And for the data, I really want the option to be able to dump as much data as possible, though that definitely should not be the default.
1 Like
Idk whether these are enough. (Surely not because multiple-worlds ones haven’t been implemented.)
Update: Settings are merged into a menu button.
I think a new tab or a popup would be a good idea.
I mean for a csv file, wouldn’t it be better to just dump everything and whoever is dealing with the file can then just ignore some columns if they want. I think it would be a bit of a wasted effort to make a GUI for selecting which columns to export.
I put a sample output folder (zipped up) in programming channel. There are literally too many things to export, and many of them can’t be easily put into a spreadsheet.
Such as: Map → Patches (This can be row header) → Basic stats (This can be column header), Species in patch (This is dynamic so I wrote some code to generate it) → Species stats (I have nowhere to put this); → Patch history data (So does this)
I see, so the problem is that the data would fill multiple tables?
Then maybe it makes sense to select the dataset to export. Or maybe we’d need an ods file exporter library (I don’t want the game to export in a proprietary format).
That would probably be better.
Could you put some of them into json files? Not everything has to be csv. Also, for many of these, the format you chose is not great. If you agree it might be worth it for me to go through and change the way stuff is outputted. For example, per patch history could easily be two tables with fewer columns.
One other thing I also thought of is that what if the export was done by selecting a folder in which the game would then dump various separate .csv files with specific names? That way the problem of dumping data that doesn’t make sense to be in the same table can all be output at once with minimal effort. I imagine that arrangement would also be pretty good for example using python to process the files in some way as such a script would just need just a path to the folder to find the various data files in.
It’s currently dumped in a ton of different tables, as you would see in that zip.
On second thought, maybe I should reformat the tables to show, at least what I think, is best. I would struggle to describe it, but I see a few ways they could be much improved.