Add CI workflow (#18)

* Add CI workflow and apply ruff formatting

* Disable cdk synth — email_processor uses pre-built package dir

The email_processor Lambda bundles deps into a gitignored package/
directory. cdk synth fails in CI without a build step to recreate it.
Disabling until packaging is standardized.

* Use CDK BundlingOptions for email_processor Lambda packaging

Replaces the pre-built gitignored package/ directory with CDK's
built-in bundling. Deps are now installed inside a Docker container
during cdk synth, so the build works identically locally and in CI.
Re-enables run-cdk-synth in the CI workflow.

.github/workflows/ci.yaml

@@ -0,0 +1,12 @@
+name: CI
+on:
+  pull_request:
+    branches: [main]
+
+jobs:
+  ci:
+    uses: Sea-Haven-Industries/.github/.github/workflows/ci-python-sam.yaml@main
+    with:
+      source-dirs: "lambdas cdk"
+      run-cdk-synth: true
+      run-sam-validate: false

cdk/app.py

@@ -3,7 +3,9 @@ import aws_cdk as cdk
 from stack import PoIngestStack
 
 app = cdk.App()
-PoIngestStack(app, "po-ingest",
+PoIngestStack(
+    app,
+    "po-ingest",
     stack_name="po-ingest",
     env=cdk.Environment(region="us-east-1"),
 )

cdk/stack.py

@@ -14,7 +14,6 @@ from aws_cdk import (
     aws_ses as ses,
     aws_ses_actions as ses_actions,
     aws_secretsmanager as secretsmanager,
-    aws_iam as iam,
 )
 from constructs import Construct
 
@@ -25,7 +24,8 @@ class PoIngestStack(Stack):
 
         # --- S3 bucket for raw emails ---
         email_bucket = s3.Bucket(
-            self, "EmailBucket",
+            self,
+            "EmailBucket",
             bucket_name=f"po-ingest-emails-{self.account}",
             removal_policy=RemovalPolicy.RETAIN,
             lifecycle_rules=[
@@ -37,7 +37,8 @@ class PoIngestStack(Stack):
         # Owned by this stack. Streams enabled for the site-extractor pipeline.
         # Other stacks (seahaven-slack-bot) reference this table via fromTableName().
         po_table = dynamodb.Table(
-            self, "PurchaseOrdersTable",
+            self,
+            "PurchaseOrdersTable",
             table_name="purchase-orders",
             partition_key=dynamodb.Attribute(
                 name="po_number",
@@ -50,7 +51,8 @@ class PoIngestStack(Stack):
 
         # --- Secrets Manager for Anthropic API key ---
         anthropic_secret = secretsmanager.Secret(
-            self, "AnthropicApiKey",
+            self,
+            "AnthropicApiKey",
             secret_name="po-ingest/anthropic-api-key",
             description="Anthropic API key for Coupa PO email parsing",
             removal_policy=RemovalPolicy.RETAIN,
@@ -58,12 +60,23 @@ class PoIngestStack(Stack):
 
         # --- Lambda function ---
         email_processor = lambda_.Function(
-            self, "EmailProcessor",
+            self,
+            "EmailProcessor",
             function_name="po-email-processor",
             runtime=lambda_.Runtime.PYTHON_3_12,
             architecture=lambda_.Architecture.ARM_64,
             handler="handler.handler",
-            code=lambda_.Code.from_asset("../lambdas/email_processor/package"),
+            code=lambda_.Code.from_asset(
+                "../lambdas/email_processor",
+                bundling=cdk.BundlingOptions(
+                    image=lambda_.Runtime.PYTHON_3_12.bundling_image,
+                    command=[
+                        "bash",
+                        "-c",
+                        "pip install -r requirements.txt -t /asset-output && cp handler.py /asset-output/",
+                    ],
+                ),
+            ),
             timeout=Duration.seconds(60),
             memory_size=256,
             log_retention=logs.RetentionDays.TWO_MONTHS,
@@ -88,7 +101,9 @@ class PoIngestStack(Stack):
         # --- SES Receipt Rule ---
         # Reuse the existing INBOUND_MAIL rule set (shared with workorder-ingest)
         rule_set = ses.ReceiptRuleSet.from_receipt_rule_set_name(
-            self, "ExistingRuleSet", "INBOUND_MAIL",
+            self,
+            "ExistingRuleSet",
+            "INBOUND_MAIL",
         )
 
         rule_set.add_rule(
@@ -104,7 +119,8 @@ class PoIngestStack(Stack):
 
         # --- Web UI Lambda ---
         web_ui = lambda_.Function(
-            self, "WebUI",
+            self,
+            "WebUI",
             function_name="po-web-ui",
             runtime=lambda_.Runtime.PYTHON_3_12,
             architecture=lambda_.Architecture.ARM_64,
@@ -125,11 +141,14 @@ class PoIngestStack(Stack):
             auth_type=lambda_.FunctionUrlAuthType.NONE,
         )
 
-        cdk.CfnOutput(self, "WebUIUrl", value=web_url.url, description="PO Dashboard URL")
+        cdk.CfnOutput(
+            self, "WebUIUrl", value=web_url.url, description="PO Dashboard URL"
+        )
 
         # --- Verified sites table (extracted from PO ship-to addresses) ---
         verified_sites_table = dynamodb.Table(
-            self, "VerifiedSitesTable",
+            self,
+            "VerifiedSitesTable",
             table_name="verified-sites",
             partition_key=dynamodb.Attribute(
                 name="siteCode",
@@ -149,7 +168,8 @@ class PoIngestStack(Stack):
 
         # --- Site extractor Lambda (DynamoDB Streams → verified-sites) ---
         site_extractor = lambda_.Function(
-            self, "SiteExtractor",
+            self,
+            "SiteExtractor",
             function_name="po-ingest-site-extractor",
             runtime=lambda_.Runtime.PYTHON_3_12,
             architecture=lambda_.Architecture.ARM_64,
@@ -177,14 +197,17 @@ class PoIngestStack(Stack):
             )
         )
 
-        cdk.CfnOutput(self, "VerifiedSitesTableName",
+        cdk.CfnOutput(
+            self,
+            "VerifiedSitesTableName",
             value=verified_sites_table.table_name,
             description="Verified site addresses extracted from POs",
         )
 
         # --- Pending site review table (POs with no extractable site code) ---
         pending_review_table = dynamodb.Table(
-            self, "PendingSiteReviewTable",
+            self,
+            "PendingSiteReviewTable",
             table_name="pending-site-review",
             partition_key=dynamodb.Attribute(
                 name="po_number",

lambdas/email_processor/handler.py

@@ -290,7 +290,7 @@ def pad_zip(zip_code: str | None) -> str | None:
         return zip_code
     clean = zip_code.strip().split("-")[0]
     if clean.isdigit() and len(clean) < 5:
-        return clean.zfill(5) + zip_code.strip()[len(clean):]
+        return clean.zfill(5) + zip_code.strip()[len(clean) :]
     return zip_code
 
 
@@ -369,7 +369,9 @@ def handler(event, context):
         logger.info(f"Subject: {email_data['subject']}")
 
         parsed = extract_with_claude(email_data)
-        logger.info(f"Parsed: type={parsed.get('email_type')}, po={parsed.get('po_number')}")
+        logger.info(
+            f"Parsed: type={parsed.get('email_type')}, po={parsed.get('po_number')}"
+        )
 
         if not parsed.get("po_number"):
             logger.warning(f"No PO number found in email, skipping: {key}")

lambdas/site_extractor/handler.py

@@ -49,9 +49,7 @@ def extract_site_code(record: dict) -> str | None:
         if m and SITE_CODE_PATTERN.match(m.group(1)):
             return m.group(1)
 
-        m = re.search(
+        m = re.search(r"(?:Station|DS)\s*[-–]?\s*([A-Z0-9]{3,5})\b", ship_to_name)
-            r"(?:Station|DS)\s*[-–]?\s*([A-Z0-9]{3,5})\b", ship_to_name
-        )
         if m and SITE_CODE_PATTERN.match(m.group(1)):
             return m.group(1)
 
@@ -85,7 +83,11 @@ def parse_address(record: dict) -> dict:
         result["city"] = ship_to.get("city")
         result["state"] = ship_to["state"]
         result["zip"] = ship_to.get("zip")
-        parts = [p for p in [result["address"], result["city"], result["state"], result["zip"]] if p]
+        parts = [
+            p
+            for p in [result["address"], result["city"], result["state"], result["zip"]]
+            if p
+        ]
         result["fullAddress"] = ", ".join(parts)
         return result
 
@@ -104,11 +106,13 @@ def parse_address(record: dict) -> dict:
         result["city"] = m.group(1).strip()
         result["state"] = m.group(2)
         result["zip"] = m.group(3)
-        result["fullAddress"] = f"{result['address']}, {result['city']}, {result['state']} {result['zip']}"
+        result["fullAddress"] = (
+            f"{result['address']}, {result['city']}, {result['state']} {result['zip']}"
+        )
         return result
 
     # Fallback: try newline-separated format
-    lines = [l.strip() for l in clean.split("\n") if l.strip()]
+    lines = [line.strip() for line in clean.split("\n") if line.strip()]
     city_re = re.compile(r"^(.+?),\s+([A-Z]{2})\s+(\d{5}(?:-\d{4})?)")
     for i, line in enumerate(lines):
         m = city_re.match(line)
@@ -118,7 +122,9 @@ def parse_address(record: dict) -> dict:
             result["state"] = m.group(2)
             result["zip"] = m.group(3)
             street = result["address"] or result["city"]
-            result["fullAddress"] = f"{street}, {result['city']}, {result['state']} {result['zip']}"
+            result["fullAddress"] = (
+                f"{street}, {result['city']}, {result['state']} {result['zip']}"
+            )
             return result
 
     result["address"] = clean
@@ -278,7 +284,10 @@ def handler(event, context):
             ship_to_name = (po.get("ship_to") or {}).get("name")
             write_pending_review(po_number, address, ship_to_name)
             pending += 1
-            logger.info("PO %s added to pending review (no site code or address match)", po_number)
+            logger.info(
+                "PO %s added to pending review (no site code or address match)",
+                po_number,
+            )
 
     logger.info(
         "Batch complete: %d processed, %d extracted (%d via address), %d pending review",

lambdas/web_ui/handler.py

@@ -5,7 +5,6 @@ Serves a simple HTML dashboard for viewing purchase orders.
 Accessed via Lambda Function URL.
 """
 
-import json
 import os
 from decimal import Decimal
 
@@ -103,7 +102,12 @@ def render_po_detail(po):
 
     view_url = po.get("view_order_url")
     if view_url:
-        fields.append(("Coupa Link", f'<a href="{view_url}" target="_blank" style="color:#3b82f6;">View in Coupa</a>'))
+        fields.append(
+            (
+                "Coupa Link",
+                f'<a href="{view_url}" target="_blank" style="color:#3b82f6;">View in Coupa</a>',
+            )
+        )
 
     details_html = ""
     for label, value in fields:
@@ -120,17 +124,17 @@ def render_po_detail(po):
     if line_items:
         rows = ""
         for item in line_items:
-            qty = item.get('quantity', '') or ''
+            qty = item.get("quantity", "") or ""
-            unit = item.get('unit', '') or ''
+            unit = item.get("unit", "") or ""
-            price = item.get('price', '') or ''
+            price = item.get("price", "") or ""
             rows += f"""
             <tr style="border-bottom:1px solid #f1f5f9;">
-                <td style="padding:10px;font-size:14px;">{item.get('description', '')}</td>
+                <td style="padding:10px;font-size:14px;">{item.get("description", "")}</td>
                 <td style="padding:10px;font-size:13px;text-align:right;">{qty}</td>
                 <td style="padding:10px;font-size:13px;">{unit}</td>
                 <td style="padding:10px;font-size:13px;text-align:right;">{price}</td>
-                <td style="padding:10px;font-size:14px;text-align:right;">{fmt_currency(item.get('amount'))}</td>
+                <td style="padding:10px;font-size:14px;text-align:right;">{fmt_currency(item.get("amount"))}</td>
-                <td style="padding:10px;font-size:13px;color:#64748b;">{item.get('need_by', '') or ''}</td>
+                <td style="padding:10px;font-size:13px;color:#64748b;">{item.get("need_by", "") or ""}</td>
             </tr>"""
 
         items_html = f"""