Logging fixups

diff --git a/tweet/jobrunner.py b/tweet/jobrunner.py
index 5a52fc44c44a9a7fb866be48321de79b714e5f7e..e4dee94eba2fcaa86536b340de5cc640dafdefac 100644 (file)
--- a/tweet/jobrunner.py
+++ b/tweet/jobrunner.py
@@ -28,9 +28,17 @@ class JobHandler:
             )
             for filename in self.config.excel_names() ]
 
-    def log(self, s):
+    def warning(self, s):
+        logger.warning(s)
+        self.job.add_log("WARNING {}: {}".format(datetime.datetime.strptime(datetime.datetime.now(), "%c"), s))
+
+    def info(self, s):
         logger.info(s)
-        self.job.add_log(s)
+        self.job.add_log("INFO {}: {}".format(datetime.datetime.strptime(datetime.datetime.now(), "%c"), s))
+
+    def debug(self, s):
+        logger.debug(s)
+        self.job.add_log("DEBUG {}: {}".format(datetime.datetime.strptime(datetime.datetime.now(), "%c"), s))
 
     def tweet(self, query_name, status):
         self.job.num_tweets += 1
@@ -74,7 +82,6 @@ def _get_rate_status(twitter):
 
 
 def _fetch(query_name, terms, max_pages, page_size, handler):
-    handler.log("_fetch()")
     # TODO use global twitter object?
     twitter = Twython(settings.TWEET_OAUTH_CONSUMER_KEY,
             settings.TWEET_OAUTH_CONSUMER_SECRET, 
@@ -87,13 +94,12 @@ def _fetch(query_name, terms, max_pages, page_size, handler):
     max_id = -1
     previous_last_id = float('inf')
 
-    handler.log("max_pages: {}".format(max_pages))
     for k in range(max_pages):
         while remaining<10:
             seconds = int(rate_status["reset"] - time.time())
-            handler.log("Twitter rate limiting, need to wait {} seconds...".format(seconds))
+            handler.info("Twitter rate limiting, need to wait {} seconds...".format(seconds))
             handler.status(Job.STATUS_WAITING, seconds)
-            for i in range(min(60, seconds)):
+            for i in range(min(120, seconds)):
                 handler.status(Job.STATUS_WAITING, seconds-i)
                 time.sleep(1)
             rate_status = _get_rate_status(twitter)
@@ -108,33 +114,33 @@ def _fetch(query_name, terms, max_pages, page_size, handler):
         try:
             results = twitter.search(**query)
         except TwythonError as e:
-            handler.log("search exception, error code = {}: {}".format(e.error_code, e))
+            handler.warning("search exception, error code = {}: {}".format(e.error_code, e))
             handler.status(Job.STATUS_FAILED)
             # FIXME
             break
 
         remaining = int(twitter.get_lastfunction_header('x-rate-limit-remaining'))
-        handler.log("remaining: {}".format(remaining))
+        handler.debug("remaining: {}".format(remaining))
 
         max_id = results["search_metadata"]["max_id"]
-        handler.log("Number of results: {}".format(results["search_metadata"]["count"]))
+        handler.debug("Number of results: {}".format(results["search_metadata"]["count"]))
         last_id = float('inf')
         for status in results["statuses"]:
             handler.tweet(query_name, status)
             if status["id"]<last_id:
                 last_id = status["id"]
 
-        handler.log("last_id: {}; max_id: {}".format(last_id, max_id))
+        handler.debug("last_id: {}; max_id: {}".format(last_id, max_id))
         max_id = last_id
 
         time.sleep(1)
 
         if max_id>=previous_last_id:
-            handler.log("max_id not descending, quitting")
+            handler.info("max_id not descending, quitting")
             break
         previous_last_id = max_id
         if k >= max_pages-1:
-            handler.log("Warning, more than {} pages of results, ignoring the rest!!!".format(max_pages));
+            handler.warning("Warning, more than {} pages of results, ignoring the rest!!!".format(max_pages));
 
     #results = twitter.search(q='nlalert', result_type='recent') # since_id=, max_id=,count=,
     #for result in results:
@@ -142,13 +148,12 @@ def _fetch(query_name, terms, max_pages, page_size, handler):
 
 # Runs in a thread pool thread
 def _execute_job(job, handler):
-    logger.debug("_execute_job")
     try:
         config = job.get_config()
         handler.status(Job.STATUS_RUNNING)
         print(str(config.query_names()))
         for query_name in config.query_names():
-            handler.log("Running query {}".format(query_name));
+            handler.info("Running query {}".format(query_name));
             _fetch(query_name, config.query(query_name), config.max_pages, config.page_size, handler)
         handler.status(Job.STATUS_DONE)
     except Exception as e:

diff --git a/tweet/streamrunner.py b/tweet/streamrunner.py
index 8063a87c69412ac13eef1632f1291af52fc60a5d..f87de6cb362c906d46018ad0b566fe7915e3de50 100644 (file)
--- a/tweet/streamrunner.py
+++ b/tweet/streamrunner.py
@@ -23,7 +23,7 @@ class Streamer(TwythonStreamer):
     count = 0
     def on_success(self, data):
         if 'text' in data:
-            #logger.info("Stream text: "+data['text'])
+            logger.info("Stream text: "+data['text'])
             #logger.info("Stream tweet: {}".format(data))
             tweet = Tweet.from_status(data)
             tweet.conforms_to_terms = check_match(data)

diff --git a/tweet/utils.py b/tweet/utils.py
index 7d7fb7b8b461adb5fee2b95d2a4b41b4101aa427..33eb1d79f006762766a938eda0cb1c17edda5178 100644 (file)
--- a/tweet/utils.py
+++ b/tweet/utils.py
@@ -45,7 +45,7 @@ class ExcelExporter:
 
     def add_tweet(self, tweet):
         if tweet.id in self.ids:
-            logger.info("Skipping tweet {}, already in".format(tweet.id))
+            logger.debug("Skipping tweet {}, already in".format(tweet.id))
             return
         self.ids.add(tweet.id)