Changeset 466

Timestamp:

27/11/07 01:09:26 (2 years ago)

Author:

athomas

Message:

pyndexter: Documents are now referenced by a user-defined key rather than a
URI. It was a needless restriction. Fixed hyperestraier unit tests up a bit.

Location:

pyndexter/branches/simplification

Files:

• pyndexter/__init__.py (modified) (10 diffs)
• pyndexter/indexers/_hyperestraier.py (modified) (2 diffs)
• setup.py (modified) (1 diff)

pyndexter/branches/simplification/pyndexter/__init__.py

@@ -59 +59 @@
     Defines three useful attributes:
 
-    ``uri``
-        The unique URI for this document.
+    ``key``
+        The unique key for this document.
 
     ``quality``
@@ -66 +66 @@
     """
 
-    def __init__(self, uri, quality=1.0, attributes=None, text=None,
+    def __init__(self, key, quality=1.0, attributes=None, text=None,
                  texts=None):
-        self.uri = URI(uri)
+        self.key = key
         self.quality = quality
         self._attributes = {}
@@ -121 +121 @@
 
     def __hash__(self):
-        return hash(self.uri)
+        return hash(self.key)
 
 
@@ -142 +142 @@
         raise NotImplementedError
 
-    def discard(self, uri):
+    def discard(self, key):
         """ Discard a document. """
         raise NotImplementedError
@@ -156 +156 @@
         raise NotImplementedError
 
-    def fetch(self, uri):
+    def fetch(self, key):
         """Attempt to fetch indexer representation of the document.
 
@@ -162 +162 @@
         and 1.0, representing the quality of the document in comparison to the
         original."""
-        raise DocumentNotFound(uri)
+        raise DocumentNotFound(key)
 
     def replace(self, document):
         """Replace a document in the index. Default is to `discard()` and
         `index()`."""
-        self.discard(document.uri)
+        self.discard(document.key)
         self.index(document)
 
@@ -227 +227 @@
         A Document object corresponding to the Hit.
 
-    ``uri``
-        The URI of the hit.
+    ``key``
+        The key of the hit.
     """
 
-    def __init__(self, uri, document=None, score=0.0, attributes=None):
-        self.uri = uri
+    def __init__(self, key, document=None, score=0.0, attributes=None):
+        self.key = key
         if callable(document):
-            self.get_document = lambda: document(uri)
+            self.get_document = lambda: document(key)
         else:
             self._document = document
@@ -268 +268 @@
 
     def __hash__(self):
-        return hash(self.uri)
+        return hash(self.key)
 
 
@@ -289 +289 @@
         return iter(self.indexer)
 
-    def discard(self, uri):
-        if isinstance(uri, Document):
-            uri = uri.uri
-        return self.indexer.discard(uri)
+    def discard(self, key):
+        if isinstance(key, Document):
+            key = key.key
+        return self.indexer.discard(key)
 
     def search(self, query):
@@ -299 +299 @@
         return self.indexer.search(query)
 
-    def fetch(self, uri):
-        return self.indexer.fetch(URI(uri))
+    def fetch(self, key):
+        return self.indexer.fetch(key)
 
 

pyndexter/branches/simplification/pyndexter/indexers/_hyperestraier.py

@@ -78 +78 @@
 
     def index(self, document):
-        uri = unicode(document.uri)
+        key = unicode(document.key)
         hdoc = hyperestraier.Document()
         for k, v in document.iteritems():
             hdoc.add_attr(u'@' + k, v)
-        hdoc.add_attr(u'@uri', uri)
+        hdoc.add_attr(u'@uri', key)
         for line in document.texts:
             hdoc.add_text(line)
         if not self.db.put_doc(hdoc):
-            raise errors.IndexerError('Failed to index %s' % document.uri)
+            raise errors.IndexerError('Failed to index %s' % document.key)
 
-    def discard(self, uri):
-        uri = unicode(uri)
-        if not self.db.out_doc_by_uri(uri):
-            raise errors.DocumentNotFound(uri)
+    def discard(self, key):
+        key = unicode(key)
+        try:
+            self.db.out_doc_by_uri(key)
+        except hyperestraier.HyperestraierError, e:
+            raise errors.DocumentNotFound(key)
 
-    def fetch(self, uri):
-        uri = unicode(uri)
-        doc = self.db.get_doc_by_uri(uri)
+    def fetch(self, key):
+        key = unicode(key)
+        doc = self.db.get_doc_by_uri(key)
         if not doc:
-            raise errors.DocumentNotFound(uri)
+            raise errors.DocumentNotFound(key)
         attributes = self._translate_attributes(doc)
-        return Document(uri, texts=doc.dtexts, quality=0.99,
+        return Document(key, texts=doc.dtexts, quality=0.99,
                         attributes=attributes)
 
@@ -152 +154 @@
     # Internal methods
     def _translate(self, doc):
-        return Hit(uri=doc.uri, document=self.indexer.fetch,
+        return Hit(key=doc.uri, document=self.indexer.fetch,
                    attributes=self.indexer._translate_attributes(doc))

pyndexter/branches/simplification/setup.py

@@ -28 +28 @@
     ],
     extras_require={
-        'hyperestraier': ['hyperestraier>=0.9.3'],
+        'hyperestraier': ['hyperestraier>=0.10.4'],
         'snowball': ['PyStemmer>=1.0.1'],
         # TODO put more indexer requirements in here