With todays release of CubeL4 the size of the package really surprised me. CubeMX downloaded 1790MBs worth of data (800 of that is music, 2 copies of each file). That is insane. I managed to ignore the 4-600MBs of past package sizes but this is crazy. Most of that package is example projects of multiple different boards. I'm not sure how others work but we don't have ST-made dev boards, only F4 discovery panels for "historic" reasons, using them as ST-Link programmers.
I would love to see _most_ of the example stuff removed from the actual patch releases that are downloaded through CubeMX with an option (even if opt-out) to include or exclude example code. Yes, I find myself browsing through some of the middleware-specific code sometimes, but since I don't have any of the dev boards, most of the times the 15-20 different board-specific projects are exactly the same for me. This results in wasted bandwidth and wasted disk space on my end, wasted server time on ST's end.
Currently I have 15 different Cube packages downloaded (as a side note, all but one are "major" releases and one F7 package is a patch release that is <100MB, which is 8MB of source code and 80MB of user manual, lol...). That's a lot of wasted space. I get that the trend is "if there is space/memory/cpu time, we can waste it" but that does not mean it should be this way.
I would like to hear other people's thoughts on this, if only I'm overly sensitive and these package sizes are fine, and if not, is there something that ST could do going forward that would improve this situation. Thanks.