From d0d3c8b2a66f6244b38ee76d0cb8da72d406031b Mon Sep 17 00:00:00 2001
From: JustAnotherArchivist <JustAnotherArchivist@users.noreply.github.com>
Date: Mon, 1 Oct 2018 03:24:29 +0200
Subject: [PATCH] Better log output for temporary failures (fixes #2)

---
 snscrape/base.py            | 29 +++++++++++++++++++++++++----
 snscrape/modules/twitter.py |  5 ++---
 2 files changed, 27 insertions(+), 7 deletions(-)

diff --git a/snscrape/base.py b/snscrape/base.py
index a4a6f21..443efc2 100644
--- a/snscrape/base.py
+++ b/snscrape/base.py
@@ -59,11 +59,32 @@ class Scraper:
 				logger.debug(f'... with data: {data!r}')
 			try:
 				r = self._session.send(req, timeout = timeout)
-				if responseOkCallback is None or responseOkCallback(r):
-					logger.debug(f'{req.url} retrieved successfully')
-					return r
 			except requests.exceptions.RequestException as exc:
-				logger.error(f'Error retrieving {url}: {exc!r}')
+				if attempt < self._retries:
+					retrying = ', retrying'
+					level = logging.WARNING
+				else:
+					retrying = ''
+					level = logging.ERROR
+				logger.log(level, f'Error retrieving {req.url}: {exc!r}{retrying}')
+			else:
+				if responseOkCallback is not None:
+					success, msg = responseOkCallback(r)
+				else:
+					success, msg = (True, None)
+				msg = f': {msg}' if msg else ''
+
+				if success:
+					logger.debug(f'{req.url} retrieved successfully{msg}')
+					return r
+				else:
+					if attempt < self._retries:
+						retrying = ', retrying'
+						level = logging.WARNING
+					else:
+						retrying = ''
+						level = logging.ERROR
+					logger.log(level, f'Error retrieving {req.url}{msg}{retrying}')
 			if attempt < self._retries:
 				sleepTime = 1.0 * 2**attempt # exponential backoff: sleep 1 second after first attempt, 2 after second, 4 after third, etc.
 				logger.info(f'Waiting {sleepTime:.0f} seconds')
diff --git a/snscrape/modules/twitter.py b/snscrape/modules/twitter.py
index 8a8436c..30e64c2 100644
--- a/snscrape/modules/twitter.py
+++ b/snscrape/modules/twitter.py
@@ -27,9 +27,8 @@ class TwitterSearchScraper(snscrape.base.Scraper):
 
 	def _check_json_callback(self, r):
 		if r.headers['content-type'] != 'application/json;charset=utf-8':
-			logger.error(f'Content type of {r.url} is not JSON')
-			return False
-		return True
+			return False, f'content type is not JSON'
+		return True, None
 
 	def get_items(self):
 		headers = {'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36'}