remove restriction on frequent one word names

[nominatim.git] / nominatim / server / falcon / server.py

diff --git a/nominatim/server/falcon/server.py b/nominatim/server/falcon/server.py
index 81e6ed396846d34cd47f534231ade5ab8da03a7b..bdae350744e6774ff927f8bec597da8b05ee90e6 100644 (file)
--- a/nominatim/server/falcon/server.py
+++ b/nominatim/server/falcon/server.py
@@ -2,81 +2,193 @@
 #
 # This file is part of Nominatim. (https://nominatim.org)
 #
 #
 # This file is part of Nominatim. (https://nominatim.org)
 #

-# Copyright (C) 2022 by the Nominatim developer community.
+# Copyright (C) 2023 by the Nominatim developer community.

 # For a full list of authors see the git log.
 """
 Server implementation using the falcon webserver framework.
 """
 # For a full list of authors see the git log.
 """
 Server implementation using the falcon webserver framework.
 """

-from typing import Type, Any, Optional, Mapping
+from typing import Optional, Mapping, cast, Any, List

 from pathlib import Path
 from pathlib import Path

+import datetime as dt
+import asyncio

 
 

-import falcon
-import falcon.asgi
+from falcon.asgi import App, Request, Response

 
 from nominatim.api import NominatimAPIAsync
 
 from nominatim.api import NominatimAPIAsync

-from nominatim.apicmd.status import StatusResult
-import nominatim.result_formatter.v1 as formatting
+import nominatim.api.v1 as api_impl
+import nominatim.api.logging as loglib
+from nominatim.config import Configuration
+
+class HTTPNominatimError(Exception):
+    """ A special exception class for errors raised during processing.
+    """
+    def __init__(self, msg: str, status: int, content_type: str) -> None:
+        self.msg = msg
+        self.status = status
+        self.content_type = content_type
+
+
+async def nominatim_error_handler(req: Request, resp: Response, #pylint: disable=unused-argument
+                                  exception: HTTPNominatimError,
+                                  _: Any) -> None:
+    """ Special error handler that passes message and content type as
+        per exception info.
+    """
+    resp.status = exception.status
+    resp.text = exception.msg
+    resp.content_type = exception.content_type
+
+
+async def timeout_error_handler(req: Request, resp: Response, #pylint: disable=unused-argument
+                                exception: TimeoutError, #pylint: disable=unused-argument
+                                _: Any) -> None:
+    """ Special error handler that passes message and content type as
+        per exception info.
+    """
+    resp.status = 503
+
+    loglib.log().comment('Aborted: Query took too long to process.')
+    logdata = loglib.get_and_disable()
+    if logdata:
+        resp.text = logdata
+        resp.content_type = 'text/html; charset=utf-8'
+    else:
+        resp.text = "Query took too long to process."
+        resp.content_type = 'text/plain; charset=utf-8'

 
 

-CONTENT_TYPE = {
-  'text': falcon.MEDIA_TEXT,
-  'xml': falcon.MEDIA_XML
-}

 
 

-class NominatimV1:
-    """ Implementation of V1 version of the Nominatim API.
+class ParamWrapper(api_impl.ASGIAdaptor):
+    """ Adaptor class for server glue to Falcon framework.

     """
 
     """
 

-    def __init__(self, project_dir: Path, environ: Optional[Mapping[str, str]]) -> None:
-        self.api = NominatimAPIAsync(project_dir, environ)
-        self.formatters = {}
+    def __init__(self, req: Request, resp: Response,
+                 config: Configuration) -> None:
+        self.request = req
+        self.response = resp
+        self._config = config
+
+
+    def get(self, name: str, default: Optional[str] = None) -> Optional[str]:
+        return cast(Optional[str], self.request.get_param(name, default=default))
+
+
+    def get_header(self, name: str, default: Optional[str] = None) -> Optional[str]:
+        return cast(Optional[str], self.request.get_header(name, default=default))
+
+
+    def error(self, msg: str, status: int = 400) -> HTTPNominatimError:
+        return HTTPNominatimError(msg, status, self.content_type)
+
+
+    def create_response(self, status: int, output: str, num_results: int) -> None:
+        self.response.context.num_results = num_results
+        self.response.status = status
+        self.response.text = output
+        self.response.content_type = self.content_type
+

 
 

-        for rtype in (StatusResult, ):
-            self.formatters[rtype] = formatting.create(rtype)
+    def base_uri(self) -> str:
+        return cast (str, self.request.forwarded_prefix)

 
 

+    def config(self) -> Configuration:
+        return self._config

 
 

-    def parse_format(self, req: falcon.asgi.Request, rtype: Type[Any], default: str) -> None:
-        """ Get and check the 'format' parameter and prepare the formatter.
-            `rtype` describes the expected return type and `default` the
-            format value to assume when no parameter is present.
+
+class EndpointWrapper:
+    """ Converter for server glue endpoint functions to Falcon request handlers.
+    """
+
+    def __init__(self, name: str, func: api_impl.EndpointFunc, api: NominatimAPIAsync) -> None:
+        self.name = name
+        self.func = func
+        self.api = api
+
+
+    async def on_get(self, req: Request, resp: Response) -> None:
+        """ Implementation of the endpoint.

         """
         """

-        req.context.format = req.get_param('format', default=default)
-        req.context.formatter = self.formatters[rtype]
+        await self.func(self.api, ParamWrapper(req, resp, self.api.config))

 
 

-        if not req.context.formatter.supports_format(req.context.format):
-            raise falcon.HTTPBadRequest(
-                description="Parameter 'format' must be one of: " +
-                            ', '.join(req.context.formatter.list_formats()))

 
 

+class FileLoggingMiddleware:
+    """ Middleware to log selected requests into a file.
+    """
+
+    def __init__(self, file_name: str):
+        self.fd = open(file_name, 'a', buffering=1, encoding='utf8') # pylint: disable=R1732

 
 

-    def format_response(self, req: falcon.asgi.Request, resp: falcon.asgi.Response,
-                        result: Any) -> None:
-        """ Render response into a string according to the formatter
-            set in `parse_format()`.
+
+    async def process_request(self, req: Request, _: Response) -> None:
+        """ Callback before the request starts timing.

         """
         """

-        resp.text = req.context.formatter.format(result, req.context.format)
-        resp.content_type = CONTENT_TYPE.get(req.context.format, falcon.MEDIA_JSON)
+        req.context.start = dt.datetime.now(tz=dt.timezone.utc)

 
 
 
 

-    async def on_get_status(self, req: falcon.asgi.Request, resp: falcon.asgi.Response) -> None:
-        """ Implementation of status endpoint.
+    async def process_response(self, req: Request, resp: Response,
+                               resource: Optional[EndpointWrapper],
+                               req_succeeded: bool) -> None:
+        """ Callback after requests writes to the logfile. It only
+            writes logs for successful requests for search, reverse and lookup.

         """
         """

-        self.parse_format(req, StatusResult, 'text')
+        if not req_succeeded or resource is None or resp.status != 200\
+            or resource.name not in ('reverse', 'search', 'lookup', 'details'):
+            return

 
 

-        result = await self.api.status()
+        finish = dt.datetime.now(tz=dt.timezone.utc)
+        duration = (finish - req.context.start).total_seconds()
+        params = req.scope['query_string'].decode('utf8')
+        start = req.context.start.replace(tzinfo=None)\
+                                 .isoformat(sep=' ', timespec='milliseconds')

 
 

-        self.format_response(req, resp, result)
-        if result.status and req.context.format == 'text':
-            resp.status = 500
+        self.fd.write(f"[{start}] "
+                      f"{duration:.4f} {getattr(resp.context, 'num_results', 0)} "
+                      f'{resource.name} "{params}"\n')

 
 
 
 

-def get_application(project_dir: Path,
-                    environ: Optional[Mapping[str, str]] = None) -> falcon.asgi.App:
-    """ Create a Nominatim falcon ASGI application.
+class APIShutdown:
+    """ Middleware that closes any open database connections.

     """
     """

-    app = falcon.asgi.App()

 
 

-    api = NominatimV1(project_dir, environ)
+    def __init__(self, api: NominatimAPIAsync) -> None:
+        self.api = api

 
 

-    app.add_route('/status', api, suffix='status')
+    async def process_shutdown(self, *_: Any) -> None:
+        """Process the ASGI lifespan shutdown event.
+        """
+        await self.api.close()
+
+
+def get_application(project_dir: Path,
+                    environ: Optional[Mapping[str, str]] = None) -> App:
+    """ Create a Nominatim Falcon ASGI application.
+    """
+    api = NominatimAPIAsync(project_dir, environ)
+
+    middleware: List[object] = [APIShutdown(api)]
+    log_file = api.config.LOG_FILE
+    if log_file:
+        middleware.append(FileLoggingMiddleware(log_file))
+
+    app = App(cors_enable=api.config.get_bool('CORS_NOACCESSCONTROL'),
+              middleware=middleware)
+    app.add_error_handler(HTTPNominatimError, nominatim_error_handler)
+    app.add_error_handler(TimeoutError, timeout_error_handler)
+    # different from TimeoutError in Python <= 3.10
+    app.add_error_handler(asyncio.TimeoutError, timeout_error_handler)
+
+    legacy_urls = api.config.get_bool('SERVE_LEGACY_URLS')
+    for name, func in api_impl.ROUTES:
+        endpoint = EndpointWrapper(name, func, api)
+        app.add_route(f"/{name}", endpoint)
+        if legacy_urls:
+            app.add_route(f"/{name}.php", endpoint)

 
     return app
 
     return app

+
+
+def run_wsgi() -> App:
+    """ Entry point for uvicorn.
+
+        Make sure uvicorn is run from the project directory.
+    """
+    return get_application(Path('.'))