Merge lp:~jelmer/bzr/default-urllib into lp:bzr

[Jelmer Vernooij]

Change the default HTTPS implementation from pycurl to urllib.

Change the default HTTPS implementation from pycurl to urllib.

[Vincent Ladeuil (vila)]

We can't do that until we get sound default values for ssl.ca_certs which is the point of bug #920455

I revert this on purpose as I think we need some feedback with 2.5 before switching the default.

Rest assured that I still have #125055 on my radar but I still think it's too early to break the setup of people relying on pycurl being the default as they rely on it to verify their certs.

I also have a gut feeling that some bugs are lurking around, may be related to proxies, so I didn't update bug #125055 asking for the default certs but consider it a master bug that we will act upon when we're ready to switch ;)

Can we agree on that and postpone this proposal ?

review: Needs Information

[Vincent Ladeuil (vila)]

To clarify further, I think the cat is out of the bag for 2.5 so we need to fix all the bugs that may be blocking #125055 before 2.5.0

[Jelmer Vernooij (jelmer)]

Am 23/01/12 18:22, schrieb Vincent Ladeuil:
> To clarify further, I think the cat is out of the bag for 2.5 so we need to fix all the bugs that may be blocking #125055 before 2.5.0
Note that this is against lp:bzr (intentionally), and not against 2.5. I
don't think not having the right default paths should be a blocker for that.

Cheers,

Jelmer

[Vincent Ladeuil (vila)]

Oh ! Silly me ! We *want* it in ttrunk if only to make sure there are no remaining bugs in 2.5 !!!

review: Approve

[Jelmer Vernooij (jelmer)]

sent to pqm by email

=== modified file 'bzrlib/transport/__init__.py'
--- bzrlib/transport/__init__.py	2012-01-23 15:26:31 +0000
+++ bzrlib/transport/__init__.py	2012-01-23 15:26:32 +0000
@@ -1782,15 +1782,15 @@
                  help="Read-only access of branches exported on the web.")
 register_transport_proto('https://',
             help="Read-only access of branches exported on the web using SSL.")
-# The default http implementation is urllib, but https uses pycurl if available
+# The default http implementation is urllib
 register_lazy_transport('http://', 'bzrlib.transport.http._pycurl',
                         'PyCurlTransport')
 register_lazy_transport('http://', 'bzrlib.transport.http._urllib',
                         'HttpTransport_urllib')
+register_lazy_transport('https://', 'bzrlib.transport.http._pycurl',
+                        'PyCurlTransport')
 register_lazy_transport('https://', 'bzrlib.transport.http._urllib',
                         'HttpTransport_urllib')
-register_lazy_transport('https://', 'bzrlib.transport.http._pycurl',
-                        'PyCurlTransport')
 
 register_transport_proto('ftp://', help="Access using passive FTP.")
 register_lazy_transport('ftp://', 'bzrlib.transport.ftp', 'FtpTransport')

=== modified file 'doc/en/release-notes/bzr-2.6.txt'
--- doc/en/release-notes/bzr-2.6.txt	2012-01-16 13:49:34 +0000
+++ doc/en/release-notes/bzr-2.6.txt	2012-01-23 15:26:32 +0000
@@ -26,6 +26,11 @@
 .. Improvements to existing commands, especially improved performance 
    or memory usage, or better results.
 
+ * Access to HTTPS URLs now uses the urrllib implementation by default.
+   For the old pycurl-based implementation, specify ``https+pycurl://`` as
+   the URL scheme when accessing a HTTPS location.
+   (Jelmer Vernooij, #125055)
+
 Bug Fixes
 *********