When the curl remote helper is called, e.g., as git-remote-https foo https://john@xxxxxxx it looks up remote.foo.url rather than taking the provided url, at least as far as http_init() is concerned (authentication). (It does use the provided url at later stages.) The problem is that for pushing, "git push" looks up the pushurl, which may be different, and passes that down to the remote helper, so that the remote helper should take that when provided. Note that at the init stage, the remote helper does not know what kind of transcation is going to be requested, but the caller does. Change it so that the remote helper obeys the passed url rather than trying to look it up. Signed-off-by: Michael J Gruber <git@xxxxxxxxxxxxxxxxxxxx> --- While this passes all tests and fixes the described problem, I don't know about any side effects. Also, I always feel a bit insecure about simply dropping pointers to allocated memory (remote->url[0]). Should I free() it? Neither remote-helpers nor memory management are exactly homeground for me... --- remote-curl.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/remote-curl.c b/remote-curl.c index b8cf45a..1fa396a 100644 --- a/remote-curl.c +++ b/remote-curl.c @@ -846,6 +846,7 @@ int main(int argc, const char **argv) if (argc > 2) { end_url_with_slash(&buf, argv[2]); + remote->url[0] = xstrdup(argv[2]); } else { end_url_with_slash(&buf, remote->url[0]); } -- 1.7.7.rc2.451.g15e150 -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html