On Sun, May 13, 2018, at 8:12 AM, Carl Zwanzig wrote: > > Um, when someone _reads_ the text, they're parsing it.
The ticket seems obvious to me that it is referring to machine parsing. > Maybe it's not intended to be _machine_ parsed, but I've had no trouble with > breaking down the text with some code. It sounds like the better argument > is why the informational (not error) output intended for a human to see is > being sent to stderr. IMHO, the answer in the ticket is specious. The comments in the ticket are valid, and you should not be machine parsing the stderr output to get input information. The output is not guaranteed to be accurate (for example it shows a shorthand "29.97 fps" instead of 30000/1001). Processing it requires additional tools or code, and the resulting processed output may be easily broken with a variety of assorted inputs. Use ffprobe instead: it is a proper tool for the job. _______________________________________________ ffmpeg-user mailing list [email protected] http://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
