On 19/01/2021 03:07, James Almer wrote:
On 1/18/2021 7:44 PM, Mark Thompson wrote:+#define BSF_ELEMENT_OPTIONS_PIR(name, help, field, unit_name) \You could reuse name instead of also passing unit_name. The only bsf where it differs is h264_metadata with display_orientation vs disp_or, and changing the latter into the former should be fine (Is it even user facing?).
I don't think it can have any user effect, though technically they can inspect it. So, sure, that's simpler.
+ { name, help, OFFSET(field), AV_OPT_TYPE_INT, \ + { .i64 = BSF_ELEMENT_PASS }, \ + BSF_ELEMENT_PASS, BSF_ELEMENT_REMOVE, FLAGS, unit_name }, \This depends on FLAGS being defined before this macro is invoked. It's probably safer and/or more robust to pass the flags as an argument.
All BSFs with options use this idiom, so it felt safe ("grep -r
AV_OPT_FLAG_BSF_PARAM libavcodec/").
I guess it is slightly surprising to have that extra implicit argument, though,
so no objection to making it explicit.
+ { "pass", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_PASS }, .flags = FLAGS, .unit = unit_name }, \ + { "insert", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_INSERT }, .flags = FLAGS, .unit = unit_name }, \ + { "remove", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_REMOVE }, .flags = FLAGS, .unit = unit_name } + +#define BSF_ELEMENT_OPTIONS_PIRE(name, help, field, unit_name) \ + { name, help, OFFSET(field), AV_OPT_TYPE_INT, \ + { .i64 = BSF_ELEMENT_PASS }, \ + BSF_ELEMENT_PASS, BSF_ELEMENT_EXTRACT, FLAGS, unit_name }, \ + { "pass", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_PASS }, .flags = FLAGS, .unit = unit_name }, \ + { "insert", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_INSERT }, .flags = FLAGS, .unit = unit_name }, \ + { "remove", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_REMOVE }, .flags = FLAGS, .unit = unit_name }, \ + { "extract", NULL, 0, AV_OPT_TYPE_CONST, \ + { .i64 = BSF_ELEMENT_EXTRACT }, .flags = FLAGS, .unit = unit_name } \
Thanks, - Mark _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
