This is a work in progress.
By default it makes a capture every 5 minutes, if the video is shorter than the capture interval, it refuses to continue. You can change the default behaviour with the command line options -i and -n.
-i (also --interval) changes de default capture interval.
If e.g. you want to take a snap every 3 minutes you would use -i 3m.
See below for more details on the time format accepted.
-n (also --numcaps) sets the number of captures
regardless of the video duration, so it is the simplest way to use
vcs on short videos.
If e.g. you want to get four captures, use -n 4.
Options accepting time as an argument support a format which will be described here. As it is probably easier understood by example, here are some valid examples: 1h30m10s, 2h, 30m, 2h20.
Explained in more detail, the format follows the pattern:
XhYmZs (no spaces allowed) with Xh,
Ym, Z and s being optional and meaning:
Xh: X is the number of hours, followed by a literal
h, hence 2h stands for 2 hours.
Ym: Y is the number of minutes, followed by a literal
m, hence 30m stans for 30 minutes, 5m for
5 minutes and 90m for 90 minutes.
Zs: Z is the number of seconds, optionally
followed by a literal s, hence both 30s and
30 stand for 30 seconds.
You don't have to write time in clock-correct form, i.e. 1h90m90s is accepted and the same as 2h31m30.
In short, the extended mode adds an extra set of (smaller, but more) captures.
The extra captures are ¼ the size of the standard captures (half width and half height) and, by default4x times the standard number of captures (in future versions this number will be rounded to be a multiple of the number of columns).
An example will probably be of help: extended mode sampe.
You can influence the number of extended captures, changing the multiplier applied, by adding a number after the -e (or --extended) option (e.g. -e4 or --extended=4). This number is multiplied by the number of standard captures to get the number of extended captures.
The syntax for this option is more strict than for the rest, when setting a multiplier.
If you use -e the number must be directly after the -e (e.g.: -e5 is ok, -e 5 is not).
If you use --extended the number must be directly after the --extended but separated by a = sign (e.g.: --extended=5 is ok, --extended 5 is not.
Use -H (or --height) to change the height. Width is changed accordingly, it can't be manually set, change to aspect ratio to change the width freely.
Note also that height is always the reference, so when you change aspect ratio height will remain the same and width will continue to be the one being derived.
Use -a (or --aspect) to change the aspect ratio. A fraction (e.g. 4/3) or a floating point number (e.g. 1.33) is accepted (fractions are more precise).
Use -A (or --autoaspect) to let vcs try to guess the aspect ratio. Guesses are only made on well known resolutions (i.e. VCD and DVD family of resolutions), note that for cases in which both a 4:3 and a 16:9 resolution are possible 4:3 is chosen. If it can't be guessed the original aspect ratio will be used.
Use -c (or --columns) to change the number of columns in which the captures are arranged (2 by default).
Looks better if the number of captures can be divided by the number of columns.
Note that with one column the heading often gets messed if there's too much information to show and/or the captures are not wide enough to accomodate all of it.
Use -f (or --from) and -t (or --to) to set the starting and ending points of
captures. Accepts the same time format as -i.
Captures explicitly set with -S or -l aren't restricted by these limits.
Use -S (or --stamp) to manually include a specific image. Accepts the same time format as -i. The count of captures when using -n doesn't include these.
Future versions will allow more fine grained control, right now you can only manually add images at a specific second.
Use -l (or --highlight) to manually emphasize single frames.
Highlights are much like manual captures, but they're placed atop the contact sheet in a light golden background. They're put in the same order as they appear on the video and not on the commandline.
