feat: some clean up and validation of multiboot header

generators/scancodes/transform_scancodes.py

@@ -0,0 +1,124 @@
+#!/usr/bin/env python3
+import functools
+import itertools
+import json
+import unicodedata
+from typing import NamedTuple, List, Optional
+import pyperclip
+
+
+class KeyCode(NamedTuple):
+    scan_code: int
+    name_lower: str
+    name_upper: str
+    ascii_lower: Optional[str]
+    ascii_upper: Optional[str]
+
+
+# scancodes retrieved from https://wiki.osdev.org/PS/2_Keyboard
+# scancodes transformed with https://www.convertjson.com/html-table-to-json.htm
+
+
+blacklist_names = {'(zero)'}
+
+special_map = {
+    '\\': 'forwardSlash',
+    '/': 'backSlash',
+    '[': 'bracketLeft',
+    ']': 'bracketRight',
+    ';': 'colonPressed',
+    "'": None,
+    '`': None,
+    ',': 'comma',
+    '.': 'dot',
+    # todo keypad *
+    '-': 'dash',
+    '=': 'equals',
+}
+
+
+def get_unknown(i):
+    return KeyCode(i, f"Unknown{i:02x}Lower", f"Unknown{i:02x}Upper", None, None)
+
+
+def not_in_blacklist(name: str) -> bool:
+    return name not in blacklist_names
+
+
+def remove_braces(name: str) -> str:
+    name = name.lstrip('(').rstrip(')')
+    if name in special_map:
+        return special_map[name]
+    return name
+
+
+def key_name_to_name(names: List[str]) -> str:
+    _names: List[str] = list(filter(lambda i: i is not None, map(remove_braces, filter(not_in_blacklist, names))))
+    _names_cap = itertools.chain([_names[0][0].lower() + _names[0][1:]],
+                                 map(lambda i: i[0].upper() + i[1:], _names[1:]))
+    return ''.join(_names_cap)
+
+
+def get_name(item: str):
+    item = item.replace('  ', ' ')
+    if len(item) == 1:
+        item = unicodedata.name(item).title()
+    if item.startswith('Latin'):
+        return 'Letter' + item.split()[-1]
+    return item.replace(' ', '_') if item else None
+
+
+def transform_set1(path):
+    codes = []
+    with open(path) as f:
+        data = json.load(f)
+
+    for ibm_number, make_break_1, _, _, lower, upper in data:
+        make_code, break_code = make_break_1.split('/')
+        if make_code.startswith('E0') or make_code.startswith('E1'):
+            # skipping multi codes for now
+            continue
+        make_code_i = int(make_code, base=16)
+        break_code_i = int(break_code, base=16)
+        if make_code_i != break_code_i ^ 0x80:
+            raise ValueError("Make/Break not equal" + ibm_number)
+        codes.append(KeyCode(
+            make_code_i,
+            get_name(lower),
+            get_name(upper),
+            lower,
+            upper,
+        ))
+
+    codes_map = {}
+
+    for code in codes:
+        if code.scan_code in codes_map:
+            raise ValueError(f"Duplicate code {code!r}")
+        codes_map[code.scan_code] = code
+
+    # max_val = functools.reduce(lambda a, b: max(a, b), map(lambda x: x.scan_code, codes), 0)
+    return codes_map
+    # txts = []
+    # for i in range(max_val):
+    #     code = codes_map.get(i) or get_unknown(i)
+    #     txts.append(f'{code.scan_code}\t{code.name_lower}\t{code.name_upper}\t{code.ascii_lower or None}\t{code.ascii_upper or None}')
+    # pyperclip.copy('\n'.join(txts))
+
+
+def get_ascii_val(val: str):
+    if not val:
+        return 'NULL'
+    if len(val) > 1:
+        return 'NULL'
+    if val == "'":
+        return "'\\''"
+    return repr(val)
+
+
+def main():
+    transform_set1()
+
+
+if __name__ == '__main__':
+    main()