Rkg script #23
Rkg script #23
Conversation
| assert AButton == 0 or AButton == 1, \ | ||
| f"A button input has value `{AButton}` at line {csv_line}" | ||
|
|
||
| assert BButton == 0 or BButton == 1, \ | ||
| f"B button input has value `{BButton}` at line {csv_line}" | ||
|
|
||
| assert ItemButton == 0 or ItemButton == 1, \ | ||
| f"Item button input has value `{ItemButton}` at line {csv_line}" |
There was a problem hiding this comment.
"<button> button input on line {csv_line} is out of range! Expected 0 - 1, got `{<button>}`"
| assert -7 <= YDirection <= 7, \ | ||
| f"Y direction input has value `{YDirection}` at line {csv_line}" | ||
|
|
||
| assert -7 <= XDirection <= 7, \ | ||
| f"X direction input has value `{XDirection}` at line {csv_line}" |
There was a problem hiding this comment.
We (regrettably) shouldn't assume the CSV is 0-centered.
"<direction> direction input has inconsistent centering at line {csv_line}"
"<direction> direction input on line {csv_line} is out of range! Expected {range} - {range + 14}, got `{<direction>}`"
There was a problem hiding this comment.
In what situation would this not be the case?
How do we know then how to parse the csv if this is not always the case? A parameter that specifies the format?
There was a problem hiding this comment.
There are still people who use 7-centered.
The most we can do is infer the center by checking unshared boundaries (> 7, < 0). If that fails, we may want to default to whichever number has the highest frequency, since neutral inputs are most likely the more common of the two. Alternatively, we can ask the user via input what center the script is.
| assert 0 <= TrickInput <= 4, \ | ||
| f"Trick input has value `{TrickInput}` at line {csv_line}" |
There was a problem hiding this comment.
"Trick input on line {csv_line} is out of range! Expected 0 - 4, got `{TrickInput}`"
|
|
||
| def get_dummy_data(): | ||
| """Generate dummy metadata to put in the header of the rkg file""" | ||
| return MetaData(0, 0, 0, True) |
There was a problem hiding this comment.
Dummy metadata is invalid - this will generate a ghost with Mario on Standard Kart S.
| if len(inputs) > 0x2774: | ||
| print("Too many ghost inputs for a standard RKG file") | ||
| else: | ||
| # Pad input data until it is of length 0x2774 | ||
| inputs.extend([0x00] * (0x2774 - len(inputs))) |
There was a problem hiding this comment.
Good idea for general use, though you may want to add a boolean for this project since we can rewrite the ghost parsing to extend the size.
|
With regards to metadata, there are two approaches that I think would be good. If no arguments are passed to the script, we can query with If there are arguments passed into the script, we assume that the arguments are IDs and use them directly. On a separate note, due to the Mario/Standard Kart S thing, we may want to check that the character/vehicle combination is valid. |
#5 Initial commit, the following still needs to happen:
Personally I'm not a fan of getting the metadata through a prompt because then you would have to fill in the id's of the character and vehicle which people don't know by heart.
How will this script be used exactly? If it is used by a discord bot or something then imo it makes more sense to just pass all the metadata as arguments to the script directly.