Merge edbeae8 into dc37e4f

annif/backend/dummy.py

@@ -10,4 +10,5 @@ class DummyBackend(backend.AnnifBackend):
 
     def _analyze(self, text, project, params):
         score = float(params.get('score', 1.0))
-        return [AnalysisHit('http://example.org/dummy', 'dummy', score)]
+        return [AnalysisHit(uri='http://example.org/dummy',
+                            label='dummy', score=score)]

annif/backend/http.py

@@ -13,6 +13,6 @@ class HTTPBackend(backend.AnnifBackend):
     def _analyze(self, text, project, params):
         data = {'text': text, 'project': params['project']}
         req = requests.post(params['endpoint'], data=data)
-        return [AnalysisHit(h['uri'], h['label'], h['score'])
+        return [AnalysisHit(uri=h['uri'], label=h['label'], score=h['score'])
                 for h in req.json()
                 if h['score'] > 0.0]

annif/backend/tfidf.py

@@ -62,10 +62,9 @@ def _merge_chunk_results(self, chunk_results, project):
             subject = project.subjects[subject_id]
             results.append(
                 AnalysisHit(
-                    subject[0],
-                    subject[1],
-                    score /
-                    len(chunk_results)))
+                    uri=subject[0],
+                    label=subject[1],
+                    score=score / len(chunk_results)))
         return results
 
     def _analyze(self, text, project, params):

annif/corpus/subject.py

@@ -1,16 +1,13 @@
 """A directory of files as a subject corpus"""
 
 
+import collections
 import glob
 import os.path
 import re
 
 
-class Subject:
-    def __init__(self, uri, label, text):
-        self.uri = uri
-        self.label = label
-        self.text = text
+Subject = collections.namedtuple('Subject', 'uri label text')
 
 
 class SubjectDirectory:
@@ -25,7 +22,7 @@ def __iter__(self):
             with open(filename) as subjfile:
                 uri, label = subjfile.readline().strip().split(' ', 1)
                 text = ' '.join(subjfile.readlines())
-                yield Subject(uri, label, text)
+                yield Subject(uri=uri, label=label, text=text)
 
 
 class SubjectIndex:
@@ -64,6 +61,6 @@ def file_as_corpus(path):
                 for line in subjfile:
                     uri, label = line.strip().split(None, 1)
                     uri = uri[1:-1]
-                    yield Subject(uri, label, None)
+                    yield Subject(uri=uri, label=label, text=None)
 
         return cls(file_as_corpus(path))

annif/hit.py

@@ -1,14 +1,6 @@
 """Class representing a single hit from analysis."""
 
+import collections
 
-class AnalysisHit:
-    """A single hit resulting from analysis."""
 
-    def __init__(self, uri, label, score):
-        self.uri = uri
-        self.label = label
-        self.score = score
-
-    def dump(self):
-        """return this object as a dict"""
-        return {'uri': self.uri, 'label': self.label, 'score': self.score}
+AnalysisHit = collections.namedtuple('AnalysisHit', 'uri label score')

annif/project.py

@@ -73,8 +73,7 @@ def _merge_hits(cls, hits_by_uri):
         merged_hits = []
         for score_hits in hits_by_uri.values():
             total = sum([sh[0] for sh in score_hits])
-            hit = annif.hit.AnalysisHit(
-                score_hits[0][1].uri, score_hits[0][1].label, total)
+            hit = score_hits[0][1]._replace(score=total)
             merged_hits.append(hit)
         return merged_hits
 

annif/rest.py

@@ -40,4 +40,4 @@ def analyze(project_id, text, limit, threshold):
         return "Project '{}' not found".format(project_id), 404
 
     hits = project.analyze(text, limit, threshold)
-    return [hit.dump() for hit in hits]
+    return [hit._asdict() for hit in hits]