| Message ID | 20190505185109.20957-2-cus@passwd.hu |
|---|---|
| State | Accepted |
| Commit | c61d16cf3b815388c076c2a35b5c91e84b45d7be |
| Headers | show |
Am So., 5. Mai 2019 um 20:51 Uhr schrieb Marton Balint <cus@passwd.hu>: > > This reverts commit dd06f022b07438d650c82255dff16908ba04244a. > > Fixes ticket #7871 and reopens ticket #7816. I'll send an alternative patch in a moment. Carl Eugen
On Sun, 5 May 2019, Carl Eugen Hoyos wrote: > Am So., 5. Mai 2019 um 20:51 Uhr schrieb Marton Balint <cus@passwd.hu>: >> >> This reverts commit dd06f022b07438d650c82255dff16908ba04244a. >> >> Fixes ticket #7871 and reopens ticket #7816. > > I'll send an alternative patch in a moment. Ping for this, I still think the revert is the best we can do here. Regards, Marton
On Tue, May 14, 2019 at 10:34 PM Marton Balint <cus@passwd.hu> wrote: > > > > On Sun, 5 May 2019, Carl Eugen Hoyos wrote: > > > Am So., 5. Mai 2019 um 20:51 Uhr schrieb Marton Balint <cus@passwd.hu>: > >> > >> This reverts commit dd06f022b07438d650c82255dff16908ba04244a. > >> > >> Fixes ticket #7871 and reopens ticket #7816. > > > > I'll send an alternative patch in a moment. > > Ping for this, I still think the revert is the best we can do here. > I agree. Slashes are flat out not allowed in that part of the URI and will always result in ambigious parsing. - Hendrik
On Tue, 14 May 2019, Hendrik Leppkes wrote: > On Tue, May 14, 2019 at 10:34 PM Marton Balint <cus@passwd.hu> wrote: >> >> >> >> On Sun, 5 May 2019, Carl Eugen Hoyos wrote: >> >> > Am So., 5. Mai 2019 um 20:51 Uhr schrieb Marton Balint <cus@passwd.hu>: >> >> >> >> This reverts commit dd06f022b07438d650c82255dff16908ba04244a. >> >> >> >> Fixes ticket #7871 and reopens ticket #7816. >> > >> > I'll send an alternative patch in a moment. >> >> Ping for this, I still think the revert is the best we can do here. >> > > I agree. Slashes are flat out not allowed in that part of the URI and > will always result in ambigious parsing. Will apply soon. Regards, Marton
> Am 18.05.2019 um 16:48 schrieb Marton Balint <cus@passwd.hu>: > > > >>> On Tue, 14 May 2019, Hendrik Leppkes wrote: >>> >>> On Tue, May 14, 2019 at 10:34 PM Marton Balint <cus@passwd.hu> wrote: >>> >>> >>> >>>> On Sun, 5 May 2019, Carl Eugen Hoyos wrote: >>>> >>>>> Am So., 5. Mai 2019 um 20:51 Uhr schrieb Marton Balint <cus@passwd.hu>: >>>>> >>>>> This reverts commit dd06f022b07438d650c82255dff16908ba04244a. >>>>> >>>>> Fixes ticket #7871 and reopens ticket #7816. >>>> >>>> I'll send an alternative patch in a moment. >>> >>> Ping for this, I still think the revert is the best we can do here. >> >> I agree. Slashes are flat out not allowed in that part of the URI and >> will always result in ambigious parsing. > > Will apply soon. Please do if you don’t see a better solution. Carl Eugen
On Sat, 18 May 2019, Carl Eugen Hoyos wrote: > > >> Am 18.05.2019 um 16:48 schrieb Marton Balint <cus@passwd.hu>: >> >> >> >>>> On Tue, 14 May 2019, Hendrik Leppkes wrote: >>>> >>>> On Tue, May 14, 2019 at 10:34 PM Marton Balint <cus@passwd.hu> wrote: >>>> >>>> >>>> >>>>> On Sun, 5 May 2019, Carl Eugen Hoyos wrote: >>>>> >>>>>> Am So., 5. Mai 2019 um 20:51 Uhr schrieb Marton Balint <cus@passwd.hu>: >>>>>> >>>>>> This reverts commit dd06f022b07438d650c82255dff16908ba04244a. >>>>>> >>>>>> Fixes ticket #7871 and reopens ticket #7816. >>>>> >>>>> I'll send an alternative patch in a moment. >>>> >>>> Ping for this, I still think the revert is the best we can do here. >>> >>> I agree. Slashes are flat out not allowed in that part of the URI and >>> will always result in ambigious parsing. >> >> Will apply soon. > > Please do if you don’t see a better solution. Thanks, applied. Regards, Marton
diff --git a/libavformat/utils.c b/libavformat/utils.c index a63d71b0f4..6ef94239a4 100644 --- a/libavformat/utils.c +++ b/libavformat/utils.c @@ -4730,7 +4730,7 @@ void av_url_split(char *proto, int proto_size, char *hostname, int hostname_size, int *port_ptr, char *path, int path_size, const char *url) { - const char *p, *ls, *ls2, *ls3, *at, *at2, *col, *brk; + const char *p, *ls, *ls2, *at, *at2, *col, *brk; if (port_ptr) *port_ptr = -1; @@ -4760,9 +4760,6 @@ void av_url_split(char *proto, int proto_size, /* separate path from hostname */ ls = strchr(p, '/'); ls2 = strchr(p, '?'); - ls3 = strchr(p, '@'); - if (ls3 && ls3 > ls && (!ls2 || ls2 > ls3)) - ls = strchr(ls3, '/'); if (!ls) ls = ls2; else if (ls && ls2)
This reverts commit dd06f022b07438d650c82255dff16908ba04244a. Fixes ticket #7871 and reopens ticket #7816. The introduced regression caused URL's with @ in them to be parsed incorrectly which is a bigger issue then not being able to specify the slash character as a password. I think there are better ways to fix the original issue, like being able to specify HTTP username and password as a protocol option, or adding a protocol option to percent-decode the URL first. Signed-off-by: Marton Balint <cus@passwd.hu> --- libavformat/utils.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-)