Rev 3035: Fix 165061 by using the correct _max_readv_combine attribute in file:///home/pqm/archives/thelove/bzr/%2Btrunk/

[Canonical.com Patch Queue Manager]

At file:///home/pqm/archives/thelove/bzr/%2Btrunk/

------------------------------------------------------------
revno: 3035
revision-id:pqm at pqm.ubuntu.com-20071127101459-zcdf4aia34927kt0
parent: pqm at pqm.ubuntu.com-20071127023739-a1ajr28wi7so2up6
parent: v.ladeuil+lp at free.fr-20071127093633-vs6mdkj2pxwo18r0
committer: Canonical.com Patch Queue Manager <pqm at pqm.ubuntu.com>
branch nick: +trunk
timestamp: Tue 2007-11-27 10:14:59 +0000
message:
  Fix 165061 by using the correct _max_readv_combine attribute
modified:
  NEWS                           NEWS-20050323055033-4e00b5db738777ff
  bzrlib/transport/__init__.py   transport.py-20050711165921-4978aa7ce1285ad5
  bzrlib/transport/http/__init__.py http_transport.py-20050711212304-506c5fd1059ace96
    ------------------------------------------------------------
    revno: 3034.1.1
    revision-id:v.ladeuil+lp at free.fr-20071127093633-vs6mdkj2pxwo18r0
    parent: pqm at pqm.ubuntu.com-20071127023739-a1ajr28wi7so2up6
    parent: v.ladeuil+lp at free.fr-20071127093537-gruuxzmso2r6c5pg
    committer: Vincent Ladeuil <v.ladeuil+lp at free.fr>
    branch nick: trunk
    timestamp: Tue 2007-11-27 10:36:33 +0100
    message:
      Fix 165061 by using the correct _max_readv_combine attribute
    modified:
      NEWS                           NEWS-20050323055033-4e00b5db738777ff
      bzrlib/transport/__init__.py   transport.py-20050711165921-4978aa7ce1285ad5
      bzrlib/transport/http/__init__.py http_transport.py-20050711212304-506c5fd1059ace96
    ------------------------------------------------------------
    revno: 3024.2.2
    revision-id:v.ladeuil+lp at free.fr-20071127093537-gruuxzmso2r6c5pg
    parent: v.ladeuil+lp at free.fr-20071127082600-y4hnhy03v0jdywgc
    committer: Vincent Ladeuil <v.ladeuil+lp at free.fr>
    branch nick: 165061
    timestamp: Tue 2007-11-27 10:35:37 +0100
    message:
      Update NEWS.
    modified:
      NEWS                           NEWS-20050323055033-4e00b5db738777ff
    ------------------------------------------------------------
    revno: 3024.2.1
    revision-id:v.ladeuil+lp at free.fr-20071127082600-y4hnhy03v0jdywgc
    parent: pqm at pqm.ubuntu.com-20071126080312-mc2sagjsq7u3lemo
    committer: Vincent Ladeuil <v.ladeuil+lp at free.fr>
    branch nick: 165061
    timestamp: Tue 2007-11-27 09:26:00 +0100
    message:
      Fix 165061 by using the correct _max_readv_combine attribute.
      
      * bzrlib/transport/http/__init__.py:
      (HttpTransportBase): Typo corrected. That should be some kind of
      record in the performance_gained / size(code_modified) ratio
      league.
    modified:
      bzrlib/transport/__init__.py   transport.py-20050711165921-4978aa7ce1285ad5
      bzrlib/transport/http/__init__.py http_transport.py-20050711212304-506c5fd1059ace96
=== modified file 'NEWS'
--- a/NEWS	2007-11-27 02:37:39 +0000
+++ b/NEWS	2007-11-27 09:36:33 +0000
@@ -98,6 +98,10 @@
    * Fix multiple connections during checkout --lightweight.
      (Vincent Ladeuil, #159150)
 
+   * Fix typo in limiting offsets coalescing for http, leading to
+     whole files being downloaded instead of parts.
+     (Vincent Ladeuil, #165061)
+
    * FTP server errors don't error in the error handling code.
      (Robert Collins, #161240)
 

=== modified file 'bzrlib/transport/__init__.py'
--- a/bzrlib/transport/__init__.py	2007-11-19 13:44:25 +0000
+++ b/bzrlib/transport/__init__.py	2007-11-27 08:26:00 +0000
@@ -645,7 +645,7 @@
 
         :param relpath: The path to read data from.
         :param offsets: A list of (offset, size) tuples.
-        :param adjust_for_latency: Adjust the requested offsets to accomdate
+        :param adjust_for_latency: Adjust the requested offsets to accomodate
             transport latency. This may re-order the offsets, expand them to
             grab adjacent data when there is likely a high cost to requesting
             data relative to delivering it.
@@ -661,9 +661,12 @@
         if adjust_for_latency:
             # Design note: We may wish to have different algorithms for the
             # expansion of the offsets per-transport. E.g. for local disk to
-            # use page-aligned expansion. If that is the case consider the following structure:
-            #  - a test that transport.readv uses self._offset_expander or some similar attribute, to do the expansion
-            #  - a test for each transport that it has some known-good offset expander
+            # use page-aligned expansion. If that is the case consider the
+            # following structure:
+            #  - a test that transport.readv uses self._offset_expander or some
+            #    similar attribute, to do the expansion
+            #  - a test for each transport that it has some known-good offset
+            #    expander
             #  - unit tests for each offset expander
             #  - a set of tests for the offset expander interface, giving
             #    baseline behaviour (which the current transport

=== modified file 'bzrlib/transport/http/__init__.py'
--- a/bzrlib/transport/http/__init__.py	2007-08-26 22:10:51 +0000
+++ b/bzrlib/transport/http/__init__.py	2007-11-27 08:26:00 +0000
@@ -250,7 +250,7 @@
     _bytes_to_read_before_seek = 128
     # No limit on the offset number that get combined into one, we are trying
     # to avoid downloading the whole file.
-    _max_readv_combined = 0
+    _max_readv_combine = 0
 
     def _readv(self, relpath, offsets):
         """Get parts of the file at the given relative path.