timedhosts: defer 1st update until a 1MB+ download. BZ 851178

python-urlgrabber.spec

@@ -3,7 +3,7 @@
 Summary: A high-level cross-protocol url-grabber
 Name: python-urlgrabber
 Version: 3.9.1
-Release: 18%{?dist}
+Release: 19%{?dist}
 Source0: urlgrabber-%{version}.tar.gz
 Patch1: urlgrabber-HEAD.patch
 
@@ -44,6 +44,9 @@ rm -rf $RPM_BUILD_ROOT
 %attr(0755,root,root) %{_libexecdir}/urlgrabber-ext-down
 
 %changelog
+* Mon Aug 27 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-19
+- timedhosts: defer 1st update until a 1MB+ download.  BZ 851178
+
 * Wed Aug 22 2012 Zdeněk Pavlas <zpavlas@redhat.com> - 3.9.1-18
 - Update to latest HEAD, lots of enhancements.
 

urlgrabber-HEAD.patch

@@ -233,7 +233,7 @@ index 3e5f3b7..8eeaeda 100644
      return (fb,lb)
  
 diff --git a/urlgrabber/grabber.py b/urlgrabber/grabber.py
-index e090e90..ffd5a10 100644
+index e090e90..daa478d 100644
 --- a/urlgrabber/grabber.py
 +++ b/urlgrabber/grabber.py
 @@ -49,11 +49,26 @@ GENERAL ARGUMENTS (kwargs)
@@ -1225,7 +1225,7 @@ index e090e90..ffd5a10 100644
  
  #####################################################################
  # DEPRECATED FUNCTIONS
-@@ -1621,6 +1924,458 @@ def retrygrab(url, filename=None, copy_local=0, close_connection=0,
+@@ -1621,6 +1924,460 @@ def retrygrab(url, filename=None, copy_local=0, close_connection=0,
  
          
  #####################################################################
@@ -1637,17 +1637,19 @@ index e090e90..ffd5a10 100644
 +
 +    @staticmethod
 +    def update(url, dl_size, dl_time, ug_err, baseurl=None):
-+        _TH.load()
-+
 +        # Use hostname from URL.  If it's a file:// URL, use baseurl.
 +        # If no baseurl, do not update timedhosts.
 +        host = urlparse.urlsplit(url).netloc.split('@')[-1] or baseurl
 +        if not host: return
 +
++        _TH.load()
 +        speed, fail, ts = _TH.hosts.get(host) or (0, 0, 0)
 +        now = time.time()
 +
 +        if ug_err is None:
++            # defer first update if the file was small.  BZ 851178.
++            if not ts and dl_size < 1e6: return
++
 +            # k1: the older, the less useful
 +            # k2: <500ms readings are less reliable
 +            # speeds vary, use 10:1 smoothing