| Message ID | 20230825053027.6610-2-lq@chinaffmpeg.org |
|---|---|
| State | New |
| Headers | show |
| Series | [FFmpeg-devel,v2,1/2] avformat/rtmppkt: add ff_amf_write_array for write array strings | expand |
| Context | Check | Description |
|---|---|---|
| yinshiyou/make_loongarch64 | success | Make finished |
| yinshiyou/make_fate_loongarch64 | success | Make fate finished |
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | success | Make fate finished |
On Fri, 25 Aug 2023, Steven Liu wrote: > Add option named rtmp_enhanced_codec, > it would support hvc1,av01,vp09 now, > the fourcc is using Array of strings. > > Signed-off-by: Steven Liu <lq@chinaffmpeg.org> > --- > doc/protocols.texi | 6 ++++++ > libavformat/rtmpproto.c | 35 +++++++++++++++++++++++++++++++++++ > 2 files changed, 41 insertions(+) > > diff --git a/doc/protocols.texi b/doc/protocols.texi > index b3fad55591..bd2b25e502 100644 > --- a/doc/protocols.texi > +++ b/doc/protocols.texi > @@ -896,6 +896,12 @@ be named, by prefixing the type with 'N' and specifying the name before > the value (i.e. @code{NB:myFlag:1}). This option may be used multiple > times to construct arbitrary AMF sequences. > > +@item rtmp_enhanced_codecs > +Specify the list of codecs the client advertises to support in an > +enhanced RTMP stream. This option should set a string like @code{hvc1,av01,vp09} > +for multiple codecs, or @code{hvc1} for only one codec, > +set codec fourcc into fourCcLive property into Connect Command Message, > + > @item rtmp_flashver > Version of the Flash plugin used to run the SWF player. The default > is LNX 9,0,124,2. (When publishing, the default is FMLE/3.0 (compatible; > diff --git a/libavformat/rtmpproto.c b/libavformat/rtmpproto.c > index f0ef223f05..ea25454362 100644 > --- a/libavformat/rtmpproto.c > +++ b/libavformat/rtmpproto.c > @@ -127,6 +127,7 @@ typedef struct RTMPContext { > int nb_streamid; ///< The next stream id to return on createStream calls > double duration; ///< Duration of the stream in seconds as returned by the server (only valid if non-zero) > int tcp_nodelay; ///< Use TCP_NODELAY to disable Nagle's algorithm if set to 1 > + char *enhanced_codecs; ///< codec list in enhanced rtmp > char username[50]; > char password[50]; > char auth_params[500]; > @@ -336,6 +337,39 @@ static int gen_connect(URLContext *s, RTMPContext *rt) > ff_amf_write_field_name(&p, "app"); > ff_amf_write_string2(&p, rt->app, rt->auth_params); > > + if (rt->enhanced_codecs) { > + uint32_t list_len = 0; > + char *fourcc_data = rt->enhanced_codecs; > + int fourcc_str_len = strlen(fourcc_data); > + > + // check the string, fourcc + ',' + ... + end fourcc correct length should be (4+1)*n+4 > + if ((fourcc_str_len + 1) % 5 != 0) > + return AVERROR(EINVAL); If we trigger this error, we definitely should tell the user what was wrong. 'Malformed rtmp_enhanched_codecs, should be of the form "fourcc[,fourcc..]"' or something like that? (Would it be better to use "frcc" or just "hvc1" or something else there, to show that it needs to be exactly a four char string for each element?) > + > + list_len = (fourcc_str_len + 1) / 5; > + // write the fourCcList field name > + ff_amf_write_field_name(&p, "fourCcList"); > + > + // write the fourcc array length > + ff_amf_write_array_start(&p, list_len); > + > + while(fourcc_data - rt->enhanced_codecs < fourcc_str_len) { > + unsigned char fourcc[5]; > + switch (AV_RN32(fourcc_data)) { > + case MKTAG('h', 'v', 'c', '1'): This feels a bit convoluted to use AV_RN32 + MKTAG on data that is already just a string - just using strncmp(fourcc_data, "hvc1", 4) would work just as well. That doesn't fit quite as neatly into a switch, but would be a more unwieldy if statement though. > + case MKTAG('a', 'v', '0', '1'): > + case MKTAG('v', 'p', '0', '9'): > + av_strlcpy(fourcc, fourcc_data, sizeof(fourcc)); > + ff_amf_write_string(&p, fourcc); > + break; > + default: When this fails, the user definitely needs to know what was wrong, so we should print the string that didn't match the expectation. E.g. ("unsupported codec fourcc, %.*s", 4, fourcc_data) or something like that? // Martin
Martin Storsjö <martin@martin.st> 于2023年8月25日周五 17:04写道： > > On Fri, 25 Aug 2023, Steven Liu wrote: > > > Add option named rtmp_enhanced_codec, > > it would support hvc1,av01,vp09 now, > > the fourcc is using Array of strings. > > > > Signed-off-by: Steven Liu <lq@chinaffmpeg.org> > > --- > > doc/protocols.texi | 6 ++++++ > > libavformat/rtmpproto.c | 35 +++++++++++++++++++++++++++++++++++ > > 2 files changed, 41 insertions(+) > > > > diff --git a/doc/protocols.texi b/doc/protocols.texi > > index b3fad55591..bd2b25e502 100644 > > --- a/doc/protocols.texi > > +++ b/doc/protocols.texi > > @@ -896,6 +896,12 @@ be named, by prefixing the type with 'N' and specifying the name before > > the value (i.e. @code{NB:myFlag:1}). This option may be used multiple > > times to construct arbitrary AMF sequences. > > > > +@item rtmp_enhanced_codecs > > +Specify the list of codecs the client advertises to support in an > > +enhanced RTMP stream. This option should set a string like @code{hvc1,av01,vp09} > > +for multiple codecs, or @code{hvc1} for only one codec, > > +set codec fourcc into fourCcLive property into Connect Command Message, > > + > > @item rtmp_flashver > > Version of the Flash plugin used to run the SWF player. The default > > is LNX 9,0,124,2. (When publishing, the default is FMLE/3.0 (compatible; > > diff --git a/libavformat/rtmpproto.c b/libavformat/rtmpproto.c > > index f0ef223f05..ea25454362 100644 > > --- a/libavformat/rtmpproto.c > > +++ b/libavformat/rtmpproto.c > > @@ -127,6 +127,7 @@ typedef struct RTMPContext { > > int nb_streamid; ///< The next stream id to return on createStream calls > > double duration; ///< Duration of the stream in seconds as returned by the server (only valid if non-zero) > > int tcp_nodelay; ///< Use TCP_NODELAY to disable Nagle's algorithm if set to 1 > > + char *enhanced_codecs; ///< codec list in enhanced rtmp > > char username[50]; > > char password[50]; > > char auth_params[500]; > > @@ -336,6 +337,39 @@ static int gen_connect(URLContext *s, RTMPContext *rt) > > ff_amf_write_field_name(&p, "app"); > > ff_amf_write_string2(&p, rt->app, rt->auth_params); > > > > + if (rt->enhanced_codecs) { > > + uint32_t list_len = 0; > > + char *fourcc_data = rt->enhanced_codecs; > > + int fourcc_str_len = strlen(fourcc_data); > > + > > + // check the string, fourcc + ',' + ... + end fourcc correct length should be (4+1)*n+4 > > + if ((fourcc_str_len + 1) % 5 != 0) > > + return AVERROR(EINVAL); > > If we trigger this error, we definitely should tell the user what was > wrong. 'Malformed rtmp_enhanched_codecs, should be of the form > "fourcc[,fourcc..]"' or something like that? (Would it be better to use > "frcc" or just "hvc1" or something else there, to show that it needs to be > exactly a four char string for each element?) Agreed > > > + > > + list_len = (fourcc_str_len + 1) / 5; > > + // write the fourCcList field name > > + ff_amf_write_field_name(&p, "fourCcList"); > > + > > + // write the fourcc array length > > + ff_amf_write_array_start(&p, list_len); > > + > > + while(fourcc_data - rt->enhanced_codecs < fourcc_str_len) { > > + unsigned char fourcc[5]; > > + switch (AV_RN32(fourcc_data)) { > > + case MKTAG('h', 'v', 'c', '1'): > > This feels a bit convoluted to use AV_RN32 + MKTAG on data that is already > just a string - just using strncmp(fourcc_data, "hvc1", 4) would work just I want use strncmp at first, but i think strncmp will slower than switch uint32_t value, so i chose use uint32_t. or should i use strncmp? > as well. That doesn't fit quite as neatly into a switch, but would be a > more unwieldy if statement though. > > > + case MKTAG('a', 'v', '0', '1'): > > + case MKTAG('v', 'p', '0', '9'): > > + av_strlcpy(fourcc, fourcc_data, sizeof(fourcc)); > > + ff_amf_write_string(&p, fourcc); > > + break; > > + default: > > When this fails, the user definitely needs to know what was wrong, so we > should print the string that didn't match the expectation. E.g. > ("unsupported codec fourcc, %.*s", 4, fourcc_data) or something like that? good suggestion. > > // Martin > > _______________________________________________ > 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 Fri, 25 Aug 2023, Steven Liu wrote: >> > + list_len = (fourcc_str_len + 1) / 5; >> > + // write the fourCcList field name >> > + ff_amf_write_field_name(&p, "fourCcList"); >> > + >> > + // write the fourcc array length >> > + ff_amf_write_array_start(&p, list_len); >> > + >> > + while(fourcc_data - rt->enhanced_codecs < fourcc_str_len) { >> > + unsigned char fourcc[5]; >> > + switch (AV_RN32(fourcc_data)) { >> > + case MKTAG('h', 'v', 'c', '1'): >> >> This feels a bit convoluted to use AV_RN32 + MKTAG on data that is already >> just a string - just using strncmp(fourcc_data, "hvc1", 4) would work just > I want use strncmp at first, but i think strncmp will slower than > switch uint32_t value, so i chose use uint32_t. or should i use > strncmp? Yes, strncmp will be slower than switching on an uint32_t - but this isn't really performance sensitive code. We do this once only, on startup, on a handful of fourccs. So I believe code clarity is more important than absolute performance here. // Martin
Martin Storsjö <martin@martin.st> 于2023年8月25日周五 17:38写道： > > On Fri, 25 Aug 2023, Steven Liu wrote: > > >> > + list_len = (fourcc_str_len + 1) / 5; > >> > + // write the fourCcList field name > >> > + ff_amf_write_field_name(&p, "fourCcList"); > >> > + > >> > + // write the fourcc array length > >> > + ff_amf_write_array_start(&p, list_len); > >> > + > >> > + while(fourcc_data - rt->enhanced_codecs < fourcc_str_len) { > >> > + unsigned char fourcc[5]; > >> > + switch (AV_RN32(fourcc_data)) { > >> > + case MKTAG('h', 'v', 'c', '1'): > >> > >> This feels a bit convoluted to use AV_RN32 + MKTAG on data that is already > >> just a string - just using strncmp(fourcc_data, "hvc1", 4) would work just > > I want use strncmp at first, but i think strncmp will slower than > > switch uint32_t value, so i chose use uint32_t. or should i use > > strncmp? > > Yes, strncmp will be slower than switching on an uint32_t - but this isn't > really performance sensitive code. We do this once only, on startup, on a > handful of fourccs. So I believe code clarity is more important than > absolute performance here. Okay, i use strncasecmp next version patch. But i must told you, I think MKTAG is clarity than use if+strncmp to me, Because there looks like a fourcc codec list align code block. > > // Martin > > _______________________________________________ > 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 Fri, 25 Aug 2023, Steven Liu wrote: > Martin Storsjö <martin@martin.st> 于2023年8月25日周五 17:38写道： >> >> On Fri, 25 Aug 2023, Steven Liu wrote: >> >> >> > + list_len = (fourcc_str_len + 1) / 5; >> >> > + // write the fourCcList field name >> >> > + ff_amf_write_field_name(&p, "fourCcList"); >> >> > + >> >> > + // write the fourcc array length >> >> > + ff_amf_write_array_start(&p, list_len); >> >> > + >> >> > + while(fourcc_data - rt->enhanced_codecs < fourcc_str_len) { >> >> > + unsigned char fourcc[5]; >> >> > + switch (AV_RN32(fourcc_data)) { >> >> > + case MKTAG('h', 'v', 'c', '1'): >> >> >> >> This feels a bit convoluted to use AV_RN32 + MKTAG on data that is already >> >> just a string - just using strncmp(fourcc_data, "hvc1", 4) would work just >> > I want use strncmp at first, but i think strncmp will slower than >> > switch uint32_t value, so i chose use uint32_t. or should i use >> > strncmp? >> >> Yes, strncmp will be slower than switching on an uint32_t - but this isn't >> really performance sensitive code. We do this once only, on startup, on a >> handful of fourccs. So I believe code clarity is more important than >> absolute performance here. > Okay, i use strncasecmp next version patch. I don't see a need for this to be case insensitive? > But i must told you, I think MKTAG is clarity than use if+strncmp to me, > Because there looks like a fourcc codec list align code block. If you prefer MKTAG for clarity, then I guess the current solution is fine here for me as well - it's not a hard opinion from me. Let's see if someone else wants to comment on it first maybe. // Martin
diff --git a/doc/protocols.texi b/doc/protocols.texi index b3fad55591..bd2b25e502 100644 --- a/doc/protocols.texi +++ b/doc/protocols.texi @@ -896,6 +896,12 @@ be named, by prefixing the type with 'N' and specifying the name before the value (i.e. @code{NB:myFlag:1}). This option may be used multiple times to construct arbitrary AMF sequences. +@item rtmp_enhanced_codecs +Specify the list of codecs the client advertises to support in an +enhanced RTMP stream. This option should set a string like @code{hvc1,av01,vp09} +for multiple codecs, or @code{hvc1} for only one codec, +set codec fourcc into fourCcLive property into Connect Command Message, + @item rtmp_flashver Version of the Flash plugin used to run the SWF player. The default is LNX 9,0,124,2. (When publishing, the default is FMLE/3.0 (compatible; diff --git a/libavformat/rtmpproto.c b/libavformat/rtmpproto.c index f0ef223f05..ea25454362 100644 --- a/libavformat/rtmpproto.c +++ b/libavformat/rtmpproto.c @@ -127,6 +127,7 @@ typedef struct RTMPContext { int nb_streamid; ///< The next stream id to return on createStream calls double duration; ///< Duration of the stream in seconds as returned by the server (only valid if non-zero) int tcp_nodelay; ///< Use TCP_NODELAY to disable Nagle's algorithm if set to 1 + char *enhanced_codecs; ///< codec list in enhanced rtmp char username[50]; char password[50]; char auth_params[500]; @@ -336,6 +337,39 @@ static int gen_connect(URLContext *s, RTMPContext *rt) ff_amf_write_field_name(&p, "app"); ff_amf_write_string2(&p, rt->app, rt->auth_params); + if (rt->enhanced_codecs) { + uint32_t list_len = 0; + char *fourcc_data = rt->enhanced_codecs; + int fourcc_str_len = strlen(fourcc_data); + + // check the string, fourcc + ',' + ... + end fourcc correct length should be (4+1)*n+4 + if ((fourcc_str_len + 1) % 5 != 0) + return AVERROR(EINVAL); + + list_len = (fourcc_str_len + 1) / 5; + // write the fourCcList field name + ff_amf_write_field_name(&p, "fourCcList"); + + // write the fourcc array length + ff_amf_write_array_start(&p, list_len); + + while(fourcc_data - rt->enhanced_codecs < fourcc_str_len) { + unsigned char fourcc[5]; + switch (AV_RN32(fourcc_data)) { + case MKTAG('h', 'v', 'c', '1'): + case MKTAG('a', 'v', '0', '1'): + case MKTAG('v', 'p', '0', '9'): + av_strlcpy(fourcc, fourcc_data, sizeof(fourcc)); + ff_amf_write_string(&p, fourcc); + break; + default: + return AVERROR(EINVAL); + } + + fourcc_data += 5; + } + } + if (!rt->is_input) { ff_amf_write_field_name(&p, "type"); ff_amf_write_string(&p, "nonprivate"); @@ -3104,6 +3138,7 @@ static const AVOption rtmp_options[] = { {"rtmp_conn", "Append arbitrary AMF data to the Connect message", OFFSET(conn), AV_OPT_TYPE_STRING, {.str = NULL }, 0, 0, DEC|ENC}, {"rtmp_flashver", "Version of the Flash plugin used to run the SWF player.", OFFSET(flashver), AV_OPT_TYPE_STRING, {.str = NULL }, 0, 0, DEC|ENC}, {"rtmp_flush_interval", "Number of packets flushed in the same request (RTMPT only).", OFFSET(flush_interval), AV_OPT_TYPE_INT, {.i64 = 10}, 0, INT_MAX, ENC}, + {"rtmp_enhanced_codecs", "Specify the codec(s) to use in an enhanced rtmp live stream", OFFSET(enhanced_codecs), AV_OPT_TYPE_STRING, {.str = NULL }, 0, 0, ENC}, {"rtmp_live", "Specify that the media is a live stream.", OFFSET(live), AV_OPT_TYPE_INT, {.i64 = -2}, INT_MIN, INT_MAX, DEC, "rtmp_live"}, {"any", "both", 0, AV_OPT_TYPE_CONST, {.i64 = -2}, 0, 0, DEC, "rtmp_live"}, {"live", "live stream", 0, AV_OPT_TYPE_CONST, {.i64 = -1}, 0, 0, DEC, "rtmp_live"},
Add option named rtmp_enhanced_codec, it would support hvc1,av01,vp09 now, the fourcc is using Array of strings. Signed-off-by: Steven Liu <lq@chinaffmpeg.org> --- doc/protocols.texi | 6 ++++++ libavformat/rtmpproto.c | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 41 insertions(+)