From 3ccc5184e0fe1f00dec69293c097d513da54a410 Mon Sep 17 00:00:00 2001
From: Erik Johnston <erikj@element.io>
Date: Fri, 31 Oct 2025 13:16:47 +0000
Subject: [PATCH] Fix schema lint script to understand `CREATE TABLE IF NOT
 EXISTS` (#19020)

The schema lint tries to make sure we don't add or remove indices in
schema files (rather than as background updates), *unless* the table was
created in the same schema file.

The regex to pull out the `CREATE TABLE` SQL incorrectly didn't
recognise `IF NOT EXISTS`.

There is a test delta file that shows that we accept different types of
`CREATE TABLE` and `CREATE INDEX` statements, as well as an index
creation that doesn't have a matching create table (to show that we do
still catch it). The test delta should be removed before merge.
---
 changelog.d/19020.misc            |  1 +
 scripts-dev/check_schema_delta.py | 20 +++++++++++++++-----
 2 files changed, 16 insertions(+), 5 deletions(-)
 create mode 100644 changelog.d/19020.misc

diff --git a/changelog.d/19020.misc b/changelog.d/19020.misc
new file mode 100644
index 0000000000..f5775ff194
--- /dev/null
+++ b/changelog.d/19020.misc
@@ -0,0 +1 @@
+Fix CI linter for schema delta files to correctly handle all types of `CREATE TABLE` syntax.
diff --git a/scripts-dev/check_schema_delta.py b/scripts-dev/check_schema_delta.py
index 7b2dec25d4..dd96c904bb 100755
--- a/scripts-dev/check_schema_delta.py
+++ b/scripts-dev/check_schema_delta.py
@@ -11,9 +11,13 @@ import click
 import git
 
 SCHEMA_FILE_REGEX = re.compile(r"^synapse/storage/schema/(.*)/delta/(.*)/(.*)$")
-INDEX_CREATION_REGEX = re.compile(r"CREATE .*INDEX .*ON ([a-z_]+)", flags=re.IGNORECASE)
-INDEX_DELETION_REGEX = re.compile(r"DROP .*INDEX ([a-z_]+)", flags=re.IGNORECASE)
-TABLE_CREATION_REGEX = re.compile(r"CREATE .*TABLE ([a-z_]+)", flags=re.IGNORECASE)
+INDEX_CREATION_REGEX = re.compile(
+    r"CREATE .*INDEX .*ON ([a-z_0-9]+)", flags=re.IGNORECASE
+)
+INDEX_DELETION_REGEX = re.compile(r"DROP .*INDEX ([a-z_0-9]+)", flags=re.IGNORECASE)
+TABLE_CREATION_REGEX = re.compile(
+    r"CREATE .*TABLE.* ([a-z_0-9]+)\s*\(", flags=re.IGNORECASE
+)
 
 # The base branch we want to check against. We use the main development branch
 # on the assumption that is what we are developing against.
@@ -173,11 +177,14 @@ def main(force_colors: bool) -> None:
                 clause = match.group()
 
                 click.secho(
-                    f"Found delta with index deletion: '{clause}' in {delta_file}\nThese should be in background updates.",
+                    f"Found delta with index deletion: '{clause}' in {delta_file}",
                     fg="red",
                     bold=True,
                     color=force_colors,
                 )
+                click.secho(
+                    " ↪ These should be in background updates.",
+                )
                 return_code = 1
 
             # Check for index creation, which is only allowed for tables we've
@@ -188,11 +195,14 @@ def main(force_colors: bool) -> None:
                 table_name = match.group(1)
                 if table_name not in created_tables:
                     click.secho(
-                        f"Found delta with index creation: '{clause}' in {delta_file}\nThese should be in background updates.",
+                        f"Found delta with index creation for existing table: '{clause}' in {delta_file}",
                         fg="red",
                         bold=True,
                         color=force_colors,
                     )
+                    click.secho(
+                        " ↪ These should be in background updates (or the table should be created in the same delta).",
+                    )
                     return_code = 1
 
     click.get_current_context().exit(return_code)