support multiple instances of same document, extend url from api url

pandora_client/__init__.py

@@ -812,23 +812,33 @@ class Client(object):
                    if f['id'] not in available and f['extension'] in DOCUMENT_FORMATS]
         return missing
 
+    def find_document(self, oshash):
+        r = self.api.findDocuments({
+            'keys': ['id'],
+            'query': {
+                'conditions': [
+                    {'key': 'oshash', 'value': oshash, 'operator': '=='}
+                ],
+                'operator': '&'
+            }
+        })
+        if r['data']['items']:
+            return r['data']['items'][0]['id']
+        return None
+
     def _add_document(self, f, item=None):
         if f.split('.')[-1] not in DOCUMENT_FORMATS:
             return False
-        url = '%supload/document/' % self._config['url']
-        r = self.api.upload_chunks(url, f, {
-            'filename': os.path.basename(f)
-        })
-        if r and item:
-            oshash = ox.oshash(f)
-            r = self.api.findDocuments({
-                "keys": ['id'],
-                "query": {
-                    "conditions": [{"key": 'oshash', "value": oshash, "operator": '=='}],
-                    "operator": '&'
-                }
+        oshash = ox.oshash(f)
+        did = self.find_document(oshash)
+        if not did:
+            url = '%supload/document/' % self._config['url']
+            r = self.api.upload_chunks(url, f, {
+                'filename': os.path.basename(f)
             })
-            did = r['data']['items'][0]['id']
+            if r and item:
+                did = self.find_document(oshash)
+        if item:
             r = self.api.addDocument({
                 'id': did,
                 'item': item
@@ -1061,12 +1071,16 @@ class API(ox.API):
 
         print(filename)
         hide_cursor()
-        result_url = data.get('url')
-        if 'uploadUrl' in data:
-            uploadUrl = data['uploadUrl']
-            if uploadUrl.startswith('/'):
+
+        def full_url(path):
+            if path.startswith('/'):
                 u = urlparse(url)
-                uploadUrl = '%s://%s%s' % (u.scheme, u.netloc, uploadUrl)
+                path = '%s://%s%s' % (u.scheme, u.netloc, path)
+            return path
+
+        result_url = full_url(data.get('url'))
+        if 'uploadUrl' in data:
+            uploadUrl = full_url(data['uploadUrl'])
             f = open(filename, 'rb')
             fsize = os.stat(filename).st_size
             done = 0