[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-wget] wget-1.12-2359 alpha version
From: |
Giuseppe Scrivano |
Subject: |
Re: [Bug-wget] wget-1.12-2359 alpha version |
Date: |
Sun, 23 May 2010 00:42:04 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.0.50 (gnu/linux) |
Ray Satiro <address@hidden> writes:
> Is there a web interface for the new repository? Somewhere we'll be able to
> click a link and download a zip of different revisions, like
> the old repository. I couldn't find it so I tried bazaar 2.1.1 for windows
> but it tells me the wget tree isn't available and I can't pull
> anything:
>
> This branch has no working tree. Last revision is 2359. Action not yet
> supported by current app_suite.
can you try this command?
"bzr branch http://bzr.savannah.gnu.org/r/wget/trunk"
You will also need git and rsync installed for the `boostrap' script.
> As for msys fixes I see you fixed the linking issue in the last revision,
> bravo :) ipv6 and ssl config tests are still broken. In windows
> if you are going do a static link test for openssl you should link with
> winsock2 (openssl now uses winsock2) and gdi as well. So this will
> work:
>
> gcc a.c -lssl -lcrypto -lws2_32 -lgdi32
>
> regardless when wget is compiled with-ssl using mingw one would expect that
> instead of ssl crypto gdi it not be linked to any of those,
> just to eay32 and ssl32 (openssl dlls).
>
> also the config ac shows lwsock32 -lws2_32 and indeed on compile both are
> linked. is there any reason for this, like maybe a trick for a
> universal binary or something? it doesn't seem logical.
>
AFAIK, -lws2_32 is needed to build successfully, is there a way to don't
use it?
> ipv6 fix is a little different and this goes to an argument that's come up
> before where if you are going legacy there apparently is a way
> to make a universal binary so that wget built with ipv6 support could?
> function on windows systems with winsock 1.1. well ok but really is
> wget set up for this and is it possible and is there any reason for it? i
> don't get that. regardless config is testing for getaddrinfo
> which is only winsock2, and according to the tcpip header file I have for it
> actually to be declared _WIN32_WINNT >= 0x0501. my suggestion
> is continue to test for getaddrinfo but fix for mingw
>
I will take a look at these issues ASAP.
I think we can declare _WIN32_WINNT >= 0x0501 without big problems.
Thanks,
Giuseppe