| Message ID | 20220202234953.6337-1-vittorio.giovara@gmail.com |
|---|---|
| State | Accepted |
| Commit | c1228315c97acbe7a6220b4a5f282bd93edbeb42 |
| Headers | show |
| Series | [FFmpeg-devel,1/2] http: Improve handling of Content-Range with Transfer-Encoding:chunked | expand |
| Context | Check | Description |
|---|---|---|
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | success | Make fate finished |
| andriy/make_ppc | success | Make finished |
| andriy/make_fate_ppc | success | Make fate finished |
On Wed, Feb 2, 2022 at 3:50 PM Vittorio Giovara <vittorio.giovara@gmail.com> wrote: > From: Justin Ruggles <justin.ruggles@gmail.com> > > When Transfer-Encoding:chunked is used, the client must ignore a > Content-Length header, if present. However, it should not ignore a > Content-Range header, which also includes the full size of the > entity. > --- > libavformat/http.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/libavformat/http.c b/libavformat/http.c > index 8f39d11a88..c89f8a5517 100644 > --- a/libavformat/http.c > +++ b/libavformat/http.c > @@ -132,6 +132,7 @@ typedef struct HTTPContext { > int64_t expires; > char *new_location; > AVDictionary *redirect_cache; > + uint64_t filesize_from_content_range; > } HTTPContext; > > #define OFFSET(x) offsetof(HTTPContext, x) > @@ -839,7 +840,7 @@ static void parse_content_range(URLContext *h, const > char *p) > p += 6; > s->off = strtoull(p, NULL, 10); > if ((slash = strchr(p, '/')) && strlen(slash) > 0) > - s->filesize = strtoull(slash + 1, NULL, 10); > + s->filesize_from_content_range = strtoull(slash + 1, NULL, > 10); > The size part of the range header is optional, and can be '*' as well. See also https://patchwork.ffmpeg.org/project/ffmpeg/patch/20211005233244.37582-1-ffmpeg@tmm1.net/ > } > if (s->seekable == -1 && (!s->is_akamai || s->filesize != 2147483647)) > h->is_streamed = 0; /* we _can_ in fact seek */ > @@ -1341,6 +1342,7 @@ static int http_read_header(URLContext *h) > av_freep(&s->new_location); > s->expires = 0; > s->chunksize = UINT64_MAX; > + s->filesize_from_content_range = UINT64_MAX; > > for (;;) { > if ((err = http_get_line(s, line, sizeof(line))) < 0) > @@ -1356,6 +1358,10 @@ static int http_read_header(URLContext *h) > s->line_count++; > } > > + // filesize from Content-Range can always be used, even if using > chunked Transfer-Encoding > + if (s->filesize_from_content_range != UINT64_MAX) > + s->filesize = s->filesize_from_content_range; > + > if (s->seekable == -1 && s->is_mediagateway && s->filesize == > 2000000000) > h->is_streamed = 1; /* we can in fact _not_ seek */ > > -- > 2.34.1 > > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
On 2/3/2022 1:26 AM, Aman Karmani wrote: > The size part of the range header is optional, and can be '*' as well. > > See also > https://patchwork.ffmpeg.org/project/ffmpeg/patch/20211005233244.37582-1-ffmpeg@tmm1.net/ Isn't this an orthogonal problem to what this patch is changing? - Derek
On Thu, Feb 3, 2022 at 3:12 PM Derek Buitenhuis <derek.buitenhuis@gmail.com> wrote: > On 2/3/2022 1:26 AM, Aman Karmani wrote: > > The size part of the range header is optional, and can be '*' as well. > > > > See also > > > https://patchwork.ffmpeg.org/project/ffmpeg/patch/20211005233244.37582-1-ffmpeg@tmm1.net/ > > Isn't this an orthogonal problem to what this patch is changing? > > - Derek > _______________________________________________ > Are there any other comments/objections to the patchset? Thanks
On Wed, Feb 9, 2022 at 1:56 AM Vittorio Giovara <vittorio.giovara@gmail.com> wrote: > On Thu, Feb 3, 2022 at 3:12 PM Derek Buitenhuis < > derek.buitenhuis@gmail.com> > wrote: > > > On 2/3/2022 1:26 AM, Aman Karmani wrote: > > > The size part of the range header is optional, and can be '*' as well. > > > > > > See also > > > > > > https://patchwork.ffmpeg.org/project/ffmpeg/patch/20211005233244.37582-1-ffmpeg@tmm1.net/ > > > > Isn't this an orthogonal problem to what this patch is changing? > > > > - Derek > > _______________________________________________ > > > > Are there any other comments/objections to the patchset? > Nope. Just apply it. > Thanks > -- > Vittorio > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
On Wed, Feb 9, 2022 at 9:49 AM Paul B Mahol <onemda@gmail.com> wrote: > On Wed, Feb 9, 2022 at 1:56 AM Vittorio Giovara < > vittorio.giovara@gmail.com> > wrote: > > > On Thu, Feb 3, 2022 at 3:12 PM Derek Buitenhuis < > > derek.buitenhuis@gmail.com> > > wrote: > > > > > On 2/3/2022 1:26 AM, Aman Karmani wrote: > > > > The size part of the range header is optional, and can be '*' as > well. > > > > > > > > See also > > > > > > > > > > https://patchwork.ffmpeg.org/project/ffmpeg/patch/20211005233244.37582-1-ffmpeg@tmm1.net/ > > > > > > Isn't this an orthogonal problem to what this patch is changing? > > > > > > - Derek > > > _______________________________________________ > > > > > > > Are there any other comments/objections to the patchset? > > > > Nope. Just apply it. > Pushed o/
diff --git a/libavformat/http.c b/libavformat/http.c index 8f39d11a88..c89f8a5517 100644 --- a/libavformat/http.c +++ b/libavformat/http.c @@ -132,6 +132,7 @@ typedef struct HTTPContext { int64_t expires; char *new_location; AVDictionary *redirect_cache; + uint64_t filesize_from_content_range; } HTTPContext; #define OFFSET(x) offsetof(HTTPContext, x) @@ -839,7 +840,7 @@ static void parse_content_range(URLContext *h, const char *p) p += 6; s->off = strtoull(p, NULL, 10); if ((slash = strchr(p, '/')) && strlen(slash) > 0) - s->filesize = strtoull(slash + 1, NULL, 10); + s->filesize_from_content_range = strtoull(slash + 1, NULL, 10); } if (s->seekable == -1 && (!s->is_akamai || s->filesize != 2147483647)) h->is_streamed = 0; /* we _can_ in fact seek */ @@ -1341,6 +1342,7 @@ static int http_read_header(URLContext *h) av_freep(&s->new_location); s->expires = 0; s->chunksize = UINT64_MAX; + s->filesize_from_content_range = UINT64_MAX; for (;;) { if ((err = http_get_line(s, line, sizeof(line))) < 0) @@ -1356,6 +1358,10 @@ static int http_read_header(URLContext *h) s->line_count++; } + // filesize from Content-Range can always be used, even if using chunked Transfer-Encoding + if (s->filesize_from_content_range != UINT64_MAX) + s->filesize = s->filesize_from_content_range; + if (s->seekable == -1 && s->is_mediagateway && s->filesize == 2000000000) h->is_streamed = 1; /* we can in fact _not_ seek */
From: Justin Ruggles <justin.ruggles@gmail.com> When Transfer-Encoding:chunked is used, the client must ignore a Content-Length header, if present. However, it should not ignore a Content-Range header, which also includes the full size of the entity. --- libavformat/http.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)