2015-05-30 18:11:16 -04:00
|
|
|
"""
|
|
|
|
Process raw qstr file and output qstr data with length, hash and data bytes.
|
|
|
|
|
2020-05-28 08:40:56 -04:00
|
|
|
This script works with Python 2.7, 3.3 and 3.4.
|
2020-05-28 12:29:28 -04:00
|
|
|
|
|
|
|
For documentation about the format of compressed translated strings, see
|
2022-05-27 15:59:54 -04:00
|
|
|
supervisor/shared/translate/translate.h
|
2015-05-30 18:11:16 -04:00
|
|
|
"""
|
|
|
|
|
2014-03-10 03:07:35 -04:00
|
|
|
from __future__ import print_function
|
|
|
|
|
2021-07-09 12:23:55 -04:00
|
|
|
import bisect
|
2014-01-21 16:40:13 -05:00
|
|
|
import re
|
2014-03-08 10:03:25 -05:00
|
|
|
import sys
|
2014-01-24 17:22:00 -05:00
|
|
|
|
2018-07-31 19:53:54 -04:00
|
|
|
import collections
|
|
|
|
import gettext
|
2018-08-16 03:27:01 -04:00
|
|
|
import os.path
|
2018-07-31 19:53:54 -04:00
|
|
|
|
2021-03-15 09:57:36 -04:00
|
|
|
if hasattr(sys.stdout, "reconfigure"):
|
|
|
|
sys.stdout.reconfigure(encoding="utf-8")
|
|
|
|
sys.stderr.reconfigure(errors="backslashreplace")
|
2020-09-12 20:39:35 -04:00
|
|
|
|
2018-08-16 03:27:01 -04:00
|
|
|
py = os.path.dirname(sys.argv[0])
|
|
|
|
top = os.path.dirname(py)
|
|
|
|
|
2016-04-14 09:37:04 -04:00
|
|
|
# Python 2/3 compatibility:
|
|
|
|
# - iterating through bytes is different
|
|
|
|
# - codepoint2name lives in a different module
|
2014-01-24 17:22:00 -05:00
|
|
|
import platform
|
2021-03-15 09:57:36 -04:00
|
|
|
|
|
|
|
if platform.python_version_tuple()[0] == "2":
|
2016-09-02 00:32:47 -04:00
|
|
|
bytes_cons = lambda val, enc=None: bytearray(val)
|
2014-01-24 17:22:00 -05:00
|
|
|
from htmlentitydefs import codepoint2name
|
2021-03-15 09:57:36 -04:00
|
|
|
elif platform.python_version_tuple()[0] == "3":
|
2016-09-02 00:32:47 -04:00
|
|
|
bytes_cons = bytes
|
2014-01-24 17:22:00 -05:00
|
|
|
from html.entities import codepoint2name
|
2016-09-02 00:32:47 -04:00
|
|
|
# end compatibility code
|
|
|
|
|
2021-03-15 09:57:36 -04:00
|
|
|
codepoint2name[ord("-")] = "hyphen"
|
2014-01-21 16:40:13 -05:00
|
|
|
|
2014-02-15 06:34:50 -05:00
|
|
|
# add some custom names to map characters that aren't in HTML
|
2021-03-15 09:57:36 -04:00
|
|
|
codepoint2name[ord(" ")] = "space"
|
|
|
|
codepoint2name[ord("'")] = "squot"
|
|
|
|
codepoint2name[ord(",")] = "comma"
|
|
|
|
codepoint2name[ord(".")] = "dot"
|
|
|
|
codepoint2name[ord(":")] = "colon"
|
|
|
|
codepoint2name[ord(";")] = "semicolon"
|
|
|
|
codepoint2name[ord("/")] = "slash"
|
|
|
|
codepoint2name[ord("%")] = "percent"
|
|
|
|
codepoint2name[ord("#")] = "hash"
|
|
|
|
codepoint2name[ord("(")] = "paren_open"
|
|
|
|
codepoint2name[ord(")")] = "paren_close"
|
|
|
|
codepoint2name[ord("[")] = "bracket_open"
|
|
|
|
codepoint2name[ord("]")] = "bracket_close"
|
|
|
|
codepoint2name[ord("{")] = "brace_open"
|
|
|
|
codepoint2name[ord("}")] = "brace_close"
|
|
|
|
codepoint2name[ord("*")] = "star"
|
|
|
|
codepoint2name[ord("!")] = "bang"
|
|
|
|
codepoint2name[ord("\\")] = "backslash"
|
|
|
|
codepoint2name[ord("+")] = "plus"
|
|
|
|
codepoint2name[ord("$")] = "dollar"
|
|
|
|
codepoint2name[ord("=")] = "equals"
|
|
|
|
codepoint2name[ord("?")] = "question"
|
|
|
|
codepoint2name[ord("@")] = "at_sign"
|
|
|
|
codepoint2name[ord("^")] = "caret"
|
|
|
|
codepoint2name[ord("|")] = "pipe"
|
|
|
|
codepoint2name[ord("~")] = "tilde"
|
2014-02-15 06:34:50 -05:00
|
|
|
|
2018-08-09 18:58:45 -04:00
|
|
|
C_ESCAPES = {
|
|
|
|
"\a": "\\a",
|
|
|
|
"\b": "\\b",
|
|
|
|
"\f": "\\f",
|
2018-08-10 19:17:03 -04:00
|
|
|
"\n": "\\n",
|
2018-08-09 18:58:45 -04:00
|
|
|
"\r": "\\r",
|
|
|
|
"\t": "\\t",
|
|
|
|
"\v": "\\v",
|
2021-03-15 09:57:36 -04:00
|
|
|
"'": "\\'",
|
|
|
|
'"': '\\"',
|
2018-08-09 18:58:45 -04:00
|
|
|
}
|
|
|
|
|
2021-04-22 20:55:39 -04:00
|
|
|
# static qstrs, should be sorted
|
|
|
|
# These are qstrs that are always included and always have the same number. It allows mpy files to omit them.
|
|
|
|
static_qstr_list = [
|
|
|
|
"",
|
|
|
|
"__dir__", # Put __dir__ after empty qstr for builtin dir() to work
|
|
|
|
"\n",
|
|
|
|
" ",
|
|
|
|
"*",
|
|
|
|
"/",
|
|
|
|
"<module>",
|
|
|
|
"_",
|
|
|
|
"__call__",
|
|
|
|
"__class__",
|
|
|
|
"__delitem__",
|
|
|
|
"__enter__",
|
|
|
|
"__exit__",
|
|
|
|
"__getattr__",
|
|
|
|
"__getitem__",
|
|
|
|
"__hash__",
|
|
|
|
"__init__",
|
|
|
|
"__int__",
|
|
|
|
"__iter__",
|
|
|
|
"__len__",
|
|
|
|
"__main__",
|
|
|
|
"__module__",
|
|
|
|
"__name__",
|
|
|
|
"__new__",
|
|
|
|
"__next__",
|
|
|
|
"__qualname__",
|
|
|
|
"__repr__",
|
|
|
|
"__setitem__",
|
|
|
|
"__str__",
|
|
|
|
"ArithmeticError",
|
|
|
|
"AssertionError",
|
|
|
|
"AttributeError",
|
|
|
|
"BaseException",
|
|
|
|
"EOFError",
|
|
|
|
"Ellipsis",
|
|
|
|
"Exception",
|
|
|
|
"GeneratorExit",
|
|
|
|
"ImportError",
|
|
|
|
"IndentationError",
|
|
|
|
"IndexError",
|
|
|
|
"KeyError",
|
|
|
|
"KeyboardInterrupt",
|
|
|
|
"LookupError",
|
|
|
|
"MemoryError",
|
|
|
|
"NameError",
|
|
|
|
"NoneType",
|
|
|
|
"NotImplementedError",
|
|
|
|
"OSError",
|
|
|
|
"OverflowError",
|
|
|
|
"RuntimeError",
|
|
|
|
"StopIteration",
|
|
|
|
"SyntaxError",
|
|
|
|
"SystemExit",
|
|
|
|
"TypeError",
|
|
|
|
"ValueError",
|
|
|
|
"ZeroDivisionError",
|
|
|
|
"abs",
|
|
|
|
"all",
|
|
|
|
"any",
|
|
|
|
"append",
|
|
|
|
"args",
|
|
|
|
"bool",
|
|
|
|
"builtins",
|
|
|
|
"bytearray",
|
|
|
|
"bytecode",
|
|
|
|
"bytes",
|
|
|
|
"callable",
|
|
|
|
"chr",
|
|
|
|
"classmethod",
|
|
|
|
"clear",
|
|
|
|
"close",
|
|
|
|
"const",
|
|
|
|
"copy",
|
|
|
|
"count",
|
|
|
|
"dict",
|
|
|
|
"dir",
|
|
|
|
"divmod",
|
|
|
|
"end",
|
|
|
|
"endswith",
|
|
|
|
"eval",
|
|
|
|
"exec",
|
|
|
|
"extend",
|
|
|
|
"find",
|
|
|
|
"format",
|
|
|
|
"from_bytes",
|
|
|
|
"get",
|
|
|
|
"getattr",
|
|
|
|
"globals",
|
|
|
|
"hasattr",
|
|
|
|
"hash",
|
|
|
|
"id",
|
|
|
|
"index",
|
|
|
|
"insert",
|
|
|
|
"int",
|
|
|
|
"isalpha",
|
|
|
|
"isdigit",
|
|
|
|
"isinstance",
|
|
|
|
"islower",
|
|
|
|
"isspace",
|
|
|
|
"issubclass",
|
|
|
|
"isupper",
|
|
|
|
"items",
|
|
|
|
"iter",
|
|
|
|
"join",
|
|
|
|
"key",
|
|
|
|
"keys",
|
|
|
|
"len",
|
|
|
|
"list",
|
|
|
|
"little",
|
|
|
|
"locals",
|
|
|
|
"lower",
|
|
|
|
"lstrip",
|
|
|
|
"main",
|
|
|
|
"map",
|
|
|
|
"micropython",
|
|
|
|
"next",
|
|
|
|
"object",
|
|
|
|
"open",
|
|
|
|
"ord",
|
|
|
|
"pop",
|
|
|
|
"popitem",
|
|
|
|
"pow",
|
|
|
|
"print",
|
|
|
|
"range",
|
|
|
|
"read",
|
|
|
|
"readinto",
|
|
|
|
"readline",
|
|
|
|
"remove",
|
|
|
|
"replace",
|
|
|
|
"repr",
|
|
|
|
"reverse",
|
|
|
|
"rfind",
|
|
|
|
"rindex",
|
|
|
|
"round",
|
|
|
|
"rsplit",
|
|
|
|
"rstrip",
|
|
|
|
"self",
|
|
|
|
"send",
|
|
|
|
"sep",
|
|
|
|
"set",
|
|
|
|
"setattr",
|
|
|
|
"setdefault",
|
|
|
|
"sort",
|
|
|
|
"sorted",
|
|
|
|
"split",
|
|
|
|
"start",
|
|
|
|
"startswith",
|
|
|
|
"staticmethod",
|
|
|
|
"step",
|
|
|
|
"stop",
|
|
|
|
"str",
|
|
|
|
"strip",
|
|
|
|
"sum",
|
|
|
|
"super",
|
|
|
|
"throw",
|
|
|
|
"to_bytes",
|
|
|
|
"tuple",
|
|
|
|
"type",
|
|
|
|
"update",
|
|
|
|
"upper",
|
|
|
|
"utf-8",
|
|
|
|
"value",
|
|
|
|
"values",
|
|
|
|
"write",
|
|
|
|
"zip",
|
|
|
|
]
|
|
|
|
|
2023-08-14 00:47:22 -04:00
|
|
|
|
2014-01-21 16:40:13 -05:00
|
|
|
# this must match the equivalent function in qstr.c
|
2015-07-20 07:03:13 -04:00
|
|
|
def compute_hash(qstr, bytes_hash):
|
2014-03-25 11:27:15 -04:00
|
|
|
hash = 5381
|
2016-09-02 00:32:47 -04:00
|
|
|
for b in qstr:
|
|
|
|
hash = (hash * 33) ^ b
|
2014-06-06 16:55:27 -04:00
|
|
|
# Make sure that valid hash is never zero, zero means "hash not computed"
|
2015-07-20 07:03:13 -04:00
|
|
|
return (hash & ((1 << (8 * bytes_hash)) - 1)) or 1
|
2014-01-21 16:40:13 -05:00
|
|
|
|
2021-03-15 09:57:36 -04:00
|
|
|
|
2016-01-31 07:59:59 -05:00
|
|
|
def qstr_escape(qst):
|
2016-04-13 17:12:39 -04:00
|
|
|
def esc_char(m):
|
|
|
|
c = ord(m.group(0))
|
|
|
|
try:
|
|
|
|
name = codepoint2name[c]
|
|
|
|
except KeyError:
|
2021-03-15 09:57:36 -04:00
|
|
|
name = "0x%02x" % c
|
|
|
|
return "_" + name + "_"
|
|
|
|
|
|
|
|
return re.sub(r"[^A-Za-z0-9_]", esc_char, qst)
|
|
|
|
|
2016-01-31 07:59:59 -05:00
|
|
|
|
2023-08-12 15:44:27 -04:00
|
|
|
def parse_input_headers_with_translations(infiles):
|
2015-01-11 12:52:45 -05:00
|
|
|
qcfgs = {}
|
2014-01-23 17:22:00 -05:00
|
|
|
qstrs = {}
|
2023-08-12 15:44:27 -04:00
|
|
|
translations = set()
|
2021-04-22 20:55:39 -04:00
|
|
|
|
|
|
|
# add static qstrs
|
|
|
|
for qstr in static_qstr_list:
|
|
|
|
# work out the corresponding qstr name
|
|
|
|
ident = qstr_escape(qstr)
|
|
|
|
|
|
|
|
# don't add duplicates
|
|
|
|
assert ident not in qstrs
|
|
|
|
|
|
|
|
# add the qstr to the list, with order number to retain original order in file
|
|
|
|
order = len(qstrs) - 300000
|
|
|
|
qstrs[ident] = (order, ident, qstr)
|
|
|
|
|
|
|
|
# read the qstrs in from the input files
|
2014-01-21 16:40:13 -05:00
|
|
|
for infile in infiles:
|
2021-03-15 09:57:36 -04:00
|
|
|
with open(infile, "rt") as f:
|
2014-01-21 16:40:13 -05:00
|
|
|
for line in f:
|
2015-01-11 12:52:45 -05:00
|
|
|
line = line.strip()
|
|
|
|
|
|
|
|
# is this a config line?
|
2021-03-15 09:57:36 -04:00
|
|
|
match = re.match(r"^QCFG\((.+), (.+)\)", line)
|
2015-01-11 12:52:45 -05:00
|
|
|
if match:
|
|
|
|
value = match.group(2)
|
2021-03-15 09:57:36 -04:00
|
|
|
if value[0] == "(" and value[-1] == ")":
|
2015-01-11 12:52:45 -05:00
|
|
|
# strip parenthesis from config value
|
|
|
|
value = value[1:-1]
|
|
|
|
qcfgs[match.group(1)] = value
|
|
|
|
continue
|
|
|
|
|
2018-07-31 19:53:54 -04:00
|
|
|
match = re.match(r'^TRANSLATE\("(.*)"\)$', line)
|
|
|
|
if match:
|
2023-08-12 15:44:27 -04:00
|
|
|
translations.add(match.group(1))
|
2018-07-31 19:53:54 -04:00
|
|
|
continue
|
|
|
|
|
2014-05-02 15:10:47 -04:00
|
|
|
# is this a QSTR line?
|
2021-03-15 09:57:36 -04:00
|
|
|
match = re.match(r"^Q\((.*)\)$", line)
|
2014-05-02 15:10:47 -04:00
|
|
|
if not match:
|
2014-04-13 08:16:51 -04:00
|
|
|
continue
|
2014-01-21 16:40:13 -05:00
|
|
|
|
|
|
|
# get the qstr value
|
|
|
|
qstr = match.group(1)
|
2016-04-14 10:22:36 -04:00
|
|
|
|
2021-04-23 15:26:42 -04:00
|
|
|
# special cases to specify control characters
|
2021-03-15 09:57:36 -04:00
|
|
|
if qstr == "\\n":
|
|
|
|
qstr = "\n"
|
2021-04-23 15:26:42 -04:00
|
|
|
elif qstr == "\\r\\n":
|
|
|
|
qstr = "\r\n"
|
2016-04-14 10:22:36 -04:00
|
|
|
|
|
|
|
# work out the corresponding qstr name
|
2016-01-31 07:59:59 -05:00
|
|
|
ident = qstr_escape(qstr)
|
2014-01-21 16:40:13 -05:00
|
|
|
|
|
|
|
# don't add duplicates
|
2014-01-23 17:22:00 -05:00
|
|
|
if ident in qstrs:
|
2014-01-21 16:40:13 -05:00
|
|
|
continue
|
|
|
|
|
2014-01-24 17:22:00 -05:00
|
|
|
# add the qstr to the list, with order number to retain original order in file
|
2017-10-21 04:06:32 -04:00
|
|
|
order = len(qstrs)
|
|
|
|
# but put special method names like __add__ at the top of list, so
|
|
|
|
# that their id's fit into a byte
|
|
|
|
if ident == "":
|
|
|
|
# Sort empty qstr above all still
|
|
|
|
order = -200000
|
2018-05-10 09:10:46 -04:00
|
|
|
elif ident == "__dir__":
|
|
|
|
# Put __dir__ after empty qstr for builtin dir() to work
|
|
|
|
order = -190000
|
2017-10-21 04:06:32 -04:00
|
|
|
elif ident.startswith("__"):
|
|
|
|
order -= 100000
|
|
|
|
qstrs[ident] = (order, ident, qstr)
|
2014-01-21 16:40:13 -05:00
|
|
|
|
2018-07-31 19:53:54 -04:00
|
|
|
if not qcfgs and qstrs:
|
2015-10-11 04:09:57 -04:00
|
|
|
sys.stderr.write("ERROR: Empty preprocessor output - check for errors above\n")
|
|
|
|
sys.exit(1)
|
|
|
|
|
2023-08-12 15:44:27 -04:00
|
|
|
return qcfgs, qstrs, translations
|
2016-01-31 07:59:59 -05:00
|
|
|
|
2023-08-14 00:47:22 -04:00
|
|
|
|
2023-08-12 15:44:27 -04:00
|
|
|
# Used externally by mpy-tool.py. Don't pass back translations.
|
|
|
|
def parse_input_headers(infiles):
|
|
|
|
qcfgs, qstrs, translations = parse_input_headers_with_translations(infiles)
|
|
|
|
return (qcfgs, qstrs)
|
2021-04-30 16:30:13 -04:00
|
|
|
|
2023-08-14 00:47:22 -04:00
|
|
|
|
2023-08-10 20:06:32 -04:00
|
|
|
def escape_bytes(qstr, qbytes):
|
2021-04-06 12:33:03 -04:00
|
|
|
if all(32 <= ord(c) <= 126 and c != "\\" and c != '"' for c in qstr):
|
|
|
|
# qstr is all printable ASCII so render it as-is (for easier debugging)
|
|
|
|
return qstr
|
|
|
|
else:
|
|
|
|
# qstr contains non-printable codes so render entire thing as hex pairs
|
|
|
|
return "".join(("\\x%02x" % b) for b in qbytes)
|
2021-03-15 09:57:36 -04:00
|
|
|
|
2021-04-30 16:30:13 -04:00
|
|
|
|
2016-01-31 07:59:59 -05:00
|
|
|
def make_bytes(cfg_bytes_len, cfg_bytes_hash, qstr):
|
2021-03-15 09:57:36 -04:00
|
|
|
qbytes = bytes_cons(qstr, "utf8")
|
2016-09-02 00:32:47 -04:00
|
|
|
qlen = len(qbytes)
|
|
|
|
qhash = compute_hash(qbytes, cfg_bytes_hash)
|
2016-01-31 07:59:59 -05:00
|
|
|
if qlen >= (1 << (8 * cfg_bytes_len)):
|
2021-03-15 09:57:36 -04:00
|
|
|
print("qstr is too long:", qstr)
|
2016-01-31 07:59:59 -05:00
|
|
|
assert False
|
2023-08-10 20:06:32 -04:00
|
|
|
qdata = escape_bytes(qstr, qbytes)
|
2021-04-06 12:33:03 -04:00
|
|
|
return '%d, %d, "%s"' % (qhash, qlen, qdata)
|
2016-01-31 07:59:59 -05:00
|
|
|
|
2021-03-15 09:57:36 -04:00
|
|
|
|
2023-08-12 15:44:27 -04:00
|
|
|
def print_qstr_data(qcfgs, qstrs, translations):
|
2015-01-11 17:27:30 -05:00
|
|
|
# get config variables
|
2021-03-15 09:57:36 -04:00
|
|
|
cfg_bytes_len = int(qcfgs["BYTES_IN_LEN"])
|
|
|
|
cfg_bytes_hash = int(qcfgs["BYTES_IN_HASH"])
|
2015-01-11 17:27:30 -05:00
|
|
|
|
2015-07-31 07:57:36 -04:00
|
|
|
# print out the starter of the generated C header file
|
2021-03-15 09:57:36 -04:00
|
|
|
print("// This file was automatically generated by makeqstrdata.py")
|
|
|
|
print("")
|
2015-01-11 17:27:30 -05:00
|
|
|
|
2015-01-11 12:52:45 -05:00
|
|
|
# add NULL qstr with no hash or data
|
2021-04-23 15:26:42 -04:00
|
|
|
print('QDEF(MP_QSTRnull, 0, 0, "")')
|
2015-01-11 17:27:30 -05:00
|
|
|
|
2018-07-31 19:53:54 -04:00
|
|
|
total_qstr_size = 0
|
2015-01-11 17:27:30 -05:00
|
|
|
# go through each qstr and print it out
|
2014-04-11 13:36:08 -04:00
|
|
|
for order, ident, qstr in sorted(qstrs.values(), key=lambda x: x[0]):
|
2016-01-31 07:59:59 -05:00
|
|
|
qbytes = make_bytes(cfg_bytes_len, cfg_bytes_hash, qstr)
|
2021-03-15 09:57:36 -04:00
|
|
|
print("QDEF(MP_QSTR_%s, %s)" % (ident, qbytes))
|
2021-04-22 20:55:39 -04:00
|
|
|
|
2018-07-31 19:53:54 -04:00
|
|
|
total_qstr_size += len(qstr)
|
|
|
|
|
2022-06-02 14:48:56 -04:00
|
|
|
print(
|
|
|
|
"// Enumerate translated texts but don't actually include translations. Instead, the linker will link them in."
|
2021-03-15 09:57:36 -04:00
|
|
|
)
|
2023-08-12 15:44:27 -04:00
|
|
|
for i, original in enumerate(sorted(translations)):
|
2022-06-02 14:48:56 -04:00
|
|
|
print('TRANSLATION("{}", {})'.format(original, i))
|
2016-01-31 07:59:59 -05:00
|
|
|
|
2018-07-31 19:53:54 -04:00
|
|
|
print()
|
|
|
|
print("// {} bytes worth of qstr".format(total_qstr_size))
|
|
|
|
|
2021-03-15 09:57:36 -04:00
|
|
|
|
2023-08-10 20:06:32 -04:00
|
|
|
def do_work(infiles):
|
2023-08-12 15:44:27 -04:00
|
|
|
qcfgs, qstrs, translations = parse_input_headers_with_translations(infiles)
|
|
|
|
print_qstr_data(qcfgs, qstrs, translations)
|
2014-01-21 16:40:13 -05:00
|
|
|
|
2023-08-14 00:47:22 -04:00
|
|
|
|
2014-01-21 16:40:13 -05:00
|
|
|
if __name__ == "__main__":
|
2018-07-31 19:53:54 -04:00
|
|
|
import argparse
|
|
|
|
|
2021-03-15 09:57:36 -04:00
|
|
|
parser = argparse.ArgumentParser(
|
|
|
|
description="Process QSTR definitions into headers for compilation"
|
|
|
|
)
|
|
|
|
parser.add_argument(
|
|
|
|
"infiles", metavar="N", type=str, nargs="+", help="an integer for the accumulator"
|
|
|
|
)
|
2018-07-31 19:53:54 -04:00
|
|
|
|
|
|
|
args = parser.parse_args()
|
|
|
|
|
2023-08-10 20:06:32 -04:00
|
|
|
do_work(args.infiles)
|