{"id": "sql_queries:canned-queries-magic-parameters", "page": "sql_queries", "ref": "canned-queries-magic-parameters", "title": "Magic parameters", "content": "Named parameters that start with an underscore are special: they can be used to automatically add values created by Datasette that are not contained in the incoming form fields or query string. \n                 These magic parameters are only supported for canned queries: to avoid security issues (such as queries that extract the user's private cookies) they are not available to SQL that is executed by the user as a custom SQL query. \n                 Available magic parameters are: \n                 \n                     \n                         _actor_*  - e.g.  _actor_id ,  _actor_name \n                         \n                             Fields from the currently authenticated  Actors . \n                         \n                     \n                     \n                         _header_*  - e.g.  _header_user_agent \n                         \n                             Header from the incoming HTTP request. The key should be in lower case and with hyphens converted to underscores e.g.  _header_user_agent  or  _header_accept_language . \n                         \n                     \n                     \n                         _cookie_*  - e.g.  _cookie_lang \n                         \n                             The value of the incoming cookie of that name. \n                         \n                     \n                     \n                         _now_epoch \n                         \n                             The number of seconds since the Unix epoch. \n                         \n                     \n                     \n                         _now_date_utc \n                         \n                             The date in UTC, e.g.  2020-06-01 \n                         \n                     \n                     \n                         _now_datetime_utc \n                         \n                             The ISO 8601 datetime in UTC, e.g.  2020-06-24T18:01:07Z \n                         \n                     \n                     \n                         _random_chars_*  - e.g.  _random_chars_128 \n                         \n                             A random string of characters of the specified length. \n                         \n                     \n                 \n                 Here's an example configuration (this time using  metadata.yaml  since that provides better support for multi-line SQL queries) that adds a message from the authenticated user, storing various pieces of additional metadata using magic parameters: \n                 databases:\n  mydatabase:\n    queries:\n      add_message:\n        allow:\n          id: \"*\"\n        sql: |-\n          INSERT INTO messages (\n            user_id, message, datetime\n          ) VALUES (\n            :_actor_id, :message, :_now_datetime_utc\n          )\n        write: true \n                 The form presented at  /mydatabase/add_message  will have just a field for  message  - the other parameters will be populated by the magic parameter mechanism. \n                 Additional custom magic parameters can be added by plugins using the  register_magic_parameters(datasette)  hook.", "breadcrumbs": "[\"Running SQL queries\", \"Canned queries\"]", "references": "[]"}
{"id": "sql_queries:canned-queries-named-parameters", "page": "sql_queries", "ref": "canned-queries-named-parameters", "title": "Canned query parameters", "content": "Canned queries support named parameters, so if you include those in the SQL you will then be able to enter them using the form fields on the canned query page or by adding them to the URL. This means canned queries can be used to create custom JSON APIs based on a carefully designed SQL statement. \n                 Here's an example of a canned query with a named parameter: \n                 select neighborhood, facet_cities.name, state\nfrom facetable\n  join facet_cities on facetable.city_id = facet_cities.id\nwhere neighborhood like '%' || :text || '%'\norder by neighborhood; \n                 In the canned query metadata (here  Using YAML for metadata  as  metadata.yaml ) it looks like this: \n                 databases:\n  fixtures:\n    queries:\n      neighborhood_search:\n        sql: |-\n          select neighborhood, facet_cities.name, state\n          from facetable\n            join facet_cities on facetable.city_id = facet_cities.id\n          where neighborhood like '%' || :text || '%'\n          order by neighborhood\n        title: Search neighborhoods \n                 Here's the equivalent using JSON (as  metadata.json ): \n                 {\n    \"databases\": {\n        \"fixtures\": {\n            \"queries\": {\n                \"neighborhood_search\": {\n                    \"sql\": \"select neighborhood, facet_cities.name, state\\nfrom facetable\\n  join facet_cities on facetable.city_id = facet_cities.id\\nwhere neighborhood like '%' || :text || '%'\\norder by neighborhood\",\n                    \"title\": \"Search neighborhoods\"\n                }\n            }\n        }\n    }\n} \n                 Note that we are using SQLite string concatenation here - the  ||  operator - to add wildcard  %  characters to the string provided by the user. \n                 You can try this canned query out here:\n                     https://latest.datasette.io/fixtures/neighborhood_search?text=town \n                 In this example the  :text  named parameter is automatically extracted from the query using a regular expression. \n                 You can alternatively provide an explicit list of named parameters using the  \"params\"  key, like this: \n                 databases:\n  fixtures:\n    queries:\n      neighborhood_search:\n        params:\n        - text\n        sql: |-\n          select neighborhood, facet_cities.name, state\n          from facetable\n            join facet_cities on facetable.city_id = facet_cities.id\n          where neighborhood like '%' || :text || '%'\n          order by neighborhood\n        title: Search neighborhoods", "breadcrumbs": "[\"Running SQL queries\", \"Canned queries\"]", "references": "[{\"href\": \"https://latest.datasette.io/fixtures/neighborhood_search?text=town\", \"label\": \"https://latest.datasette.io/fixtures/neighborhood_search?text=town\"}]"}
{"id": "sql_queries:canned-queries-options", "page": "sql_queries", "ref": "canned-queries-options", "title": "Additional canned query options", "content": "Additional options can be specified for canned queries in the YAML or JSON configuration.", "breadcrumbs": "[\"Running SQL queries\", \"Canned queries\"]", "references": "[]"}
{"id": "sql_queries:canned-queries-writable", "page": "sql_queries", "ref": "canned-queries-writable", "title": "Writable canned queries", "content": "Canned queries by default are read-only. You can use the  \"write\": true  key to indicate that a canned query can write to the database. \n                 See  Controlling access to specific canned queries  for details on how to add permission checks to canned queries, using the  \"allow\"  key. \n                 {\n    \"databases\": {\n        \"mydatabase\": {\n            \"queries\": {\n                \"add_name\": {\n                    \"sql\": \"INSERT INTO names (name) VALUES (:name)\",\n                    \"write\": true\n                }\n            }\n        }\n    }\n} \n                 This configuration will create a page at  /mydatabase/add_name  displaying a form with a  name  field. Submitting that form will execute the configured  INSERT  query. \n                 You can customize how Datasette represents success and errors using the following optional properties: \n                 \n                     \n                         on_success_message  - the message shown when a query is successful \n                     \n                     \n                         on_success_redirect  - the path or URL the user is redirected to on success \n                     \n                     \n                         on_error_message  - the message shown when a query throws an error \n                     \n                     \n                         on_error_redirect  - the path or URL the user is redirected to on error \n                     \n                 \n                 For example: \n                 {\n    \"databases\": {\n        \"mydatabase\": {\n            \"queries\": {\n                \"add_name\": {\n                    \"sql\": \"INSERT INTO names (name) VALUES (:name)\",\n                    \"write\": true,\n                    \"on_success_message\": \"Name inserted\",\n                    \"on_success_redirect\": \"/mydatabase/names\",\n                    \"on_error_message\": \"Name insert failed\",\n                    \"on_error_redirect\": \"/mydatabase\"\n                }\n            }\n        }\n    }\n} \n                 You can use  \"params\"  to explicitly list the named parameters that should be displayed as form fields - otherwise they will be automatically detected. \n                 You can pre-populate form fields when the page first loads using a query string, e.g.  /mydatabase/add_name?name=Prepopulated . The user will have to submit the form to execute the query.", "breadcrumbs": "[\"Running SQL queries\", \"Canned queries\"]", "references": "[]"}
{"id": "sql_queries:fragment", "page": "sql_queries", "ref": "fragment", "title": "fragment", "content": "Some plugins, such as  datasette-vega , can be configured by including additional data in the fragment hash of the URL - the bit that comes after a  #  symbol. \n                     You can set a default fragment hash that will be included in the link to the canned query from the database index page using the  \"fragment\"  key. \n                     This example demonstrates both  fragment  and  hide_sql : \n                     {\n    \"databases\": {\n        \"fixtures\": {\n            \"queries\": {\n                \"neighborhood_search\": {\n                    \"sql\": \"select neighborhood, facet_cities.name, state\\nfrom facetable join facet_cities on facetable.city_id = facet_cities.id\\nwhere neighborhood like '%' || :text || '%' order by neighborhood;\",\n                    \"fragment\": \"fragment-goes-here\",\n                    \"hide_sql\": true\n                }\n            }\n        }\n    }\n} \n                     See here  for a demo of this in action.", "breadcrumbs": "[\"Running SQL queries\", \"Canned queries\", \"Additional canned query options\"]", "references": "[{\"href\": \"https://github.com/simonw/datasette-vega\", \"label\": \"datasette-vega\"}, {\"href\": \"https://latest.datasette.io/fixtures#queries\", \"label\": \"See here\"}]"}
{"id": "sql_queries:hide-sql", "page": "sql_queries", "ref": "hide-sql", "title": "hide_sql", "content": "Canned queries default to displaying their SQL query at the top of the page. If the query is extremely long you may want to hide it by default, with a \"show\" link that can be used to make it visible. \n                     Add the  \"hide_sql\": true  option to hide the SQL query by default.", "breadcrumbs": "[\"Running SQL queries\", \"Canned queries\", \"Additional canned query options\"]", "references": "[]"}
{"id": "sql_queries:id1", "page": "sql_queries", "ref": "id1", "title": "Canned queries", "content": "As an alternative to adding views to your database, you can define canned queries inside your  metadata.json  file. Here's an example: \n             {\n    \"databases\": {\n       \"sf-trees\": {\n           \"queries\": {\n               \"just_species\": {\n                   \"sql\": \"select qSpecies from Street_Tree_List\"\n               }\n           }\n       }\n    }\n} \n             Then run Datasette like this: \n             datasette sf-trees.db -m metadata.json \n             Each canned query will be listed on the database index page, and will also get its own URL at: \n             /database-name/canned-query-name \n             For the above example, that URL would be: \n             /sf-trees/just_species \n             You can optionally include  \"title\"  and  \"description\"  keys to show a title and description on the canned query page. As with regular table metadata you can alternatively specify  \"description_html\"  to have your description rendered as HTML (rather than having HTML special characters escaped).", "breadcrumbs": "[\"Running SQL queries\"]", "references": "[]"}
{"id": "sql_queries:id2", "page": "sql_queries", "ref": "id2", "title": "Pagination", "content": "Datasette's default table pagination is designed to be extremely efficient. SQL OFFSET/LIMIT pagination can have a significant performance penalty once you get into multiple thousands of rows, as each page still requires the database to scan through every preceding row to find the correct offset. \n             When paginating through tables, Datasette instead orders the rows in the table by their primary key and performs a WHERE clause against the last seen primary key for the previous page. For example: \n             select rowid, * from Tree_List where rowid > 200 order by rowid limit 101 \n             This represents page three for this particular table, with a page size of 100. \n             Note that we request 101 items in the limit clause rather than 100. This allows us to detect if we are on the last page of the results: if the query returns less than 101 rows we know we have reached the end of the pagination set. Datasette will only return the first 100 rows - the 101st is used purely to detect if there should be another page. \n             Since the where clause acts against the index on the primary key, the query is extremely fast even for records that are a long way into the overall pagination set.", "breadcrumbs": "[\"Running SQL queries\"]", "references": "[]"}
{"id": "sql_queries:id3", "page": "sql_queries", "ref": "id3", "title": "Cross-database queries", "content": "SQLite has the ability to run queries that join across multiple databases. Up to ten databases can be attached to a single SQLite connection and queried together. \n             Datasette can execute joins across multiple databases if it is started with the  --crossdb  option: \n             datasette fixtures.db extra_database.db --crossdb \n             If it is started in this way, the  /_memory  page can be used to execute queries that join across multiple databases. \n             References to tables in attached databases should be preceded by the database name and a period. \n             For example, this query will show a list of tables across both of the above databases: \n             select\n  'fixtures' as database, *\nfrom\n  [fixtures].sqlite_master\nunion\nselect\n  'extra_database' as database, *\nfrom\n  [extra_database].sqlite_master \n             Try that out here .", "breadcrumbs": "[\"Running SQL queries\"]", "references": "[{\"href\": \"https://latest.datasette.io/_memory?sql=select%0D%0A++%27fixtures%27+as+database%2C+*%0D%0Afrom%0D%0A++%5Bfixtures%5D.sqlite_master%0D%0Aunion%0D%0Aselect%0D%0A++%27extra_database%27+as+database%2C+*%0D%0Afrom%0D%0A++%5Bextra_database%5D.sqlite_master\", \"label\": \"Try that out here\"}]"}
{"id": "sql_queries:sql", "page": "sql_queries", "ref": "sql", "title": "Running SQL queries", "content": "Datasette treats SQLite database files as read-only and immutable. This means it is not possible to execute INSERT or UPDATE statements using Datasette, which allows us to expose SELECT statements to the outside world without needing to worry about SQL injection attacks. \n         The easiest way to execute custom SQL against Datasette is through the web UI. The database index page includes a SQL editor that lets you run any SELECT query you like. You can also construct queries using the filter interface on the tables page, then click \"View and edit SQL\" to open that query in the custom SQL editor. \n         Note that this interface is only available if the  execute-sql  permission is allowed. \n         Any Datasette SQL query is reflected in the URL of the page, allowing you to bookmark them, share them with others and navigate through previous queries using your browser back button. \n         You can also retrieve the results of any query as JSON by adding  .json  to the base URL.", "breadcrumbs": "[]", "references": "[]"}
{"id": "sql_queries:sql-parameters", "page": "sql_queries", "ref": "sql-parameters", "title": "Named parameters", "content": "Datasette has special support for SQLite named parameters. Consider a SQL query like this: \n             select * from Street_Tree_List\nwhere \"PermitNotes\" like :notes\nand \"qSpecies\" = :species \n             If you execute this query using the custom query editor, Datasette will extract the two named parameters and use them to construct form fields for you to provide values. \n             You can also provide values for these fields by constructing a URL: \n             /mydatabase?sql=select...&species=44 \n             SQLite string escaping rules will be applied to values passed using named parameters - they will be wrapped in quotes and their content will be correctly escaped. \n             Values from named parameters are treated as SQLite strings. If you need to perform numeric comparisons on them you should cast them to an integer or float first using  cast(:name as integer)  or  cast(:name as real) , for example: \n             select * from Street_Tree_List\nwhere latitude > cast(:min_latitude as real)\nand latitude < cast(:max_latitude as real) \n             Datasette disallows custom SQL queries containing the string PRAGMA (with a small number  of exceptions ) as SQLite pragma statements can be used to change database settings at runtime. If you need to include the string \"pragma\" in a query you can do so safely using a named parameter.", "breadcrumbs": "[\"Running SQL queries\"]", "references": "[{\"href\": \"https://github.com/simonw/datasette/issues/761\", \"label\": \"of exceptions\"}]"}
{"id": "sql_queries:sql-views", "page": "sql_queries", "ref": "sql-views", "title": "Views", "content": "If you want to bundle some pre-written SQL queries with your Datasette-hosted database you can do so in two ways. The first is to include SQL views in your database - Datasette will then list those views on your database index page. \n             The quickest way to create views is with the SQLite command-line interface: \n             $ sqlite3 sf-trees.db\nSQLite version 3.19.3 2017-06-27 16:48:08\nEnter \".help\" for usage hints.\nsqlite> CREATE VIEW demo_view AS select qSpecies from Street_Tree_List;\n<CTRL+D>", "breadcrumbs": "[\"Running SQL queries\"]", "references": "[]"}
{"id": "testing_plugins:id1", "page": "testing_plugins", "ref": "id1", "title": "Testing plugins", "content": "We recommend using  pytest  to write automated tests for your plugins. \n         If you use the template described in  Starting an installable plugin using cookiecutter  your plugin will start with a single test in your  tests/  directory that looks like this: \n         from datasette.app import Datasette\nimport pytest\n\n\n@pytest.mark.asyncio\nasync def test_plugin_is_installed():\n    datasette = Datasette(memory=True)\n    response = await datasette.client.get(\"/-/plugins.json\")\n    assert response.status_code == 200\n    installed_plugins = {p[\"name\"] for p in response.json()}\n    assert (\n        \"datasette-plugin-template-demo\"\n        in installed_plugins\n    ) \n         This test uses the  datasette.client  object to exercise a test instance of Datasette.  datasette.client  is a wrapper around the  HTTPX  Python library which can imitate HTTP requests using ASGI. This is the recommended way to write tests against a Datasette instance. \n         This test also uses the  pytest-asyncio  package to add support for  async def  test functions running under pytest. \n         You can install these packages like so: \n         pip install pytest pytest-asyncio \n         If you are building an installable package you can add them as test dependencies to your  setup.py  module like this: \n         setup(\n    name=\"datasette-my-plugin\",\n    # ...\n    extras_require={\"test\": [\"pytest\", \"pytest-asyncio\"]},\n    tests_require=[\"datasette-my-plugin[test]\"],\n) \n         You can then install the test dependencies like so: \n         pip install -e '.[test]' \n         Then run the tests using pytest like so: \n         pytest", "breadcrumbs": "[]", "references": "[{\"href\": \"https://docs.pytest.org/\", \"label\": \"pytest\"}, {\"href\": \"https://www.python-httpx.org/\", \"label\": \"HTTPX\"}, {\"href\": \"https://pypi.org/project/pytest-asyncio/\", \"label\": \"pytest-asyncio\"}]"}
{"id": "testing_plugins:testing-plugins-datasette-test-instance", "page": "testing_plugins", "ref": "testing-plugins-datasette-test-instance", "title": "Setting up a Datasette test instance", "content": "The above example shows the easiest way to start writing tests against a Datasette instance: \n             from datasette.app import Datasette\nimport pytest\n\n\n@pytest.mark.asyncio\nasync def test_plugin_is_installed():\n    datasette = Datasette(memory=True)\n    response = await datasette.client.get(\"/-/plugins.json\")\n    assert response.status_code == 200 \n             Creating a  Datasette()  instance like this as useful shortcut in tests, but there is one detail you need to be aware of. It's important to ensure that the async method  .invoke_startup()  is called on that instance. You can do that like this: \n             datasette = Datasette(memory=True)\nawait datasette.invoke_startup() \n             This method registers any  startup(datasette)  or  prepare_jinja2_environment(env, datasette)  plugins that might themselves need to make async calls. \n             If you are using  await datasette.client.get()  and similar methods then you don't need to worry about this - Datasette automatically calls  invoke_startup()  the first time it handles a request.", "breadcrumbs": "[\"Testing plugins\"]", "references": "[]"}
{"id": "testing_plugins:testing-plugins-fixtures", "page": "testing_plugins", "ref": "testing-plugins-fixtures", "title": "Using pytest fixtures", "content": "Pytest fixtures  can be used to create initial testable objects which can then be used by multiple tests. \n             A common pattern for Datasette plugins is to create a fixture which sets up a temporary test database and wraps it in a Datasette instance. \n             Here's an example that uses the  sqlite-utils library  to populate a temporary test database. It also sets the title of that table using a simulated  metadata.json  configuration: \n             from datasette.app import Datasette\nimport pytest\nimport sqlite_utils\n\n\n@pytest.fixture(scope=\"session\")\ndef datasette(tmp_path_factory):\n    db_directory = tmp_path_factory.mktemp(\"dbs\")\n    db_path = db_directory / \"test.db\"\n    db = sqlite_utils.Database(db_path)\n    db[\"dogs\"].insert_all(\n        [\n            {\"id\": 1, \"name\": \"Cleo\", \"age\": 5},\n            {\"id\": 2, \"name\": \"Pancakes\", \"age\": 4},\n        ],\n        pk=\"id\",\n    )\n    datasette = Datasette(\n        [db_path],\n        metadata={\n            \"databases\": {\n                \"test\": {\n                    \"tables\": {\n                        \"dogs\": {\"title\": \"Some dogs\"}\n                    }\n                }\n            }\n        },\n    )\n    return datasette\n\n\n@pytest.mark.asyncio\nasync def test_example_table_json(datasette):\n    response = await datasette.client.get(\n        \"/test/dogs.json?_shape=array\"\n    )\n    assert response.status_code == 200\n    assert response.json() == [\n        {\"id\": 1, \"name\": \"Cleo\", \"age\": 5},\n        {\"id\": 2, \"name\": \"Pancakes\", \"age\": 4},\n    ]\n\n\n@pytest.mark.asyncio\nasync def test_example_table_html(datasette):\n    response = await datasette.client.get(\"/test/dogs\")\n    assert \">Some dogs</h1>\" in response.text \n             Here the  datasette()  function defines the fixture, which is than automatically passed to the two test functions based on pytest automatically matching their  datasette  function parameters. \n             The  @pytest.fixture(scope=\"session\")  line here ensures the fixture is reused for the full  pytest  execution session. This means that the temporary database file will be created once and reused for each test. \n             If you want to create that test database repeatedly for every individual test function, write the fixture function like this instead. You may want to do this if your plugin modifies the database contents in some way: \n             @pytest.fixture\ndef datasette(tmp_path_factory):\n    # This fixture will be executed repeatedly for every test\n    ...", "breadcrumbs": "[\"Testing plugins\"]", "references": "[{\"href\": \"https://docs.pytest.org/en/stable/fixture.html\", \"label\": \"Pytest fixtures\"}, {\"href\": \"https://sqlite-utils.datasette.io/en/stable/python-api.html\", \"label\": \"sqlite-utils library\"}]"}
{"id": "testing_plugins:testing-plugins-pdb", "page": "testing_plugins", "ref": "testing-plugins-pdb", "title": "Using pdb for errors thrown inside Datasette", "content": "If an exception occurs within Datasette itself during a test, the response returned to your plugin will have a  response.status_code  value of 500. \n             You can add  pdb=True  to the  Datasette  constructor to drop into a Python debugger session inside your test run instead of getting back a 500 response code. This is equivalent to running the  datasette  command-line tool with the  --pdb  option. \n             Here's what that looks like in a test function: \n             def test_that_opens_the_debugger_or_errors():\n    ds = Datasette([db_path], pdb=True)\n    response = await ds.client.get(\"/\") \n             If you use this pattern you will need to run  pytest  with the  -s  option to avoid capturing stdin/stdout in order to interact with the debugger prompt.", "breadcrumbs": "[\"Testing plugins\"]", "references": "[]"}
{"id": "testing_plugins:testing-plugins-pytest-httpx", "page": "testing_plugins", "ref": "testing-plugins-pytest-httpx", "title": "Testing outbound HTTP calls with pytest-httpx", "content": "If your plugin makes outbound HTTP calls - for example datasette-auth-github or datasette-import-table - you may need to mock those HTTP requests in your tests. \n             The  pytest-httpx  package is a useful library for mocking calls. It can be tricky to use with Datasette though since it mocks all HTTPX requests, and Datasette's own testing mechanism uses HTTPX internally. \n             To avoid breaking your tests, you can return  [\"localhost\"]  from the  non_mocked_hosts()  fixture. \n             As an example, here's a very simple plugin which executes an HTTP response and returns the resulting content: \n             from datasette import hookimpl\nfrom datasette.utils.asgi import Response\nimport httpx\n\n\n@hookimpl\ndef register_routes():\n    return [\n        (r\"^/-/fetch-url$\", fetch_url),\n    ]\n\n\nasync def fetch_url(datasette, request):\n    if request.method == \"GET\":\n        return Response.html(\n            \"\"\"\n            <form action=\"/-/fetch-url\" method=\"post\">\n            <input type=\"hidden\" name=\"csrftoken\" value=\"{}\">\n            <input name=\"url\"><input type=\"submit\">\n        </form>\"\"\".format(\n                request.scope[\"csrftoken\"]()\n            )\n        )\n    vars = await request.post_vars()\n    url = vars[\"url\"]\n    return Response.text(httpx.get(url).text) \n             Here's a test for that plugin that mocks the HTTPX outbound request: \n             from datasette.app import Datasette\nimport pytest\n\n\n@pytest.fixture\ndef non_mocked_hosts():\n    # This ensures httpx-mock will not affect Datasette's own\n    # httpx calls made in the tests by datasette.client:\n    return [\"localhost\"]\n\n\nasync def test_outbound_http_call(httpx_mock):\n    httpx_mock.add_response(\n        url=\"https://www.example.com/\",\n        text=\"Hello world\",\n    )\n    datasette = Datasette([], memory=True)\n    response = await datasette.client.post(\n        \"/-/fetch-url\",\n        data={\"url\": \"https://www.example.com/\"},\n    )\n    assert response.text == \"Hello world\"\n\n    outbound_request = httpx_mock.get_request()\n    assert (\n        outbound_request.url == \"https://www.example.com/\"\n    )", "breadcrumbs": "[\"Testing plugins\"]", "references": "[{\"href\": \"https://pypi.org/project/pytest-httpx/\", \"label\": \"pytest-httpx\"}]"}
{"id": "testing_plugins:testing-plugins-register-in-test", "page": "testing_plugins", "ref": "testing-plugins-register-in-test", "title": "Registering a plugin for the duration of a test", "content": "When writing tests for plugins you may find it useful to register a test plugin just for the duration of a single test. You can do this using  pm.register()  and  pm.unregister()  like this: \n             from datasette import hookimpl\nfrom datasette.app import Datasette\nfrom datasette.plugins import pm\nimport pytest\n\n\n@pytest.mark.asyncio\nasync def test_using_test_plugin():\n    class TestPlugin:\n        __name__ = \"TestPlugin\"\n\n        # Use hookimpl and method names to register hooks\n        @hookimpl\n        def register_routes(self):\n            return [\n                (r\"^/error$\", lambda: 1 / 0),\n            ]\n\n    pm.register(TestPlugin(), name=\"undo\")\n    try:\n        # The test implementation goes here\n        datasette = Datasette()\n        response = await datasette.client.get(\"/error\")\n        assert response.status_code == 500\n    finally:\n        pm.unregister(name=\"undo\")", "breadcrumbs": "[\"Testing plugins\"]", "references": "[]"}
{"id": "writing_plugins:id1", "page": "writing_plugins", "ref": "id1", "title": "Writing plugins", "content": "You can write one-off plugins that apply to just one Datasette instance, or you can write plugins which can be installed using  pip  and can be shipped to the Python Package Index ( PyPI ) for other people to install. \n         Want to start by looking at an example? The  Datasette plugins directory  lists more than 90 open source plugins with code you can explore. The  plugin hooks  page includes links to example plugins for each of the documented hooks.", "breadcrumbs": "[]", "references": "[{\"href\": \"https://pypi.org/\", \"label\": \"PyPI\"}, {\"href\": \"https://datasette.io/plugins\", \"label\": \"Datasette plugins directory\"}]"}
{"id": "writing_plugins:writing-plugins-building-urls", "page": "writing_plugins", "ref": "writing-plugins-building-urls", "title": "Building URLs within plugins", "content": "Plugins that define their own custom user interface elements may need to link to other pages within Datasette. \n             This can be a bit tricky if the Datasette instance is using the  base_url  configuration setting to run behind a proxy, since that can cause Datasette's URLs to include an additional prefix. \n             The  datasette.urls  object provides internal methods for correctly generating URLs to different pages within Datasette, taking any  base_url  configuration into account. \n             This object is exposed in templates as the  urls  variable, which can be used like this: \n             Back to the <a href=\"{{ urls.instance() }}\">Homepage</a> \n             See  datasette.urls  for full details on this object.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[]"}
{"id": "writing_plugins:writing-plugins-configuration", "page": "writing_plugins", "ref": "writing-plugins-configuration", "title": "Writing plugins that accept configuration", "content": "When you are writing plugins, you can access plugin configuration like this using the  datasette plugin_config()  method. If you know you need plugin configuration for a specific table, you can access it like this: \n             plugin_config = datasette.plugin_config(\n    \"datasette-cluster-map\", database=\"sf-trees\", table=\"Street_Tree_List\"\n) \n             This will return the  {\"latitude_column\": \"lat\", \"longitude_column\": \"lng\"}  in the above example. \n             If there is no configuration for that plugin, the method will return  None . \n             If it cannot find the requested configuration at the table layer, it will fall back to the database layer and then the root layer. For example, a user may have set the plugin configuration option like so: \n             {\n    \"databases: {\n        \"sf-trees\": {\n            \"plugins\": {\n                \"datasette-cluster-map\": {\n                    \"latitude_column\": \"xlat\",\n                    \"longitude_column\": \"xlng\"\n                }\n            }\n        }\n    }\n} \n             In this case, the above code would return that configuration for ANY table within the  sf-trees  database. \n             The plugin configuration could also be set at the top level of  metadata.json : \n             {\n    \"title\": \"This is the top-level title in metadata.json\",\n    \"plugins\": {\n        \"datasette-cluster-map\": {\n            \"latitude_column\": \"xlat\",\n            \"longitude_column\": \"xlng\"\n        }\n    }\n} \n             Now that  datasette-cluster-map  plugin configuration will apply to every table in every database.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[]"}
{"id": "writing_plugins:writing-plugins-cookiecutter", "page": "writing_plugins", "ref": "writing-plugins-cookiecutter", "title": "Starting an installable plugin using cookiecutter", "content": "Plugins that can be installed should be written as Python packages using a  setup.py  file. \n             The quickest way to start writing one an installable plugin is to use the  datasette-plugin  cookiecutter template. This creates a new plugin structure for you complete with an example test and GitHub Actions workflows for testing and publishing your plugin. \n             Install cookiecutter  and then run this command to start building a plugin using the template: \n             cookiecutter gh:simonw/datasette-plugin \n             Read  a cookiecutter template for writing Datasette plugins  for more information about this template.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[{\"href\": \"https://github.com/simonw/datasette-plugin\", \"label\": \"datasette-plugin\"}, {\"href\": \"https://cookiecutter.readthedocs.io/en/stable/installation.html\", \"label\": \"Install cookiecutter\"}, {\"href\": \"https://simonwillison.net/2020/Jun/20/cookiecutter-plugins/\", \"label\": \"a cookiecutter template for writing Datasette plugins\"}]"}
{"id": "writing_plugins:writing-plugins-custom-templates", "page": "writing_plugins", "ref": "writing-plugins-custom-templates", "title": "Custom templates", "content": "If your plugin has a  templates/  directory, Datasette will attempt to load templates from that directory before it uses its own default templates. \n             The priority order for template loading is: \n             \n                 \n                     templates from the  --template-dir  argument, if specified \n                 \n                 \n                     templates from the  templates/  directory in any installed plugins \n                 \n                 \n                     default templates that ship with Datasette \n                 \n             \n             See  Custom pages and templates  for more details on how to write custom templates, including which filenames to use to customize which parts of the Datasette UI. \n             Templates should be bundled for distribution using the same  package_data  mechanism in  setup.py  described for static assets above, for example: \n             package_data = (\n    {\n        \"datasette_plugin_name\": [\n            \"templates/my_template.html\",\n        ],\n    },\n) \n             You can also use wildcards here such as  templates/*.html . See  datasette-edit-schema  for an example of this pattern.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[{\"href\": \"https://github.com/simonw/datasette-edit-schema\", \"label\": \"datasette-edit-schema\"}]"}
{"id": "writing_plugins:writing-plugins-designing-urls", "page": "writing_plugins", "ref": "writing-plugins-designing-urls", "title": "Designing URLs for your plugin", "content": "You can register new URL routes within Datasette using the  register_routes(datasette)  plugin hook. \n             Datasette's default URLs include these: \n             \n                 \n                     /dbname  - database page \n                 \n                 \n                     /dbname/tablename  - table page \n                 \n                 \n                     /dbname/tablename/pk  - row page \n                 \n             \n             See  Pages and API endpoints  and  Introspection  for more default URL routes. \n             To avoid accidentally conflicting with a database file that may be loaded into Datasette, plugins should register URLs using a  /-/  prefix. For example, if your plugin adds a new interface for uploading Excel files you might register a URL route like this one: \n             \n                 \n                     /-/upload-excel \n                 \n             \n             Try to avoid registering URLs that clash with other plugins that your users might have installed. There is no central repository of reserved URL paths (yet) but you can review existing plugins by browsing the  plugins directory <https://datasette.io/plugins> . \n             If your plugin includes functionality that relates to a specific database you could also register a URL route like this: \n             \n                 \n                     /dbname/-/upload-excel \n                 \n             \n             Or for a specific table like this: \n             \n                 \n                     /dbname/tablename/-/modify-table-schema \n                 \n             \n             Note that a row could have a primary key of  -  and this URL scheme will still work, because Datasette row pages do not ever have a trailing slash followed by additional path components.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[]"}
{"id": "writing_plugins:writing-plugins-one-off", "page": "writing_plugins", "ref": "writing-plugins-one-off", "title": "Writing one-off plugins", "content": "The quickest way to start writing a plugin is to create a  my_plugin.py  file and drop it into your  plugins/  directory. Here is an example plugin, which adds a new custom SQL function called  hello_world()  which takes no arguments and returns the string  Hello world! . \n             from datasette import hookimpl\n\n\n@hookimpl\ndef prepare_connection(conn):\n    conn.create_function(\n        \"hello_world\", 0, lambda: \"Hello world!\"\n    ) \n             If you save this in  plugins/my_plugin.py  you can then start Datasette like this: \n             datasette serve mydb.db --plugins-dir=plugins/ \n             Now you can navigate to  http://localhost:8001/mydb  and run this SQL: \n             select hello_world(); \n             To see the output of your plugin.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[{\"href\": \"http://localhost:8001/mydb\", \"label\": \"http://localhost:8001/mydb\"}]"}
{"id": "writing_plugins:writing-plugins-packaging", "page": "writing_plugins", "ref": "writing-plugins-packaging", "title": "Packaging a plugin", "content": "Plugins can be packaged using Python setuptools. You can see an example of a packaged plugin at  https://github.com/simonw/datasette-plugin-demos \n             The example consists of two files: a  setup.py  file that defines the plugin: \n             from setuptools import setup\n\nVERSION = \"0.1\"\n\nsetup(\n    name=\"datasette-plugin-demos\",\n    description=\"Examples of plugins for Datasette\",\n    author=\"Simon Willison\",\n    url=\"https://github.com/simonw/datasette-plugin-demos\",\n    license=\"Apache License, Version 2.0\",\n    version=VERSION,\n    py_modules=[\"datasette_plugin_demos\"],\n    entry_points={\n        \"datasette\": [\n            \"plugin_demos = datasette_plugin_demos\"\n        ]\n    },\n    install_requires=[\"datasette\"],\n) \n             And a Python module file,  datasette_plugin_demos.py , that implements the plugin: \n             from datasette import hookimpl\nimport random\n\n\n@hookimpl\ndef prepare_jinja2_environment(env):\n    env.filters[\"uppercase\"] = lambda u: u.upper()\n\n\n@hookimpl\ndef prepare_connection(conn):\n    conn.create_function(\n        \"random_integer\", 2, random.randint\n    ) \n             Having built a plugin in this way you can turn it into an installable package using the following command: \n             python3 setup.py sdist \n             This will create a  .tar.gz  file in the  dist/  directory. \n             You can then install your new plugin into a Datasette virtual environment or Docker container using  pip : \n             pip install datasette-plugin-demos-0.1.tar.gz \n             To learn how to upload your plugin to  PyPI  for use by other people, read the PyPA guide to  Packaging and distributing projects .", "breadcrumbs": "[\"Writing plugins\"]", "references": "[{\"href\": \"https://github.com/simonw/datasette-plugin-demos\", \"label\": \"https://github.com/simonw/datasette-plugin-demos\"}, {\"href\": \"https://pypi.org/\", \"label\": \"PyPI\"}, {\"href\": \"https://packaging.python.org/tutorials/distributing-packages/\", \"label\": \"Packaging and distributing projects\"}]"}
{"id": "writing_plugins:writing-plugins-static-assets", "page": "writing_plugins", "ref": "writing-plugins-static-assets", "title": "Static assets", "content": "If your plugin has a  static/  directory, Datasette will automatically configure itself to serve those static assets from the following path: \n             /-/static-plugins/NAME_OF_PLUGIN_PACKAGE/yourfile.js \n             Use the  datasette.urls.static_plugins(plugin_name, path)  method to generate URLs to that asset that take the  base_url  setting into account, see  datasette.urls . \n             To bundle the static assets for a plugin in the package that you publish to PyPI, add the following to the plugin's  setup.py : \n             package_data = (\n    {\n        \"datasette_plugin_name\": [\n            \"static/plugin.js\",\n        ],\n    },\n) \n             Where  datasette_plugin_name  is the name of the plugin package (note that it uses underscores, not hyphens) and  static/plugin.js  is the path within that package to the static file. \n             datasette-cluster-map  is a useful example of a plugin that includes packaged static assets in this way.", "breadcrumbs": "[\"Writing plugins\"]", "references": "[{\"href\": \"https://github.com/simonw/datasette-cluster-map\", \"label\": \"datasette-cluster-map\"}]"}