From 5928e50ca091c2970a30927593cc6dea904a15d7 Mon Sep 17 00:00:00 2001
From: Jesse Mortenson <jessemortenson@gmail.com>
Date: Fri, 24 Jan 2025 14:29:00 -0600
Subject: [PATCH 1/2] VA: bugfix for missing action description

---
 scrapers/utils/actions.py | 13 ++++++++-----
 scrapers/va/bills.py      | 25 +++++++++++++++++--------
 2 files changed, 25 insertions(+), 13 deletions(-)

diff --git a/scrapers/utils/actions.py b/scrapers/utils/actions.py
index ea9c5251e9..0163ddc3ca 100644
--- a/scrapers/utils/actions.py
+++ b/scrapers/utils/actions.py
@@ -46,11 +46,14 @@ def match(self, text):
         matched = False
 
         for regex in self.regexes:
-            m = regex.search(text)
-            if m:
-                matched = True
-                # add any matched attrs
-                attrs.update(m.groupdict())
+            try:
+                m = regex.search(text)
+                if m:
+                    matched = True
+                    # add any matched attrs
+                    attrs.update(m.groupdict())
+            except TypeError:
+                continue
 
         if matched:
             return attrs
diff --git a/scrapers/va/bills.py b/scrapers/va/bills.py
index f34c9f8087..1e602607d2 100644
--- a/scrapers/va/bills.py
+++ b/scrapers/va/bills.py
@@ -121,15 +121,24 @@ def add_actions(self, bill: Bill, legislation_id: str):
 
         for row in page["LegislationEvents"]:
             when = dateutil.parser.parse(row["EventDate"]).date()
-            action_attr = self.categorizer.categorize(row["Description"])
-            classification = action_attr["classification"]
+            description = row["Description"]
+            if not description and row["VoteTally"]:
+                description = f"Vote {row['VoteTally']}"
 
-            bill.add_action(
-                chamber=self.chamber_map[row["ChamberCode"]],
-                description=row["Description"],
-                date=when,
-                classification=classification,
-            )
+            if description:
+                action_attr = self.categorizer.categorize(description)
+                classification = action_attr["classification"]
+
+                bill.add_action(
+                    chamber=self.chamber_map[row["ChamberCode"]],
+                    description=description,
+                    date=when,
+                    classification=classification,
+                )
+            else:
+                self.logger.warning(
+                    f"Could not add action due to missing description for {bill.identifier} LegislataionEventID {row['LegislationEventID']}"
+                )
 
             # map reference numbers back to their actions for impact filenames
             # HB9F122.PDF > { 'HB9F122' => "Impact statement from DPB (HB9)" }

From 62a7dc493c714e24ac084f9c2975e67abc9d23f0 Mon Sep 17 00:00:00 2001
From: Jesse Mortenson <jessemortenson@gmail.com>
Date: Fri, 24 Jan 2025 14:31:39 -0600
Subject: [PATCH 2/2] Backoff on swallowing errors in action classifier

---
 scrapers/utils/actions.py | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/scrapers/utils/actions.py b/scrapers/utils/actions.py
index 0163ddc3ca..ea9c5251e9 100644
--- a/scrapers/utils/actions.py
+++ b/scrapers/utils/actions.py
@@ -46,14 +46,11 @@ def match(self, text):
         matched = False
 
         for regex in self.regexes:
-            try:
-                m = regex.search(text)
-                if m:
-                    matched = True
-                    # add any matched attrs
-                    attrs.update(m.groupdict())
-            except TypeError:
-                continue
+            m = regex.search(text)
+            if m:
+                matched = True
+                # add any matched attrs
+                attrs.update(m.groupdict())
 
         if matched:
             return attrs