jenkins-bot merged this change.

View Change

[bugfix] Fix UnboundLocalError in ProofreadPage._ocr_callback

- use for statement instead of while
- return if loop is processed completely with 5 retries
  and do not continue the method because response is not fetched
- move sleep statement out of exception handling; probably other
  exceptions can also have retries in future

Bug: T243644
Change-Id: Id3601ea4678a3a0ab158a942644d792e46c7cea5
---
M pywikibot/proofreadpage.py
1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/pywikibot/proofreadpage.py b/pywikibot/proofreadpage.py
index 5ca3dfd..662cba0 100644
--- a/pywikibot/proofreadpage.py
+++ b/pywikibot/proofreadpage.py
@@ -599,17 +599,14 @@
                 (self._OCR_METHODS, ocr_tool))
 
         # wrong link fail with Exceptions
-        retry = 0
-        while retry < 5:
+        for retry in range(5, 30, 5):
             pywikibot.debug('{0}: get URI {1!r}'.format(ocr_tool, cmd_uri),
                             _logger)
             try:
                 response = http.fetch(cmd_uri)
             except requests.exceptions.ReadTimeout as e:
-                retry += 1
+                timeout = e
                 pywikibot.warning('ReadTimeout %s: %s' % (cmd_uri, e))
-                pywikibot.warning('retrying in %s seconds ...' % (retry * 5))
-                time.sleep(retry * 5)
             except Exception as e:
                 pywikibot.error('"%s": %s' % (cmd_uri, e))
                 return (True, e)
@@ -618,6 +615,11 @@
                                 _logger)
                 break
 
+            pywikibot.warning('retrying in {} seconds ...'.format(retry))
+            time.sleep(retry)
+        else:
+            return True, timeout
+
         if 400 <= response.status < 600:
             return (True, 'Http response status {0}'.format(response.status))
 

To view, visit change 567213. To unsubscribe, or for help writing mail filters, visit settings.