Trim off fat

[corpserot]

spawned out of #241

git archive --format=zip -o sbz-e51efb16d1ee04edcf0ca8b5672c5c0bb8bf903e.zip e51efb16d1ee04edcf0ca8b5672c5c0bb8bf903e # from slimming branch
git archive --format=zip -o sbz-15de3c069b5b68686b7643c1179dea82ff1abd57.zip 15de3c069b5b68686b7643c1179dea82ff1abd57 # from master branch

get-item 'sbz-*.zip' | select-object Name, @{Name='Size MB'; Expression={([math]::Round($_.Length / 1MB, 2))}}
# Name                                             Size MB
# ----                                             -------
# sbz-15de3c069b5b68686b7643c1179dea82ff1abd57.zip   26.58
# sbz-e51efb16d1ee04edcf0ca8b5672c5c0bb8bf903e.zip   23.65

the optimize_audio.py performs lossy optimization using ffmpeg on "high quality" audio files (read the source to see what i mean by that). it seems that all of the audio is optimized to some degree, at least according to the configs i've set. because it is lossy, it will destroy data so don't run it on the same files more than once. you can avoid that by simply running it on new audio files introduced by a commit.

python devtools\optimize_audio.py 1234abcd
# optimizes all audio files involved in commit 1234abcd

the script itself skips over non-"high quality" audio files so that helps

[corpserot]

if anyone has any ideas what else to trim lemme know.

[TheEt1234]

(zander also wants specifically the contentDB zip file to be under 30mb, i believe)

[TheEt1234]

LGTM 👍

[TheEt1234]

(ok, i'll do the same for stubes too)