Systemd/hwdb.d/parse_hwdb.py
Zbigniew Jędrzejewski-Szmek 77547d5313 hwdb: check that uppercase digits are used in modalias patterns
This is all confusing as hell, becuase in some places lowercase hexadecimal
digits are used, and in other places uppercase. This adds a check for the
most common case that we and others got wrong.

I tried to extend the general grammar in hwdb_grammar() to include this check,
but it quickly became very complicated and didn't seem to work properly. Doing
initial parsing with more general rules is easier and also seems to give better
error messages:

/home/zbyszek/src/systemd-work/build/../hwdb.d/60-autosuspend.hwdb: 3 match groups, 5 matches, 3 properties
Pattern 'v058fp9540*' is invalid: Expected W:(0123...), found 'f'  (at char 4), (line:1, col:5)
2020-07-17 11:15:58 +02:00

285 lines
11 KiB
Python
Executable file

#!/usr/bin/env python3
# SPDX-License-Identifier: MIT
#
# This file is distributed under the MIT license, see below.
#
# The MIT License (MIT)
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import glob
import string
import sys
import os
try:
from pyparsing import (Word, White, Literal, ParserElement, Regex, LineEnd,
OneOrMore, Combine, Or, Optional, Suppress, Group,
nums, alphanums, printables,
stringEnd, pythonStyleComment, QuotedString,
ParseBaseException)
except ImportError:
print('pyparsing is not available')
sys.exit(77)
try:
from evdev.ecodes import ecodes
except ImportError:
ecodes = None
print('WARNING: evdev is not available')
try:
from functools import lru_cache
except ImportError:
# don't do caching on old python
lru_cache = lambda: (lambda f: f)
EOL = LineEnd().suppress()
EMPTYLINE = LineEnd()
COMMENTLINE = pythonStyleComment + EOL
INTEGER = Word(nums)
STRING = QuotedString('"')
REAL = Combine((INTEGER + Optional('.' + Optional(INTEGER))) ^ ('.' + INTEGER))
SIGNED_REAL = Combine(Optional(Word('-+')) + REAL)
UDEV_TAG = Word(string.ascii_uppercase, alphanums + '_')
# Those patterns are used in type-specific matches
TYPES = {'mouse': ('usb', 'bluetooth', 'ps2', '*'),
'evdev': ('name', 'atkbd', 'input'),
'id-input': ('modalias'),
'touchpad': ('i8042', 'rmi', 'bluetooth', 'usb'),
'joystick': ('i8042', 'rmi', 'bluetooth', 'usb'),
'keyboard': ('name', ),
'sensor': ('modalias', ),
}
# Patterns that are used to set general properties on a device
GENERAL_MATCHES = {'acpi',
'bluetooth',
'usb',
'pci',
'sdio',
'vmbus',
'OUI',
}
def upperhex_word(length):
return Word(nums + 'ABCDEF', exact=length)
@lru_cache()
def hwdb_grammar():
ParserElement.setDefaultWhitespaceChars('')
prefix = Or(category + ':' + Or(conn) + ':'
for category, conn in TYPES.items())
matchline_typed = Combine(prefix + Word(printables + ' ' + '®'))
matchline_general = Combine(Or(GENERAL_MATCHES) + ':' + Word(printables + ' ' + '®'))
matchline = (matchline_typed | matchline_general) + EOL
propertyline = (White(' ', exact=1).suppress() +
Combine(UDEV_TAG - '=' - Word(alphanums + '_=:@*.!-;, "') - Optional(pythonStyleComment)) +
EOL)
propertycomment = White(' ', exact=1) + pythonStyleComment + EOL
group = (OneOrMore(matchline('MATCHES*') ^ COMMENTLINE.suppress()) -
OneOrMore(propertyline('PROPERTIES*') ^ propertycomment.suppress()) -
(EMPTYLINE ^ stringEnd()).suppress())
commentgroup = OneOrMore(COMMENTLINE).suppress() - EMPTYLINE.suppress()
grammar = OneOrMore(Group(group)('GROUPS*') ^ commentgroup) + stringEnd()
return grammar
@lru_cache()
def property_grammar():
ParserElement.setDefaultWhitespaceChars(' ')
dpi_setting = (Optional('*')('DEFAULT') + INTEGER('DPI') + Suppress('@') + INTEGER('HZ'))('SETTINGS*')
mount_matrix_row = SIGNED_REAL + ',' + SIGNED_REAL + ',' + SIGNED_REAL
mount_matrix = (mount_matrix_row + ';' + mount_matrix_row + ';' + mount_matrix_row)('MOUNT_MATRIX')
props = (('MOUSE_DPI', Group(OneOrMore(dpi_setting))),
('MOUSE_WHEEL_CLICK_ANGLE', INTEGER),
('MOUSE_WHEEL_CLICK_ANGLE_HORIZONTAL', INTEGER),
('MOUSE_WHEEL_CLICK_COUNT', INTEGER),
('MOUSE_WHEEL_CLICK_COUNT_HORIZONTAL', INTEGER),
('ID_AUTOSUSPEND', Literal('1')),
('ID_INPUT', Literal('1')),
('ID_INPUT_ACCELEROMETER', Literal('1')),
('ID_INPUT_JOYSTICK', Literal('1')),
('ID_INPUT_KEY', Literal('1')),
('ID_INPUT_KEYBOARD', Literal('1')),
('ID_INPUT_MOUSE', Literal('1')),
('ID_INPUT_POINTINGSTICK', Literal('1')),
('ID_INPUT_SWITCH', Literal('1')),
('ID_INPUT_TABLET', Literal('1')),
('ID_INPUT_TABLET_PAD', Literal('1')),
('ID_INPUT_TOUCHPAD', Literal('1')),
('ID_INPUT_TOUCHSCREEN', Literal('1')),
('ID_INPUT_TRACKBALL', Literal('1')),
('POINTINGSTICK_SENSITIVITY', INTEGER),
('POINTINGSTICK_CONST_ACCEL', REAL),
('ID_INPUT_JOYSTICK_INTEGRATION', Or(('internal', 'external'))),
('ID_INPUT_TOUCHPAD_INTEGRATION', Or(('internal', 'external'))),
('XKB_FIXED_LAYOUT', STRING),
('XKB_FIXED_VARIANT', STRING),
('XKB_FIXED_MODEL', STRING),
('KEYBOARD_LED_NUMLOCK', Literal('0')),
('KEYBOARD_LED_CAPSLOCK', Literal('0')),
('ACCEL_MOUNT_MATRIX', mount_matrix),
('ACCEL_LOCATION', Or(('display', 'base'))),
('PROXIMITY_NEAR_LEVEL', INTEGER),
)
fixed_props = [Literal(name)('NAME') - Suppress('=') - val('VALUE')
for name, val in props]
kbd_props = [Regex(r'KEYBOARD_KEY_[0-9a-f]+')('NAME')
- Suppress('=') -
('!' ^ (Optional('!') - Word(alphanums + '_')))('VALUE')
]
abs_props = [Regex(r'EVDEV_ABS_[0-9a-f]{2}')('NAME')
- Suppress('=') -
Word(nums + ':')('VALUE')
]
grammar = Or(fixed_props + kbd_props + abs_props) + EOL
return grammar
ERROR = False
def error(fmt, *args, **kwargs):
global ERROR
ERROR = True
print(fmt.format(*args, **kwargs))
def convert_properties(group):
matches = [m[0] for m in group.MATCHES]
props = [p[0] for p in group.PROPERTIES]
return matches, props
def parse(fname):
grammar = hwdb_grammar()
try:
with open(fname, 'r', encoding='UTF-8') as f:
parsed = grammar.parseFile(f)
except ParseBaseException as e:
error('Cannot parse {}: {}', fname, e)
return []
return [convert_properties(g) for g in parsed.GROUPS]
def check_matches(groups):
matches = sum((group[0] for group in groups), [])
# This is a partial check. The other cases could be also done, but those
# two are most commonly wrong.
grammars = { 'usb' : 'v' + upperhex_word(4) + Optional('p' + upperhex_word(4)),
'pci' : 'v' + upperhex_word(8) + Optional('d' + upperhex_word(8)),
}
for match in matches:
prefix, rest = match.split(':', maxsplit=1)
gr = grammars.get(prefix)
if gr:
try:
gr.parseString(rest)
except ParseBaseException as e:
error('Pattern {!r} is invalid: {}', rest, e)
continue
if rest[-1] not in '*:':
error('pattern {} does not end with "*" or ":"', match)
matches.sort()
prev = None
for match in matches:
if match == prev:
error('Match {!r} is duplicated', match)
prev = match
def check_one_default(prop, settings):
defaults = [s for s in settings if s.DEFAULT]
if len(defaults) > 1:
error('More than one star entry: {!r}', prop)
def check_one_mount_matrix(prop, value):
numbers = [s for s in value if s not in {';', ','}]
if len(numbers) != 9:
error('Wrong accel matrix: {!r}', prop)
try:
numbers = [abs(float(number)) for number in numbers]
except ValueError:
error('Wrong accel matrix: {!r}', prop)
bad_x, bad_y, bad_z = max(numbers[0:3]) == 0, max(numbers[3:6]) == 0, max(numbers[6:9]) == 0
if bad_x or bad_y or bad_z:
error('Mount matrix is all zero in {} row: {!r}',
'x' if bad_x else ('y' if bad_y else 'z'),
prop)
def check_one_keycode(prop, value):
if value != '!' and ecodes is not None:
key = 'KEY_' + value.upper()
if not (key in ecodes or
value.upper() in ecodes or
# new keys added in kernel 5.5
'KBD_LCD_MENU' in key):
error('Keycode {} unknown', key)
def check_properties(groups):
grammar = property_grammar()
for matches, props in groups:
prop_names = set()
for prop in props:
# print('--', prop)
prop = prop.partition('#')[0].rstrip()
try:
parsed = grammar.parseString(prop)
except ParseBaseException as e:
error('Failed to parse: {!r}', prop)
continue
# print('{!r}'.format(parsed))
if parsed.NAME in prop_names:
error('Property {} is duplicated', parsed.NAME)
prop_names.add(parsed.NAME)
if parsed.NAME == 'MOUSE_DPI':
check_one_default(prop, parsed.VALUE.SETTINGS)
elif parsed.NAME == 'ACCEL_MOUNT_MATRIX':
check_one_mount_matrix(prop, parsed.VALUE)
elif parsed.NAME.startswith('KEYBOARD_KEY_'):
val = parsed.VALUE if isinstance(parsed.VALUE, str) else parsed.VALUE[0]
check_one_keycode(prop, val)
def print_summary(fname, groups):
n_matches = sum(len(matches) for matches, props in groups)
n_props = sum(len(props) for matches, props in groups)
print('{}: {} match groups, {} matches, {} properties'
.format(fname, len(groups), n_matches, n_props))
if n_matches == 0 or n_props == 0:
error('{}: no matches or props'.format(fname))
if __name__ == '__main__':
args = sys.argv[1:] or sorted(glob.glob(os.path.dirname(sys.argv[0]) + '/[67][0-9]-*.hwdb'))
for fname in args:
groups = parse(fname)
print_summary(fname, groups)
check_matches(groups)
check_properties(groups)
sys.exit(ERROR)