<br><br><div class="gmail_quote">On Sat, Jun 20, 2009 at 2:59 AM, Vincent Ladeuil <span dir="ltr"><<a href="mailto:v.ladeuil%2Blp@free.fr">v.ladeuil+lp@free.fr</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Long story short: pycurl validates certificates, some people<br> relies on certificate validation. urllib doesn't validate<br> certificates (yet). Once certificate validation is implemented<br> for urllib (easy for 2.6), we can get rid of pycurl. For python<br> 2.4 and 2.5 though, that will mean replacing the pycurl<br> dependency by a python-https[1] dependency.<br> </blockquote><div><br>Ah. Now I get it. Thank you for explaining that.<br><br> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> The probability is very high (above 80%) that this new bug is<br> triggered by the test infrastructure and doesn't influence normal<br> bzr usage (that's why <a href="https://bugs.launchpad.net/bzr/+bug/383920" target="_blank">https://bugs.launchpad.net/bzr/+bug/383920</a><br> still haven't reached the top of my stack).<br> </blockquote><div><br>Does this mean that the affected tests get KnownIssue status? I realize each community has a different social standard for what is "safely ignored" and I think you can tell my perspective is more user-centric than developer-centric.<br> </div><div> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Maritza> I figure if I can frame a clear report that is easy<br> Maritza> to reproduce then maybe that's the best I can do for<br> Maritza> now.<br> <br> Thanks for working on it.<br> </blockquote><div><br><br>No problem. The best days are the ones when I help others figure stuff out that I can't or don't do.<br><br>-M<br><br></div></div><br>