scan asserting no snowflake / google.cloud import under src/signalforge/prune/, with a planted-violation self-check and a no-false-positive check. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…ompiler

Add byte-exact Snowflake snapshot fixtures (captured from real compiler output with SNOWFLAKE_DIALECT) and snapshot tests covering the four built-in test types (full + sample modes) and custom_sql (single-table full, single-table sample, multi-table full-scan). - 11 new fixtures under tests/fixtures/prune/compiled_sql/snowflake/: "-quoted, per-component qualified names ("FAKE_PROJECT"."DATASET"."ORDERS"), UPPER-folded identifiers, MOD(ABS(HASH(*)), <bucket>) < 1 sample predicate. - Snapshot tests assert compiled == fixture (byte-exact) per variant. - custom_sql single-table sample test confirms the #116 materialised-sample substitution invariant under the Snowflake quote char: the body reads from the `sample` CTE alias and never the source table. - Guard tests: Snowflake fixtures contain `"` and never a backtick; sample fixtures contain HASH(*) and never FARM_FINGERPRINT. The 11 existing BigQuery fixtures are unchanged; compiler logic untouched (US-002). Full validation green (2245 passed). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

… + tests

Add `fakesnow` as a dev/test dependency behind a NEW gated `@pytest.mark.snowflake` marker and a gated test that feeds the compiler's real emitted Snowflake SQL into an in-memory fakesnow connection. - pyproject.toml: add `fakesnow>=0.9` to `[dependency-groups].dev` and `[project.optional-dependencies].dev` (kept in sync); register the `snowflake` marker; add `and not snowflake` to the default `addopts` deselection so the default `pytest` run does NOT collect it. - tests/prune/test_compiler_fakesnow.py: for each built-in test type (not_null / unique / accepted_values / relationships), compile the failing-rows SELECT with SNOWFLAKE_DIALECT, create a tiny matching fakesnow table, run the SQL wrapped as the adapter does (`SELECT COUNT(*) AS failures FROM (<sql>) AS t`), and assert it executes with the engineered failing-row shape. Determinism engineered by RULE SEMANTICS, not value-equality (testing-signal.md): a NULL row → failures >= 1, a duplicate → failures >= 1, etc. No HASH() value assertions. scope="full" only — fakesnow's DuckDB backend rejects the sample-mode `HASH(*)` predicate and the `sample` CTE name (DEC-005 caveat; sample-mode shape is gated by the US-003 byte-exact snapshots). Confirmed: default `uv run pytest` deselects all 8 snowflake tests; `uv run pytest -m snowflake --no-cov` runs them green. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…L validation

The deterministic-sample CTE was bound to the bare alias `sample`, but `SAMPLE` is a Snowflake reserved keyword (TABLESAMPLE) — `WITH sample AS (...)` is a syntax error on real Snowflake (confirmed via sqlglot). This is on the default `scope=sample` prune path, not an edge case. Fix: add `Dialect.sample_cte_alias` (BigQuery default bare `sample` → BQ snapshots byte-identical; Snowflake `"sample"` quoted to bypass the reserved word). Thread it through `_render_sample_cte`, the built-in `target`, the relationships `child_target`, and the custom_sql sample substitution. Regenerated the 5 Snowflake sample fixtures; added a gated sqlglot parse-guard over every Snowflake fixture (the guard that catches this class of bug, since fakesnow can't execute sample-mode SQL due to HASH(*)). Found during US-005 Quality Gate; surfaced by the US-004 fakesnow validation. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…ard hardening From the 4 code-review passes (no blocking bugs found beyond the already-fixed sample-CTE reserved-word bug): - Doc-count drift: the QG sample-CTE fix added a 5th Dialect field, so the "four fields" wording in the Dialect / POSTGRES docstrings and three test_models docstrings was stale -> "five". Added the missing sample_cte_alias default assertion to test_dialect_constructs_without_new_field_args. Clarified the POSTGRES note (sample_cte_alias="sample" is already Postgres-correct; quote_qualified_per_component + cast templates are what need correcting when Postgres ops land). - Import-guard hardening: the AST detector missed the namespace-split `from google import cloud` form (module='google' alone isn't forbidden but the imported name completes 'google.cloud'). Now flagged; planted self-check extended to 9 violations. Resolved the reviewers' split on string-literal escaping: Snowflake DOES honor backslash escapes (\') in single-quoted string literals, so the shared escape_bq_string_literal is valid for both dialects (no change). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…+ identifier_case graduation 5-surface graduation of identifier_case (declared -> active) + Snowflake compiler dialect documentation: - .claude/rules/prune-engine.md: full Dialect-field list the compiler reads; identifier_case graduation; sample_cte_alias reserved-word lesson (snapshots pin shape not validity — keep a parser/executor in the loop); supports_qualify stays forward-compat; new #121 Snowflake subsection + refs. - docs/prune-ops.md: Snowflake compiler dialect section (case-folding + residual, HASH() cross-version reproducibility caveat, QUALIFY not used, validation tiers); updated multi-warehouse deferral (Snowflake landed). - CLAUDE.md: public-API surface — compiler emits Snowflake SQL from SNOWFLAKE_DIALECT; five new Dialect fields; identifier_case graduated. - .claude/rules/warehouse-adapters.md: Dialect now carries prune-compiler SQL-fragment templates; populate them for future vendor dialects. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Copilot

Pull request overview

This PR operationalizes Snowflake-compatible SQL generation for the prune compiler by making warehouse-specific SQL emission fully dialect-driven via Dialect (rather than any warehouse SDK imports or branching on dialect name), while keeping BigQuery output byte-identical through defaults and snapshot regression gates.

Changes:

Extended Dialect with declarative SQL-fragment fields (BigQuery-defaulted) and set Snowflake-specific values on SNOWFLAKE_DIALECT.
Refactored the prune compiler to consume the dialect fields for identifier folding/quoting, qualified-name rendering, sampling hash predicate, and date/timestamp literals.
Added Snowflake snapshot fixtures + gated fakesnow/sqlglot validation, plus an AST-based import confinement guard for src/signalforge/prune/.

Reviewed changes

Copilot reviewed 23 out of 24 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
uv.lock	Adds locked dependencies for Snowflake SQL validation stack (fakesnow/duckdb/pyarrow/sqlglot).
pyproject.toml	Adds `fakesnow` to dev deps; registers/excludes `snowflake` marker by default.
src/signalforge/warehouse/models.py	Extends `Dialect` with SQL fragment fields; configures `SNOWFLAKE_DIALECT` values.
src/signalforge/prune/compiler.py	Makes compilation fully dialect-driven (folding, quoting, sampling predicate, literals, qualified names).
tests/warehouse/test_models.py	Pins new `Dialect` field defaults + Snowflake-specific values.
tests/prune/test_compiler.py	Adds unit coverage for folding/quoting + Snowflake snapshot suite and guards.
tests/prune/test_compiler_import_guard.py	AST scan ensuring no `snowflake`/`google.cloud` imports under `src/signalforge/prune/`.
tests/prune/test_compiler_fakesnow.py	Gated fakesnow execution tests + sqlglot Snowflake parse guard for fixtures.
tests/fixtures/prune/compiled_sql/snowflake/not_null.sql	Snowflake snapshot fixture for `not_null` (full).
tests/fixtures/prune/compiled_sql/snowflake/not_null_sample.sql	Snowflake snapshot fixture for `not_null` (sample).
tests/fixtures/prune/compiled_sql/snowflake/unique.sql	Snowflake snapshot fixture for `unique` (full).
tests/fixtures/prune/compiled_sql/snowflake/unique_sample.sql	Snowflake snapshot fixture for `unique` (sample).
tests/fixtures/prune/compiled_sql/snowflake/accepted_values.sql	Snowflake snapshot fixture for `accepted_values` (full).
tests/fixtures/prune/compiled_sql/snowflake/accepted_values_sample.sql	Snowflake snapshot fixture for `accepted_values` (sample).
tests/fixtures/prune/compiled_sql/snowflake/relationships.sql	Snowflake snapshot fixture for `relationships` (full).
tests/fixtures/prune/compiled_sql/snowflake/relationships_sample.sql	Snowflake snapshot fixture for `relationships` (sample).
tests/fixtures/prune/compiled_sql/snowflake/custom_sql.sql	Snowflake snapshot fixture for custom SQL (single-table, full).
tests/fixtures/prune/compiled_sql/snowflake/custom_sql_sample.sql	Snowflake snapshot fixture for custom SQL (single-table, sample).
tests/fixtures/prune/compiled_sql/snowflake/custom_sql_fullscan.sql	Snowflake snapshot fixture for custom SQL (multi-table full-scan).
plans/super/121-prune-snowflake-dialect.md	Plan document capturing decisions and breakdown for #121.
docs/prune-ops.md	Documents Snowflake compiler behavior and validation strategy.
CLAUDE.md	Updates public API surface notes for new Dialect fields + Snowflake compiler support.
.claude/rules/warehouse-adapters.md	Documents new Dialect SQL-fragment fields as part of adapter conventions.
.claude/rules/prune-engine.md	Updates pruning rules to reflect new dialect-driven compiler surfaces and Snowflake validation.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

- test_compiler_import_guard.py: fix confusing `googleftover` example in the prefix-match comment -> `google_leftover` (a name that starts with `google` but is not `google.cloud`). - plans/super/121-prune-snowflake-dialect.md: align the Phase line with reality (implementation complete, PR open awaiting review) so it no longer reads as a plan-only review. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

wjduenow · 2026-05-26T18:53:14Z

PR Review Summary

Fixed (2 items)

File	Line	Issue	Commit
`tests/prune/test_compiler_import_guard.py`	27	Confusing `googleftover` example in prefix-match comment	`530f6a4`
`plans/super/121-prune-snowflake-dialect.md`	9	Phase line read as plan-only (stale vs. implementation PR)	`530f6a4`

False Positives (0 items)

None — both comments were real and fixed inline.

Codecov flagged 1 uncovered patch line — compiler.py:149, the `identifier_case == "lower"` branch of `_fold_identifier`. No compiler test exercised a lower-folding dialect (POSTGRES_DIALECT isn't used in the snapshot suite). Added test_quote_folds_lower_for_postgres_dialect pinning `_quote("CustomerId", POSTGRES_DIALECT) == '"customerid"'`. Patch coverage now 100% for the changeset; the remaining compiler.py:312 miss is pre-existing #116 custom_sql code, outside this PR's diff. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

wjduenow · 2026-05-26T20:08:03Z

@coderabbitai review

coderabbitai · 2026-05-26T20:08:20Z

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

…flake-dialect # Conflicts: # .claude/rules/warehouse-adapters.md # CLAUDE.md

Add super plan for #121: prune compiler Snowflake dialect support

4420599

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

wjduenow and others added 12 commits May 26, 2026 09:27

#121: devolve plan to beads (epic bd_1-scaffolding-mfa)

f08f172

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Merge bead bd_1-scaffolding-mfa.1: US-001 extend Dialect with dialect…

acb210a

…-fragment fields

Merge bead bd_1-scaffolding-mfa.2: US-002 consume Dialect fields in c…

398765e

…ompiler

Merge bead bd_1-scaffolding-mfa.3: US-003 Snowflake snapshot fixtures…

bef76e2

… + tests

Merge bead bd_1-scaffolding-mfa.4: US-004 fakesnow gated Snowflake SQ…

7d59d09

…L validation

wjduenow requested a review from Copilot May 26, 2026 18:42

wjduenow marked this pull request as ready for review May 26, 2026 18:42

Copilot started reviewing on behalf of wjduenow May 26, 2026 18:42 View session

Copilot AI reviewed May 26, 2026

View reviewed changes

Comment thread tests/prune/test_compiler_import_guard.py Outdated

Comment thread plans/super/121-prune-snowflake-dialect.md Outdated

wjduenow changed the title ~~#121: prune compiler Snowflake dialect support (plan)~~ #121: prune compiler Snowflake dialect support May 26, 2026

Merge remote-tracking branch 'origin/dev' into feature/121-prune-snow…

0f06fa8

…flake-dialect # Conflicts: # .claude/rules/warehouse-adapters.md # CLAUDE.md

wjduenow merged commit 1f8cc51 into dev May 26, 2026
6 checks passed

wjduenow deleted the feature/121-prune-snowflake-dialect branch May 26, 2026 20:34

This was referenced May 27, 2026

#139: Snowflake projection-subquery sample shape #142

Merged

#169: row_count_between 6th test primitive #176

Merged

coderabbitai Bot mentioned this pull request Jun 1, 2026

#171: row_count_anomaly_by_period #181

Merged

coderabbitai Bot mentioned this pull request Jun 17, 2026

fix(#171): row_count_anomaly_by_period emits valid BigQuery SQL #248

Merged

This was referenced Jun 24, 2026

feat(warehouse): #221 Databricks adapter skeleton #254

Merged

#223: Databricks prune compiler emits valid Databricks SQL from DATABRICKS_DIALECT #256

Merged

coderabbitai Bot mentioned this pull request Jul 6, 2026

#267: aggregate/scalar dbt-expectations macros — source-table routing #269

Merged

Navigation Menu

#121: prune compiler Snowflake dialect support - #127

#121: prune compiler Snowflake dialect support#127
wjduenow merged 16 commits into
devfrom
feature/121-prune-snowflake-dialect

wjduenow commented May 26, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented May 26, 2026 •

edited

Loading

Review skipped

Walkthrough

Changes

Estimated code review effort

Possibly related issues

Possibly related PRs

Poem

Uh oh!

codecov-commenter commented May 26, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Uh oh!

wjduenow commented May 26, 2026

Uh oh!

wjduenow commented May 26, 2026

Uh oh!

coderabbitai Bot commented May 26, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

wjduenow commented May 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

Testing

Compounding Update

Uh oh!

coderabbitai Bot commented May 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Walkthrough

Changes

Estimated code review effort

Possibly related issues

Possibly related PRs

Poem

Uh oh!

codecov-commenter commented May 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

wjduenow commented May 26, 2026

PR Review Summary

Fixed (2 items)

False Positives (0 items)

Uh oh!

wjduenow commented May 26, 2026

Uh oh!

coderabbitai Bot commented May 26, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

wjduenow commented May 26, 2026 •

edited

Loading

coderabbitai Bot commented May 26, 2026 •

edited

Loading

codecov-commenter commented May 26, 2026 •

edited

Loading