From 6261b0f4e63102d50fd217652149b078a48dcdf9 Mon Sep 17 00:00:00 2001
From: Irving Popovetsky <irving@honeycomb.io>
Date: Tue, 2 Jan 2024 09:19:23 -0800
Subject: [PATCH] Add Honeycomb.io sqlfluff workflow

Signed-off-by: Irving Popovetsky <irving@honeycomb.io>
---
 menu_of_workflows/honeycomb.io/.sqlfluff      | 32 +++++++++++++
 menu_of_workflows/honeycomb.io/README.md      | 19 ++++++++
 .../honeycomb.io/sqlfluff_lint_dbt_models.yml | 48 +++++++++++++++++++
 3 files changed, 99 insertions(+)
 create mode 100644 menu_of_workflows/honeycomb.io/.sqlfluff
 create mode 100644 menu_of_workflows/honeycomb.io/README.md
 create mode 100644 menu_of_workflows/honeycomb.io/sqlfluff_lint_dbt_models.yml

diff --git a/menu_of_workflows/honeycomb.io/.sqlfluff b/menu_of_workflows/honeycomb.io/.sqlfluff
new file mode 100644
index 0000000..41493e0
--- /dev/null
+++ b/menu_of_workflows/honeycomb.io/.sqlfluff
@@ -0,0 +1,32 @@
+[sqlfluff]
+templater = jinja
+dialect = snowflake
+max_line_length = 160
+
+[sqlfluff:indentation]
+indent_unit = space
+tab_space_size = 4
+
+[sqlfluff:rules:layout.long_lines]
+ignore_comment_lines = True
+ignore_comment_clauses = True
+
+[sqlfluff:rules:capitalisation.keywords]
+# for Keywords
+capitalisation_policy = lower
+
+[sqlfluff:rules:ambiguous.column_references]
+# GROUP BY/ORDER BY column references
+group_by_and_order_by_style = explicit
+
+[sqlfluff:rules:convention.blocked_words]
+# Comma separated list of blocked words that should not be used
+blocked_words = "iff"
+blocked_regex = None
+
+[sqlfluff:templater:jinja]
+apply_dbt_builtins = True
+
+[sqlfluff:templater:jinja:context]
+database=analytics
+schema=prod
diff --git a/menu_of_workflows/honeycomb.io/README.md b/menu_of_workflows/honeycomb.io/README.md
new file mode 100644
index 0000000..bf6a8a9
--- /dev/null
+++ b/menu_of_workflows/honeycomb.io/README.md
@@ -0,0 +1,19 @@
+# honeycomb.io SQLFluff GitHub Workflow
+
+This workflow is a massively simplified and updated version of the [surfline workflow](menu_of_workflows/surfline).
+
+To add this to your repo, copy the contents of [`sqlfluff_lint_dbt_models.yml`](./sqlfluff_lint_dbt_models.yml)
+in this folder into a file named `.github/workflows/sqlfluff_lint_dbt_models.yml`.
+
+## This GitHub Workflow
+- Lints any added or modified models in `/models`, with one exclusion example
+- Uses a bare `requirements.txt` file to manage python dependencies. The contents of it are simply `sqlfluff==2.3.5`
+- Identifies only the changed files on Pull Requests so that only they are linted.
+- Annotates failures on the PR, on the line where they occur, using SQLFluff's native annotation mode.
+
+__NOTE:__ This workflow has been tested on Snowflake, but it should be possible to modify the flow to work with other data warehouses.
+
+## Setup
+### `.sqlfluff`
+
+The `.sqlfluff` found [here](./.sqlfluff) skips the dbt templater in favor of the much simpler and faster jinja templater. A few jinja contexts are supplied as an example, and these are typically enough to keep the templater happy.
diff --git a/menu_of_workflows/honeycomb.io/sqlfluff_lint_dbt_models.yml b/menu_of_workflows/honeycomb.io/sqlfluff_lint_dbt_models.yml
new file mode 100644
index 0000000..c7ac708
--- /dev/null
+++ b/menu_of_workflows/honeycomb.io/sqlfluff_lint_dbt_models.yml
@@ -0,0 +1,48 @@
+# Original source: https://github.com/sqlfluff/sqlfluff-github-actions/blob/main/menu_of_workflows/surfline/sqlfluff_lint_dbt_models.yml
+name: SQLFluff lint dbt data models
+
+on:
+  push:
+    branches:
+    - main
+  pull_request:
+    branches:
+    - main
+
+jobs:
+  sqlfluff-lint-models:
+    name: Lint dbt models using SQLFluff
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: ['ubuntu-latest']
+        python-version: ['3.11']
+
+    steps:
+      - name: Checkout branch
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0  # OR "2" -> To retrieve the preceding commit.
+
+      - name: Setup Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: 'pip'
+
+      - name: Install Python packages or restore cache
+        run: pip install -r requirements.txt
+
+      - name: Get all changed model SQL files
+        id: changed-files-models
+        uses: tj-actions/changed-files@v37
+        with:
+          files: models/**/*.sql
+          files_ignore: models/utils/date_spine.sql
+
+      - name: Lint dbt models
+        if: steps.changed-files-models.outputs.any_changed == 'true'
+        shell: bash -l {0}
+        run: |
+          sqlfluff lint --format github-annotation-native --annotation-level failure ${{ steps.changed-files-models.outputs.all_changed_files }}