refactor backend code into repository module, remove PL/Python

docker/entrypoint.py

@@ -45,14 +45,10 @@
 import os
 import configparser
 import sys
-from time import sleep
 
-from sqlalchemy import create_engine
-from sqlalchemy.exc import OperationalError
-from sqlalchemy.exc import ProgrammingError
-
-from pycsw.core import admin
+from pycsw.core.config import StaticContext
 from pycsw.core.util import EnvInterpolation
+from pycsw.core.repository import Repository, setup
 
 LOGGER = logging.getLogger(__name__)
 
@@ -77,28 +73,17 @@ def launch_pycsw(pycsw_config, workers=2, reload=False):
     for more information on how to control gunicorn by sending UNIX signals.
     """
 
-    db_url = pycsw_config.get("repository", "database")
-
-    db = db_url.partition(":")[0].partition("+")[0]
-
-    db_handler = {
-        "sqlite": handle_sqlite_db,
-        "postgresql": handle_postgresql_db,
-    }.get(db)
-
-    # FIXME: pycsw.wsgi_flask is bound to PostgreSQL backends only
-    # An update in pycsw/ogc/api/records.py is required
-    if db_handler is not None:
-        LOGGER.debug("Setting up pycsw's data repository...")
-        LOGGER.debug(f"Repository URL: {db_url}")
-        db_handler(
-            db_url,
-            pycsw_config.get("repository", "table"),
-            pycsw_config.get("server", "home")
-        )
-        pycsw_app = 'pycsw.wsgi_flask:APP'
-    else:
-        pycsw_app = 'pycsw.wsgi'
+    try:
+        setup(pycsw_config.get("repository", "database"),
+              pycsw_config.get("repository", "table"))
+    except Exception as err:
+        LOGGER.debug(err)
+
+    repo = Repository(pycsw_config.get("repository", "database"),
+                      StaticContext(),
+                      table=pycsw_config.get("repository", "table"))
+
+    repo.ping()
 
     sys.stdout.flush()
     # we're using --reload-engine=poll because there is a bug with gunicorn
@@ -119,7 +104,7 @@ def launch_pycsw(pycsw_config, workers=2, reload=False):
         "--error-logfile=-",
         f"--workers={workers}",
         f"--timeout={timeout}",
-        pycsw_app,
+        'pycsw.wsgi_flask:APP'
 
     ]
     LOGGER.debug(f"Launching pycsw with {' '.join(execution_args)} ...")
@@ -129,46 +114,6 @@ def launch_pycsw(pycsw_config, workers=2, reload=False):
     )
 
 
-def handle_sqlite_db(database_url, table_name, pycsw_home):
-    db_path = database_url.rpartition(":///")[-1]
-    if not os.path.isfile(db_path):
-        try:
-            os.makedirs(os.path.dirname(db_path))
-        except OSError as exc:
-            if exc.args[0] == 17:  # directory already exists
-                pass
-        admin.setup_db(database=database_url, table=table_name,
-                       home=pycsw_home)
-
-
-def handle_postgresql_db(database_url, table_name, pycsw_home):
-    _wait_for_postgresql_db(database_url)
-    try:
-        admin.setup_db(database=database_url, table=table_name,
-                       home=pycsw_home)
-    except ProgrammingError:
-        pass  # database tables are already created
-
-
-def _wait_for_postgresql_db(database_url, max_tries=10, wait_seconds=3):
-    LOGGER.debug(f"Waiting for {database_url}...")
-    engine = create_engine(database_url)
-    current_try = 0
-    while current_try < max_tries:
-        try:
-            engine.execute("SELECT version();")
-            LOGGER.debug("Database is already up!")
-            break
-        except OperationalError:
-            LOGGER.debug("Database not responding yet ...")
-            current_try += 1
-            sleep(wait_seconds)
-    else:
-        raise RuntimeError(
-            f"Database not responding at {database_url} after {max_tries} tries. "
-        )
-
-
 if __name__ == "__main__":
     parser = argparse.ArgumentParser()
     parser.add_argument(

docs/administration.rst

@@ -54,7 +54,7 @@ Setting up the Database
 
 .. code-block:: bash
 
-  pycsw-admin.py setup_db --config default.cfg
+  pycsw-admin.py setup_repository --config default.cfg
 
 This will create the necessary tables and values for the repository.
 
@@ -120,8 +120,8 @@ Database Specific Notes
 PostgreSQL
 ^^^^^^^^^^
 
-- if PostGIS is not enabled, pycsw makes uses of PL/Python functions.  To enable PostgreSQL support, the database user must be able to create functions within the database. In case of recent PostgreSQL versions (9.x), the PL/Python extension must be enabled prior to pycsw setup
-- `PostgreSQL Full Text Search`_ is supported for ``csw:AnyText`` based queries.  pycsw creates a tsvector column based on the text from anytext column. Then pycsw creates a GIN index against the anytext_tsvector column.  This is created automatically in ``pycsw.admin.setup_db``.  Any query against OARec's ``q`` parameter or CSW `csw:AnyText` or `apiso:AnyText` will process using PostgreSQL FTS handling
+-  To enable PostgreSQL support, the database user must be able to create functions within the database.
+- `PostgreSQL Full Text Search`_ is supported for ``csw:AnyText`` based queries.  pycsw creates a tsvector column based on the text from anytext column. Then pycsw creates a GIN index against the anytext_tsvector column.  This is created automatically in ``pycsw.core.repository.setup``.  Any query against OARec's ``q`` parameter or CSW `csw:AnyText` or `apiso:AnyText` will process using PostgreSQL FTS handling
 
 PostGIS
 ^^^^^^^
@@ -131,7 +131,7 @@ PostGIS
 - If PostGIS is detected, the ``pycsw-admin.py`` script will create both a native geometry column and a WKT column, as well as a trigger to keep both synchronized
 - In case PostGIS gets disabled, pycsw will continue to work with the `WKT`_ column
 - In case of migration from plain PostgreSQL database to PostGIS, the spatial functions of PostGIS will be used automatically
-- When migrating from plain PostgreSQL database to PostGIS, in order to enable native geometry support, a "GEOMETRY" column named "wkb_geometry" needs to be created manually (along with the update trigger in ``pycsw.admin.setup_db``). Also the native geometries must be filled manually from the `WKT`_ field. Next versions of pycsw will automate this process
+- When migrating from plain PostgreSQL database to PostGIS, in order to enable native geometry support, a "GEOMETRY" column named "wkb_geometry" needs to be created manually (along with the update trigger in ``pycsw.core.repository.setup``). Also the native geometries must be filled manually from the `WKT`_ field. Next versions of pycsw will automate this process
 
 .. _custom_repository:
 

docs/locale/zh/LC_MESSAGES/administration.po

@@ -196,26 +196,22 @@ msgstr "PostgreSQL"
 
 #: ../../administration.rst:119
 msgid ""
-"if PostGIS is not enabled, pycsw makes uses of PL/Python functions.  To enable "
-"PostgreSQL support, the database user must be able to create functions within "
-"the database. In case of recent PostgreSQL versions (9.x), the PL/Python "
-"extension must be enabled prior to pycsw setup"
+"To enable the database user must be able to create functions within the database."
 msgstr ""
-"若 PostGIS不可用，pycsw会启用 PL/Python功能。在PostgreSQL支持下，用户必须在数据"
-"库中创建这些功能。如果是最新版本PostgreSQL (9.x)，PL/ Python扩展"
+"在PostgreSQL支持下，用户必须在数据"
 
 #: ../../administration.rst:120
 msgid ""
 "`PostgreSQL Full Text Search`_ is supported for ``csw:AnyText`` based queries.  "
 "pycsw creates a tsvector column based on the text from anytext column. Then "
 "pycsw creates a GIN index against the anytext_tsvector column.  This is created "
-"automatically in ``pycsw.admin.setup_db``.  Any query against OARec's ``q`` "
+"automatically in ``pycsw.core.repository.setup``.  Any query against OARec's ``q`` "
 "parameter or CSW `csw:AnyText` or `apiso:AnyText` will process using PostgreSQL "
 "FTS handling"
 msgstr ""
 "`PostgreSQL Full Text Search`_ (全文搜索)支持基于 ``csw:AnyText`` （任意文本）的"
 "查询。pycsw创建的tsvector栏是基于文本anytext栏。然而pycsw针对"
-"anytext_tsvector（任意文本）栏创建了GIN索引。这在 ``pycsw.admin.setup_db`` 是自"
+"anytext_tsvector（任意文本）栏创建了GIN索引。这在 ``pycsw.core.repository.setup`` 是自"
 "动生成的。任何针对 `csw:AnyText` or `apiso:AnyText` 的查询都是使用PostgreSQL FT"
 "进行处理"
 
@@ -258,13 +254,13 @@ msgstr "如果plain PostgreSQL数据库移入PostGIS中，PostGIS的空间功能
 msgid ""
 "When migrating from plain PostgreSQL database to PostGIS, in order to enable "
 "native geometry support, a \"GEOMETRY\" column named \"wkb_geometry\" needs to "
-"be created manually (along with the update trigger in ``pycsw.admin."
-"setup_db``). Also the native geometries must be filled manually from the `WKT`_ "
+"be created manually (along with the update trigger in ``pycsw.core.repository."
+"setup``). Also the native geometries must be filled manually from the `WKT`_ "
 "field. Next versions of pycsw will automate this process"
 msgstr ""
 "当从普通的数据库plain PostgreSQL移入PostGIS时， 为了启用本地geometry支持，需要手"
-"动创建一个名为 \"wkb_geometry\" 的 \"GEOMETRY\" 列（随着 ``pycsw.admin."
-"setup_db`` 一起更新）。本地geometries也必须在 `WKT`_ 中手动填充。pycsw的下一个版"
+"动创建一个名为 \"wkb_geometry\" 的 \"GEOMETRY\" 列（随着 ``pycsw.core.repository."
+"setup`` 一起更新）。本地geometries也必须在 `WKT`_ 中手动填充。pycsw的下一个版"
 "本就会将此过程自动化运行"
 
 #: ../../administration.rst:135