Merge pull request #300 from NatLibFi/issue282-handling-syntax-errors…

…-in-projects-config

Issue282 handling syntax errors in projects config

annif/default_config.py

@@ -36,3 +36,7 @@ class TestingInitializeConfig(TestingConfig):
 
 class TestingNoProjectsConfig(TestingConfig):
     PROJECTS_FILE = 'tests/notfound.cfg'
+
+
+class TestingInvalidProjectsConfig(TestingConfig):
+    PROJECTS_FILE = 'tests/projects_invalid.cfg'

annif/project.py

@@ -63,10 +63,13 @@ def _init_access(self):
                 project_id=self.project_id)
 
     def _initialize_analyzer(self):
-        analyzer = self.analyzer
-        logger.debug("Project '%s': initialized analyzer: %s",
-                     self.project_id,
-                     str(analyzer))
+        try:
+            analyzer = self.analyzer
+            logger.debug("Project '%s': initialized analyzer: %s",
+                         self.project_id,
+                         str(analyzer))
+        except AnnifException as err:
+            logger.warning(err.format_message())
 
     def _initialize_subjects(self):
         try:
@@ -88,10 +91,10 @@ def _initialize_vectorizer(self):
 
     def _initialize_backend(self):
         logger.debug("Project '%s': initializing backend", self.project_id)
-        if not self.backend:
-            logger.debug("Cannot initialize backend: does not exist")
-            return
         try:
+            if not self.backend:
+                logger.debug("Cannot initialize backend: does not exist")
+                return
             self.backend.initialize()
         except AnnifException as err:
             logger.warning(err.format_message())
@@ -121,13 +124,22 @@ def _suggest_with_backend(self, text, backend_params):
 
     @property
     def analyzer(self):
-        if self._analyzer is None and self.analyzer_spec:
-            self._analyzer = annif.analyzer.get_analyzer(self.analyzer_spec)
+        if self._analyzer is None:
+            if self.analyzer_spec:
+                self._analyzer = annif.analyzer.get_analyzer(
+                    self.analyzer_spec)
+            else:
+                raise ConfigurationException(
+                    "analyzer setting is missing (and needed by the backend)",
+                    project_id=self.project_id)
         return self._analyzer
 
     @property
     def backend(self):
         if self._backend is None:
+            if 'backend' not in self.config:
+                raise ConfigurationException(
+                    "backend setting is missing", project_id=self.project_id)
             backend_id = self.config['backend']
             try:
                 backend_class = annif.backend.get_backend(backend_id)
@@ -215,7 +227,7 @@ def dump(self):
         return {'project_id': self.project_id,
                 'name': self.name,
                 'language': self.language,
-                'backend': {'backend_id': self.config['backend']}
+                'backend': {'backend_id': self.config.get('backend')}
                 }
 
     def remove_model_data(self):
@@ -242,7 +254,11 @@ def _create_projects(projects_file, datadir, init_projects):
     config = configparser.ConfigParser()
     config.optionxform = lambda option: option
     with open(projects_file, encoding='utf-8') as projf:
-        config.read_file(projf)
+        try:
+            config.read_file(projf)
+        except (configparser.DuplicateOptionError,
+                configparser.DuplicateSectionError) as err:
+            raise ConfigurationException(err)
 
     # create AnnifProject objects from the configuration file
     projects = collections.OrderedDict()

tests/projects.cfg

@@ -44,6 +44,12 @@ language=en
 backend=tfidf
 analyzer=snowball(english)
 
+[nobackend]
+name=Dummy with no backend
+language=en
+vocab=dummy
+analyzer=snowball(english)
+
 [pav]
 name=PAV Ensemble Finnish
 language=fi

tests/projects_invalid.cfg

@@ -0,0 +1,23 @@
+# Project configuration for Annif unit tests
+
+[duplicatedproject]
+name=Dummy with no backend
+language=en
+backend=dummy
+vocab=dummy
+analyzer=snowball(english)
+
+[duplicatedproject]
+name=Dummy with no backend
+language=en
+backend=dummy
+vocab=dummy
+analyzer=snowball(english)
+
+[duplicatedvocab]
+name=Dummy with no backend
+language=en
+backend=dummy
+vocab=dummy
+vocab=dummy
+analyzer=snowball(english)

tests/test_project.py

@@ -68,13 +68,35 @@ def test_get_project_nonexistent(app):
             annif.project.get_project('nonexistent')
 
 
+def test_get_project_noanalyzer(app):
+    with app.app_context():
+        project = annif.project.get_project('noanalyzer')
+        with pytest.raises(ConfigurationException):
+            analyzer = project.analyzer
+
+
 def test_get_project_novocab(app):
     with app.app_context():
         project = annif.project.get_project('novocab')
         with pytest.raises(ConfigurationException):
             vocab = project.vocab
 
 
+def test_get_project_nobackend(app):
+    with app.app_context():
+        project = annif.project.get_project('nobackend')
+        with pytest.raises(ConfigurationException):
+            backend = project.backend
+
+
+def test_get_project_invalid_config_file(app):
+    app = annif.create_app(
+        config_name='annif.default_config.TestingInvalidProjectsConfig')
+    with app.app_context():
+        with pytest.raises(ConfigurationException):
+            project = annif.project.get_project('duplicatedvocab')
+
+
 def test_project_load_vocabulary_tfidf(app, vocabulary, testdatadir):
     with app.app_context():
         project = annif.project.get_project('tfidf-fi')