Clean up qmk generate-autocorrect-data (#19710)

Author: fauxpark

docs/feature_autocorrect.md

@@ -86,7 +86,7 @@ The `qmk generate-autocorrect-data` commands can make an effort to check for ent
 
 ## Overriding Autocorrect
 
-Occasionally you might actually want to type a typo (for instance, while editing autocorrection_dict.txt) without being autocorrected. There are a couple of ways to do this:
+Occasionally you might actually want to type a typo (for instance, while editing autocorrect_dict.txt) without being autocorrected. There are a couple of ways to do this:
 
 1. Begin typing the typo.
 2. Before typing the last letter, press and release the Ctrl or Alt key.
@@ -238,13 +238,13 @@ bool apply_autocorrect(uint8_t backspaces, const char *str) {
 
 ## Appendix: Trie binary data format :id=appendix
 
-This section details how the trie is serialized to byte data in autocorrection_data. You don’t need to care about this to use this autocorrection implementation. But it is documented for the record in case anyone is interested in modifying the implementation, or just curious how it works.
+This section details how the trie is serialized to byte data in autocorrect_data. You don’t need to care about this to use this autocorrection implementation. But it is documented for the record in case anyone is interested in modifying the implementation, or just curious how it works.
 
 What I did here is fairly arbitrary, but it is simple to decode and gets the job done.
 
 ### Encoding :id=encoding
 
-All autocorrection data is stored in a single flat array autocorrection_data. Each trie node is associated with a byte offset into this array, where data for that node is encoded, beginning with root at offset 0. There are three kinds of nodes. The highest two bits of the first byte of the node indicate what kind:
+All autocorrection data is stored in a single flat array autocorrect_data. Each trie node is associated with a byte offset into this array, where data for that node is encoded, beginning with root at offset 0. There are three kinds of nodes. The highest two bits of the first byte of the node indicate what kind:
 
 * 00 ⇒ chain node: a trie node with a single child.
 * 01 ⇒ branching node: a trie node with multiple children.

lib/python/qmk/cli/generate/autocorrect_data.py

@@ -33,9 +33,11 @@
 
 from milc import cli
 
-import qmk.path
+from qmk.commands import dump_lines
+from qmk.constants import GPL2_HEADER_C_LIKE, GENERATED_HEADER_C_LIKE
 from qmk.keyboard import keyboard_completer, keyboard_folder
 from qmk.keymap import keymap_completer, locate_keymap
+from qmk.path import normpath
 
 KC_A = 4
 KC_SPC = 0x2c
@@ -63,9 +65,10 @@ def parse_file(file_name: str) -> List[Tuple[str, str]]:
     try:
         from english_words import english_words_lower_alpha_set as correct_words
     except ImportError:
-        cli.echo('Autocorrection will falsely trigger when a typo is a substring of a correctly spelled word.')
-        cli.echo('To check for this, install the english_words package and rerun this script:')
-        cli.echo('  {fg_cyan}python3 -m pip install english_words')
+        if not cli.args.quiet:
+            cli.echo('Autocorrection will falsely trigger when a typo is a substring of a correctly spelled word.')
+            cli.echo('To check for this, install the english_words package and rerun this script:')
+            cli.echo('  {fg_cyan}python3 -m pip install english_words')
         # Use a minimal word list as a fallback.
         correct_words = ('information', 'available', 'international', 'language', 'loosest', 'reference', 'wealthier', 'entertainment', 'association', 'provides', 'technology', 'statehood')
 
@@ -232,58 +235,51 @@ def encode_link(link: Dict[str, Any]) -> List[int]:
     return [byte_offset & 255, byte_offset >> 8]
 
 
-def write_generated_code(autocorrections: List[Tuple[str, str]], data: List[int], file_name: str) -> None:
-    """Writes autocorrection data as generated C code to `file_name`.
-  Args:
-    autocorrections: List of (typo, correction) tuples.
-    data: List of ints in 0-255, the serialized trie.
-    file_name: String, path of the output C file.
-  """
-    assert all(0 <= b <= 255 for b in data)
-
-    def typo_len(e: Tuple[str, str]) -> int:
-        return len(e[0])
+def typo_len(e: Tuple[str, str]) -> int:
+    return len(e[0])
 
-    min_typo = min(autocorrections, key=typo_len)[0]
-    max_typo = max(autocorrections, key=typo_len)[0]
-    generated_code = ''.join([
-        '// Generated code.\n\n', f'// Autocorrection dictionary ({len(autocorrections)} entries):\n', ''.join(sorted(f'//   {typo:<{len(max_typo)}} -> {correction}\n' for typo, correction in autocorrections)),
-        f'\n#define AUTOCORRECT_MIN_LENGTH {len(min_typo)}  // "{min_typo}"\n', f'#define AUTOCORRECT_MAX_LENGTH {len(max_typo)}  // "{max_typo}"\n\n', f'#define DICTIONARY_SIZE {len(data)}\n\n',
-        textwrap.fill('static const uint8_t autocorrect_data[DICTIONARY_SIZE] PROGMEM = {%s};' % (', '.join(map(str, data))), width=120, subsequent_indent='    '), '\n\n'
-    ])
 
-    with open(file_name, 'wt') as f:
-        f.write(generated_code)
+def to_hex(b: int) -> str:
+    return f'0x{b:02X}'
 
 
-@cli.argument('filename', default='autocorrect_dict.txt', help='The autocorrection database file')
+@cli.argument('filename', type=normpath, help='The autocorrection database file')
 @cli.argument('-kb', '--keyboard', type=keyboard_folder, completer=keyboard_completer, help='The keyboard to build a firmware for. Ignored when a configurator export is supplied.')
 @cli.argument('-km', '--keymap', completer=keymap_completer, help='The keymap to build a firmware for. Ignored when a configurator export is supplied.')
-@cli.argument('-o', '--output', arg_only=True, type=qmk.path.normpath, help='File to write to')
+@cli.argument('-o', '--output', arg_only=True, type=normpath, help='File to write to')
+@cli.argument('-q', '--quiet', arg_only=True, action='store_true', help="Quiet mode, only output error messages")
 @cli.subcommand('Generate the autocorrection data file from a dictionary file.')
 def generate_autocorrect_data(cli):
     autocorrections = parse_file(cli.args.filename)
     trie = make_trie(autocorrections)
     data = serialize_trie(autocorrections, trie)
-    # Environment processing
-    if cli.args.output == '-':
-        cli.args.output = None
 
-    if cli.args.output:
-        cli.args.output.parent.mkdir(parents=True, exist_ok=True)
-        cli.log.info('Creating autocorrect database at {fg_cyan}%s', cli.args.output)
-        write_generated_code(autocorrections, data, cli.args.output)
+    current_keyboard = cli.args.keyboard or cli.config.user.keyboard or cli.config.generate_autocorrect_data.keyboard
+    current_keymap = cli.args.keymap or cli.config.user.keymap or cli.config.generate_autocorrect_data.keymap
+
+    if current_keyboard and current_keymap:
+        cli.args.output = locate_keymap(current_keyboard, current_keymap).parent / 'autocorrect_data.h'
 
-    else:
-        current_keyboard = cli.args.keyboard or cli.config.user.keyboard or cli.config.generate_autocorrect_data.keyboard
-        current_keymap = cli.args.keymap or cli.config.user.keymap or cli.config.generate_autocorrect_data.keymap
+    assert all(0 <= b <= 255 for b in data)
 
-        if current_keyboard and current_keymap:
-            filename = locate_keymap(current_keyboard, current_keymap).parent / 'autocorrect_data.h'
-            cli.log.info('Creating autocorrect database at {fg_cyan}%s', filename)
-            write_generated_code(autocorrections, data, filename)
+    min_typo = min(autocorrections, key=typo_len)[0]
+    max_typo = max(autocorrections, key=typo_len)[0]
 
-        else:
-            write_generated_code(autocorrections, data, 'autocorrect_data.h')
+    # Build the autocorrect_data.h file.
+    autocorrect_data_h_lines = [GPL2_HEADER_C_LIKE, GENERATED_HEADER_C_LIKE, '#pragma once', '']
 
-    cli.log.info('Processed %d autocorrection entries to table with %d bytes.', len(autocorrections), len(data))
+    autocorrect_data_h_lines.append(f'// Autocorrection dictionary ({len(autocorrections)} entries):')
+    for typo, correction in autocorrections:
+        autocorrect_data_h_lines.append(f'//   {typo:<{len(max_typo)}} -> {correction}')
+
+    autocorrect_data_h_lines.append('')
+    autocorrect_data_h_lines.append(f'#define AUTOCORRECT_MIN_LENGTH {len(min_typo)} // "{min_typo}"')
+    autocorrect_data_h_lines.append(f'#define AUTOCORRECT_MAX_LENGTH {len(max_typo)} // "{max_typo}"')
+    autocorrect_data_h_lines.append(f'#define DICTIONARY_SIZE {len(data)}')
+    autocorrect_data_h_lines.append('')
+    autocorrect_data_h_lines.append('static const uint8_t autocorrect_data[DICTIONARY_SIZE] PROGMEM = {')
+    autocorrect_data_h_lines.append(textwrap.fill('    %s' % (', '.join(map(to_hex, data))), width=100, subsequent_indent='    '))
+    autocorrect_data_h_lines.append('};')
+
+    # Show the results
+    dump_lines(cli.args.output, autocorrect_data_h_lines, cli.args.quiet)