Version 0.9.2

docs/database.md

@@ -1,6 +1,15 @@
 Starlette includes optional database support. There is currently only a driver
 for Postgres databases, but MySQL and SQLite support is planned.
 
+Enabling the built-in database support requires `sqlalchemy`, and an appropriate database driver. Currently this means `asyncpg` is a requirement.
+
+The database support is completely optional - you can either include the middleware or not, or you can build alternative kinds of backends instead. It does not
+include support for an ORM, but it does support using queries built using
+[SQLAlchemy Core][sqlalchemy-core].
+
+Here's a complete example, that includes table definitions, installing the
+`DatabaseMiddleware`, and a couple of endpoints that interact with the database.
+
 ```python
 import os
 import sqlalchemy

docs/graphql.md

@@ -25,6 +25,22 @@ which you can use to interact with your GraphQL API.
 
 ![GraphiQL](img/graphiql.png)
 
+## Accessing request information
+
+The current request is available in the context.
+
+```python
+class Query(graphene.ObjectType):
+    user_agent = graphene.String()
+
+    def resolve_user_agent(self, info):
+        """
+        Return the User-Agent of the incoming request.
+        """
+        request = info.context["request"]
+        return request.headers.get("User-Agent", "<unknown>")
+```
+
 ## Sync or Async executors
 
 If you're working with a standard ORM, then just use regular function calls for

docs/release-notes.md

@@ -1,3 +1,9 @@
+## 0.9.2
+
+* Add optional database support.
+* Add `request` to GraphQL context.
+* Hide any password component in `URL.__repr__`.
+
 ## 0.9.1
 
 * Handle startup/shutdown errors properly.

starlette/__init__.py

@@ -1 +1 @@
-__version__ = "0.9.1"
+__version__ = "0.9.2"