FS#67448 - [mlt] compress lumas.

Attached to Project: Arch Linux
Opened by bartus (bartus) - Friday, 31 July 2020, 11:51 GMT
Last edited by Antonio Rojas (arojas) - Friday, 31 July 2020, 14:29 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Antonio Rojas (arojas)
Architecture All
Severity Medium
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

After switching to cmake `mlt` balloons out substantially (from `3MiB` to `260MiB`) by generating uncompressed lumas for for `luma_transition_filter.

[There's no flag in `cmake` to turn off lumas build or to build compressed lumas](https://github.com/mltframework/mlt/commit/b9732158e5c8446d5083adcc5586c6938f8b75a1#diff-cb48c84f4672eeab94c4f8908b424727) (those options exists only in `configure/make` build)
I propose using [a bit of code of code form `/mlt/src/module/lumas/create_lumas.sh` to compress lumas from `pgm`s to `png`s](https://github.com/mltframework/mlt/blob/master/src/modules/lumas/create_lumas#L68-L73).

This way we go down with `mlt` size form `~260MiB` to `8.7MiB`

btw. I've putted this as PR on GitHub, but looks like those don't count.
https://github.com/archlinux/svntogit-packages/pull/2
This task depends upon

Closed by  Antonio Rojas (arojas)
Friday, 31 July 2020, 14:29 GMT
Reason for closing:  Fixed
Additional comments about closing:  mlt 6.22.0-2

Loading...