Conversation
joa-quim
left a comment
There was a problem hiding this comment.
I can't say that I fully understand the issue but I've not worked with animations for quite some time. But you do, so I trust that this is an improvement.
|
I am quite confident that it is an imporvement necessary. I am confident (although a little less so) that the modification I made to the code is the correct one. But I guess if it is wrong, we will figure it out. My tests were fine. |
|
This PR affects the output of this test ( |
|
I added # GMT_KNOWN_FAILURE in the test script that has to be updated. I will update the reference image (in DVC) in another PR. I have doubts on how to do it. |
|
What are those error with the tests? May I merge it anyway? I would think so but I am not very sure. |
|
The tests are unrelated to changes in this PR. So please merge. |
Intro
The description about the coda phase in the events documentation mentions:
Also, at the end of the -M section, it says:
This implies that symbols without a defined duration (infinite) shouldn’t reach the coda phase. However, currently, this is what is happening (as seen in animation #8607 and animation 08).
This also means that I have to define the symbol properties via the coda phase modifiers (-M+c). In fact, I must include these modifiers to make the symbols visible (e.g., like -Mt+c0), because the default behavior is for them to become invisible.
Proposed Fix
I made a fix to address this issue. The problem was (I believe) that only events with finite duration could reach the normal phase. So, in this PR, I remove that condition. I tested it on my PC (with the same script from the previous link) and got this animation, which looks fine to me:
Tutorial_1_Test.mp4
I think this solution is better because now the symbols at the end retain the properties defined by -Sc and -G (without the -M+c modification). This means the code could be simplified:
I also try the original animation made by Paul (WED-A_Vid_1.sh) and I got the same animation.