[webvtt] Question about the parsing cue's timestamp

alastor0325 has just created a new issue for 
https://github.com/w3c/webvtt:

== Question about the parsing cue's timestamp ==
https://w3c.github.io/webvtt/#collect-a-webvtt-timestamp

Hi, @zcorpan,
I have some questions about timestamp parsing.

### 1. Characters showing before or after the timestamp?
In the step4 of the **collect a WebVTT timestamp**, it mentions,
> If the character indicated by position is not an ASCII digit, then 
return an error and abort these steps.

Does that mean we can't add any characters (except LF and CRLF) 
**before** start time?
However, in 
[space-chars.vtt](http://w3c-test.org/webvtt/webvtt-file-format-parsing/webvtt-file-parsing/support/space-chars.vtt),
 it allows to have characters **before** timestamp.

In addition,  I didn't see any constraint about adding the characters 
**after** end time.
However, in 
[space-chars.vtt](http://w3c-test.org/webvtt/webvtt-file-format-parsing/webvtt-file-parsing/support/space-chars.vtt),
 the last case doesn't allow to have vertical tabs **after** 
timestamp.

Could you help me to explain them :) ?

### 2. start time and end time
Do we need to check whether the start time and the end time are both 
valid in the parsing step? eg. the start time is later than the end 
time. Or we would throw error in next steps like rendering steps or 
something else?

Thanks!

Please view or discuss this issue at 
https://github.com/w3c/webvtt/issues/311 using your GitHub account

Received on Tuesday, 5 July 2016 11:17:55 UTC