Devmodified

pattern/vector/svm/liblinear.py

@@ -31,15 +31,15 @@
 	if sys.platform == 'win32':
 		liblinear = CDLL(path.join(dirname, r'..\windows\liblinear.dll'))
 	else:
-		liblinear = CDLL(path.join(dirname, '../liblinear.so.3'))
+		liblinear = CDLL(path.join(dirname, 'macos/liblinear-2.20/liblinear.so.3'))
 except:
 # For unix the prefix 'lib' is not considered.
 	if find_library('linear'):
 		liblinear = CDLL(find_library('linear'))
 	elif find_library('liblinear'):
 		liblinear = CDLL(find_library('liblinear'))
 	else:
-		raise Exception('LIBLINEAR library not found.')
+		libsvm = CDLL(path.join(path.dirname(__file__), 'ubuntu/liblinear-2.20/liblinear.so.3'))
 
 L2R_LR = 0
 L2R_L2LOSS_SVC_DUAL = 1

pattern/vector/svm/libsvm.py

@@ -24,15 +24,17 @@
 	if sys.platform == 'win32':
 		libsvm = CDLL(path.join(dirname, r'..\windows\libsvm.dll'))
 	else:
-		libsvm = CDLL(path.join(dirname, '../libsvm.so.2'))
+		libsvm = CDLL(path.join(dirname, 'macos/libsvm-3.22/libsvm.so.2'))
+
 except:
 # For unix the prefix 'lib' is not considered.
 	if find_library('svm'):
 		libsvm = CDLL(find_library('svm'))
 	elif find_library('libsvm'):
 		libsvm = CDLL(find_library('libsvm'))
 	else:
-		raise Exception('LIBSVM library not found.')
+		libsvm = CDLL(path.join(path.dirname(__file__), 'ubuntu/libsvm-3.22/libsvm.so.2'))
+
 
 C_SVC = 0
 NU_SVC = 1

pattern/web/__init__.py

@@ -1729,6 +1729,12 @@ def search(self, query, type=SEARCH, start=1, count=10, sort=RELEVANCY, size=Non
             - count: maximum 100.
             There is a limit of 150+ queries per 15 minutes.
         """
+
+        def f(v):
+            v = v.get('extended_tweet', {}).get('full_text', v.get('full_text', v.get('text', '')))
+            return v
+
+
         if type != SEARCH:
             raise SearchEngineTypeError
         if not query or count < 1 or (isinstance(start, (int, float)) and start < 1):
@@ -1748,6 +1754,7 @@ def search(self, query, type=SEARCH, start=1, count=10, sort=RELEVANCY, size=Non
         url.query = {
                "q": query,
           "max_id": id,
+          'tweet_mode': 'extended',
            "count": min(count, 100)
         }
         # 2) Restrict location with geo=(latitude, longitude, radius).
@@ -1778,7 +1785,7 @@ def search(self, query, type=SEARCH, start=1, count=10, sort=RELEVANCY, size=Non
             r = Result(url=None)
             r.id = self.format(x.get("id_str"))
             r.url = self.format(TWITTER_STATUS % (x.get("user", {}).get("screen_name"), x.get("id_str")))
-            r.text = self.format(x.get("text"))
+            r.text = self.format(f(x))
             r.date = self.format(x.get("created_at"))
             r.author = self.format(x.get("user", {}).get("screen_name"))
             r.language = self.format(x.get("metadata", {}).get("iso_language_code"))
@@ -1789,7 +1796,7 @@ def search(self, query, type=SEARCH, start=1, count=10, sort=RELEVANCY, size=Non
             if rt:
                 comment = re.search(r"^(.*? )RT", r.text)
                 comment = comment.group(1) if comment else ""
-                r.text = self.format("RT @%s: %s" % (rt["user"]["screen_name"], rt["text"]))
+                r.text = self.format("RT @%s: %s" % (rt["user"]["screen_name"], f(rt)))
             results.append(r)
         # Twitter.search(start=id, count=10) takes a tweet.id,
         # and returns 10 results that are older than this id.
@@ -1879,15 +1886,20 @@ def __init__(self, socket, delimiter="\n", format=lambda s: s, **kwargs):
         self.format = format
 
     def parse(self, data):
+
         """ TwitterStream.queue will populate with Result objects as
             TwitterStream.update() is called iteratively.
         """
+        def f(v):
+            v = v.get('extended_tweet', {}).get('full_text', v.get('full_text', v.get('text', '')))
+            return v
+
         if data.strip():
             x = json.loads(data)
             r = Result(url=None)
             r.id = self.format(x.get("id_str"))
             r.url = self.format(TWITTER_STATUS % (x.get("user", {}).get("screen_name"), x.get("id_str")))
-            r.text = self.format(x.get("text"))
+            r.text = self.format(f(x))
             r.date = self.format(x.get("created_at"))
             r.author = self.format(x.get("user", {}).get("screen_name"))
             r.language = self.format(x.get("metadata", {}).get("iso_language_code"))
@@ -1898,7 +1910,7 @@ def parse(self, data):
             if rt:
                 comment = re.search(r"^(.*? )RT", r.text)
                 comment = comment.group(1) if comment else ""
-                r.text = self.format("RT @%s: %s" % (rt["user"]["screen_name"], rt["text"]))
+                r.text = self.format("RT @%s: %s" % (rt["user"]["screen_name"], f(rt)))
             return r