2018-07-27 13:49:58 +00:00
|
|
|
|
"""pospell is a spellcheckers for po files containing reStructuedText.
|
|
|
|
|
"""
|
2018-07-27 19:57:44 +00:00
|
|
|
|
import io
|
2018-07-31 22:20:03 +00:00
|
|
|
|
import logging
|
2018-07-23 15:37:50 +00:00
|
|
|
|
import subprocess
|
2018-07-28 22:58:20 +00:00
|
|
|
|
import sys
|
2018-07-27 09:38:17 +00:00
|
|
|
|
import tempfile
|
2019-08-20 14:38:03 +00:00
|
|
|
|
from contextlib import redirect_stderr
|
2018-07-27 13:54:10 +00:00
|
|
|
|
from itertools import chain
|
2018-07-23 15:37:50 +00:00
|
|
|
|
from pathlib import Path
|
2019-12-10 14:10:17 +00:00
|
|
|
|
from shutil import which
|
2018-07-27 09:38:17 +00:00
|
|
|
|
|
2018-07-27 19:57:44 +00:00
|
|
|
|
import docutils.frontend
|
|
|
|
|
import docutils.nodes
|
|
|
|
|
import docutils.parsers.rst
|
2018-07-28 22:58:20 +00:00
|
|
|
|
import polib
|
2018-07-27 19:57:44 +00:00
|
|
|
|
from docutils.parsers.rst import roles
|
|
|
|
|
from docutils.utils import new_document
|
|
|
|
|
|
2018-07-31 22:20:03 +00:00
|
|
|
|
import regex
|
|
|
|
|
|
2019-10-17 07:46:30 +00:00
|
|
|
|
__version__ = "1.0.3"
|
2019-10-16 14:55:46 +00:00
|
|
|
|
|
|
|
|
|
DEFAULT_DROP_CAPITALIZED = {"fr": True, "fr_FR": True}
|
|
|
|
|
|
2018-07-28 22:58:20 +00:00
|
|
|
|
try:
|
|
|
|
|
HUNSPELL_VERSION = subprocess.check_output(
|
|
|
|
|
["hunspell", "--version"], universal_newlines=True
|
|
|
|
|
).split("\n")[0]
|
|
|
|
|
except FileNotFoundError:
|
|
|
|
|
print("hunspell not found, please install hunspell.", file=sys.stderr)
|
|
|
|
|
exit(1)
|
|
|
|
|
|
2018-07-27 19:57:44 +00:00
|
|
|
|
|
|
|
|
|
class DummyNodeClass(docutils.nodes.Inline, docutils.nodes.TextElement):
|
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def monkey_patch_role(role):
|
|
|
|
|
def role_or_generic(role_name, language_module, lineno, reporter):
|
|
|
|
|
base_role, message = role(role_name, language_module, lineno, reporter)
|
|
|
|
|
if base_role is None:
|
|
|
|
|
roles.register_generic_role(role_name, DummyNodeClass)
|
|
|
|
|
base_role, message = role(role_name, language_module, lineno, reporter)
|
|
|
|
|
return base_role, message
|
|
|
|
|
|
|
|
|
|
return role_or_generic
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
roles.role = monkey_patch_role(roles.role)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
class NodeToTextVisitor(docutils.nodes.NodeVisitor):
|
|
|
|
|
def __init__(self, document):
|
|
|
|
|
self.output = []
|
2019-07-26 15:40:48 +00:00
|
|
|
|
self.depth = 0
|
2018-07-27 19:57:44 +00:00
|
|
|
|
super().__init__(document)
|
|
|
|
|
|
2019-07-26 15:40:48 +00:00
|
|
|
|
def dispatch_visit(self, node):
|
|
|
|
|
self.depth += 1
|
|
|
|
|
super().dispatch_visit(node)
|
|
|
|
|
|
|
|
|
|
def dispatch_departure(self, node):
|
|
|
|
|
self.depth -= 1
|
|
|
|
|
super().dispatch_departure(node)
|
|
|
|
|
|
2018-07-27 19:57:44 +00:00
|
|
|
|
def unknown_visit(self, node):
|
2019-07-26 15:40:48 +00:00
|
|
|
|
"""Mandatory implementation to visit unknwon nodes.
|
|
|
|
|
"""
|
|
|
|
|
# print(" " * self.depth * 4, node.__class__.__name__, ":", node)
|
|
|
|
|
|
|
|
|
|
def unknown_departure(self, node):
|
|
|
|
|
"""To help debugging tree.
|
|
|
|
|
"""
|
|
|
|
|
# print(node, repr(node), node.__class__.__name__)
|
|
|
|
|
|
|
|
|
|
def visit_emphasis(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
|
|
|
|
|
|
|
|
|
def visit_superscript(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
|
|
|
|
|
|
|
|
|
def visit_title_reference(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
|
|
|
|
|
|
|
|
|
def visit_strong(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
|
|
|
|
|
|
|
|
|
def visit_DummyNodeClass(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
|
|
|
|
|
|
|
|
|
def visit_reference(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
|
|
|
|
|
|
|
|
|
def visit_literal(self, node):
|
|
|
|
|
raise docutils.nodes.SkipChildren
|
2018-07-27 19:57:44 +00:00
|
|
|
|
|
|
|
|
|
def visit_Text(self, node):
|
|
|
|
|
self.output.append(node.rawsource)
|
|
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
|
return " ".join(self.output)
|
|
|
|
|
|
2018-07-23 15:37:50 +00:00
|
|
|
|
|
|
|
|
|
def strip_rst(line):
|
2018-07-27 19:57:44 +00:00
|
|
|
|
if line.endswith("::"):
|
|
|
|
|
# Drop :: at the end, it would cause Literal block expected
|
|
|
|
|
line = line[:-2]
|
|
|
|
|
parser = docutils.parsers.rst.Parser()
|
|
|
|
|
components = (docutils.parsers.rst.Parser,)
|
|
|
|
|
settings = docutils.frontend.OptionParser(
|
|
|
|
|
components=components
|
|
|
|
|
).get_default_values()
|
|
|
|
|
stderr_stringio = io.StringIO()
|
|
|
|
|
with redirect_stderr(stderr_stringio):
|
2019-08-20 14:38:03 +00:00
|
|
|
|
document = new_document("<rst-doc>", settings=settings)
|
2018-07-27 19:57:44 +00:00
|
|
|
|
parser.parse(line, document)
|
|
|
|
|
stderr = stderr_stringio.getvalue()
|
|
|
|
|
if stderr:
|
|
|
|
|
print(stderr.strip(), "while parsing:", line)
|
|
|
|
|
visitor = NodeToTextVisitor(document)
|
|
|
|
|
document.walk(visitor)
|
|
|
|
|
return str(visitor)
|
2018-07-23 15:37:50 +00:00
|
|
|
|
|
|
|
|
|
|
2019-10-16 14:55:46 +00:00
|
|
|
|
def clear(po_path, line, drop_capitalized=False):
|
2018-07-27 09:01:09 +00:00
|
|
|
|
"""Clear various other syntaxes we may encounter in a line.
|
|
|
|
|
"""
|
2019-09-16 08:44:18 +00:00
|
|
|
|
# Normalize spaces
|
|
|
|
|
line = regex.sub(r"\s+", " ", line)
|
2018-07-31 22:20:03 +00:00
|
|
|
|
to_drop = {
|
|
|
|
|
r'<a href="[^"]*?">',
|
2019-10-09 11:06:45 +00:00
|
|
|
|
# Strip accronyms
|
2019-10-16 15:51:50 +00:00
|
|
|
|
r"\b[\w-]*\p{Uppercase}{2,}[0-9.\w-]*\b",
|
2018-07-31 22:20:03 +00:00
|
|
|
|
r"---?", # -- and --- separators to be ignored
|
2019-10-16 15:03:15 +00:00
|
|
|
|
r" - ", # Drop lone dashes (sometimes used in place of -- or ---)
|
2018-07-31 22:20:03 +00:00
|
|
|
|
r"-\\ ", # Ignore "MINUS BACKSLASH SPACE" typically used in
|
|
|
|
|
# formulas, like '-\ *π*' but *π* gets removed too
|
2019-11-16 13:47:22 +00:00
|
|
|
|
r"{[a-z_]*?}", # Sphinx variable
|
2019-08-20 15:34:30 +00:00
|
|
|
|
r"'?-?\b([0-9]+\.)*[0-9]+\.[0-9abcrx]+\b'?", # Versions
|
2019-05-23 18:45:04 +00:00
|
|
|
|
r"[0-9]+h", # Hours
|
2019-11-18 08:52:00 +00:00
|
|
|
|
r"%\([a-z_]+?\)[diouxXeEfFgGcrsa%]", # Sphinx variable
|
2019-09-16 08:44:18 +00:00
|
|
|
|
r"« . »", # Single letter examples (typically in Unicode documentation)
|
2019-09-30 11:37:48 +00:00
|
|
|
|
"\xad", # soft hyphen
|
2018-07-31 22:20:03 +00:00
|
|
|
|
}
|
2019-10-09 11:06:45 +00:00
|
|
|
|
if drop_capitalized:
|
2019-10-16 14:55:46 +00:00
|
|
|
|
to_drop.add(
|
2019-10-09 11:06:45 +00:00
|
|
|
|
# Strip capitalized words in sentences
|
2019-10-16 14:55:46 +00:00
|
|
|
|
r"(?<!\. |^|-)\b(\p{Letter}['’])?\b\p{Uppercase}\p{Letter}[\w.-]*\b"
|
|
|
|
|
)
|
2018-07-31 22:20:03 +00:00
|
|
|
|
if logging.getLogger().isEnabledFor(logging.DEBUG):
|
|
|
|
|
for pattern in to_drop:
|
|
|
|
|
for dropped in regex.findall(pattern, line):
|
|
|
|
|
logging.debug("%s: dropping %r due to from %r", po_path, dropped, line)
|
|
|
|
|
return regex.sub("|".join(to_drop), r"", line)
|
2018-07-27 09:01:09 +00:00
|
|
|
|
|
|
|
|
|
|
2019-10-16 14:55:46 +00:00
|
|
|
|
def po_to_text(po_path, drop_capitalized=False):
|
2018-07-27 13:54:10 +00:00
|
|
|
|
"""Converts a po file to a text file, by stripping the msgids and all
|
|
|
|
|
po syntax, but by keeping the kept lines at their same position /
|
|
|
|
|
line number.
|
|
|
|
|
"""
|
2018-07-23 15:37:50 +00:00
|
|
|
|
buffer = []
|
2018-07-23 17:24:10 +00:00
|
|
|
|
lines = 0
|
2018-07-27 13:54:10 +00:00
|
|
|
|
entries = polib.pofile(po_path)
|
2018-07-23 15:37:50 +00:00
|
|
|
|
for entry in entries:
|
2018-07-27 08:03:21 +00:00
|
|
|
|
if entry.msgid == entry.msgstr:
|
|
|
|
|
continue
|
2018-07-23 17:24:10 +00:00
|
|
|
|
while lines < entry.linenum:
|
2018-07-27 13:54:10 +00:00
|
|
|
|
buffer.append("")
|
2018-07-23 17:24:10 +00:00
|
|
|
|
lines += 1
|
2019-10-09 11:07:09 +00:00
|
|
|
|
buffer.append(clear(po_path, strip_rst(entry.msgstr), drop_capitalized))
|
2018-07-23 17:24:10 +00:00
|
|
|
|
lines += 1
|
2018-07-27 13:54:10 +00:00
|
|
|
|
return "\n".join(buffer)
|
2018-07-23 15:37:50 +00:00
|
|
|
|
|
|
|
|
|
|
2018-07-28 22:58:20 +00:00
|
|
|
|
def parse_args():
|
|
|
|
|
"""Parse command line arguments.
|
2018-07-27 13:54:10 +00:00
|
|
|
|
"""
|
2018-07-23 15:37:50 +00:00
|
|
|
|
import argparse
|
2018-07-27 13:54:10 +00:00
|
|
|
|
|
2018-07-23 15:37:50 +00:00
|
|
|
|
parser = argparse.ArgumentParser(
|
2018-07-27 13:54:10 +00:00
|
|
|
|
description="Check spelling in po files containing restructuredText."
|
|
|
|
|
)
|
|
|
|
|
parser.add_argument(
|
|
|
|
|
"-l",
|
|
|
|
|
"--language",
|
|
|
|
|
type=str,
|
|
|
|
|
default="fr",
|
|
|
|
|
help="Language to check, you'll have to install the corresponding "
|
|
|
|
|
"hunspell dictionary, on Debian see apt list 'hunspell-*'.",
|
|
|
|
|
)
|
|
|
|
|
parser.add_argument(
|
|
|
|
|
"--glob",
|
|
|
|
|
type=str,
|
|
|
|
|
help="Provide a glob pattern, to be interpreted by pospell, to find po files, "
|
|
|
|
|
"like --glob '**/*.po'.",
|
|
|
|
|
)
|
2019-10-09 11:07:09 +00:00
|
|
|
|
parser.add_argument(
|
2019-10-16 14:55:46 +00:00
|
|
|
|
"--drop-capitalized",
|
2019-10-09 11:07:09 +00:00
|
|
|
|
action="store_true",
|
2019-10-16 14:55:46 +00:00
|
|
|
|
help="Always drop capitalized words in sentences (defaults according to the language).",
|
|
|
|
|
)
|
|
|
|
|
parser.add_argument(
|
|
|
|
|
"--no-drop-capitalized",
|
|
|
|
|
action="store_true",
|
|
|
|
|
help="Never drop capitalized words in sentences (defaults according to the language).",
|
2019-10-09 11:07:09 +00:00
|
|
|
|
)
|
2018-07-27 13:54:10 +00:00
|
|
|
|
parser.add_argument(
|
|
|
|
|
"po_file",
|
|
|
|
|
nargs="*",
|
2018-07-27 14:21:28 +00:00
|
|
|
|
type=Path,
|
2018-07-27 13:54:10 +00:00
|
|
|
|
help="Files to check, can optionally be mixed with --glob, or not, "
|
|
|
|
|
"use the one that fit your needs.",
|
|
|
|
|
)
|
2018-07-31 22:20:03 +00:00
|
|
|
|
parser.add_argument(
|
|
|
|
|
"-v",
|
|
|
|
|
"--verbose",
|
|
|
|
|
action="count",
|
|
|
|
|
default=0,
|
|
|
|
|
help="More output, use -vv, -vvv, and so on.",
|
|
|
|
|
)
|
2018-07-28 22:58:20 +00:00
|
|
|
|
parser.add_argument(
|
|
|
|
|
"--version",
|
|
|
|
|
action="version",
|
|
|
|
|
version="%(prog)s " + __version__ + " using hunspell: " + HUNSPELL_VERSION,
|
|
|
|
|
)
|
2018-07-27 13:54:10 +00:00
|
|
|
|
parser.add_argument("--debug", action="store_true")
|
|
|
|
|
parser.add_argument("-p", "--personal-dict", type=str)
|
2018-12-24 14:49:54 +00:00
|
|
|
|
parser.add_argument(
|
|
|
|
|
"--modified", "-m", action="store_true", help="Use git to find modified files."
|
|
|
|
|
)
|
|
|
|
|
args = parser.parse_args()
|
2019-10-16 14:55:46 +00:00
|
|
|
|
if args.drop_capitalized and args.no_drop_capitalized:
|
|
|
|
|
print("Error: don't provide both --drop-capitalized AND --no-drop-capitalized.")
|
|
|
|
|
parser.print_help()
|
|
|
|
|
exit(1)
|
2018-12-24 14:49:54 +00:00
|
|
|
|
if not args.po_file and not args.modified:
|
|
|
|
|
parser.print_help()
|
|
|
|
|
exit(1)
|
|
|
|
|
return args
|
2018-07-28 22:58:20 +00:00
|
|
|
|
|
|
|
|
|
|
2019-10-16 14:55:46 +00:00
|
|
|
|
def spell_check(
|
|
|
|
|
po_files, personal_dict, language, drop_capitalized=False, debug_only=False
|
|
|
|
|
):
|
2018-12-24 14:49:54 +00:00
|
|
|
|
"""Check for spelling mistakes in the files po_files (po format,
|
|
|
|
|
containing restructuredtext), for the given language.
|
|
|
|
|
personal_dict allow to pass a personal dict (-p) option, to hunspell.
|
|
|
|
|
|
|
|
|
|
Debug only will show what's passed to Hunspell instead of passing it.
|
2018-07-28 22:58:20 +00:00
|
|
|
|
"""
|
2018-07-23 22:28:13 +00:00
|
|
|
|
errors = 0
|
2018-12-24 14:49:54 +00:00
|
|
|
|
personal_dict_arg = ["-p", personal_dict] if personal_dict else []
|
2018-07-23 15:37:50 +00:00
|
|
|
|
with tempfile.TemporaryDirectory() as tmpdirname:
|
|
|
|
|
tmpdir = Path(tmpdirname)
|
2018-12-24 14:49:54 +00:00
|
|
|
|
for po_file in po_files:
|
|
|
|
|
if debug_only:
|
2019-10-09 11:07:09 +00:00
|
|
|
|
print(po_to_text(str(po_file), drop_capitalized))
|
2018-07-23 22:00:52 +00:00
|
|
|
|
continue
|
2019-10-16 14:55:46 +00:00
|
|
|
|
(tmpdir / po_file.name).write_text(
|
|
|
|
|
po_to_text(str(po_file), drop_capitalized)
|
|
|
|
|
)
|
2019-11-14 22:30:52 +00:00
|
|
|
|
try:
|
|
|
|
|
output = subprocess.check_output(
|
|
|
|
|
["hunspell", "-d", language]
|
|
|
|
|
+ personal_dict_arg
|
|
|
|
|
+ ["-u3", str(tmpdir / po_file.name)],
|
|
|
|
|
universal_newlines=True,
|
|
|
|
|
)
|
|
|
|
|
except subprocess.CalledProcessError:
|
|
|
|
|
return -1
|
2018-07-27 13:54:10 +00:00
|
|
|
|
for line in output.split("\n"):
|
2018-07-31 22:20:03 +00:00
|
|
|
|
match = regex.match(
|
2018-07-27 13:54:10 +00:00
|
|
|
|
r"(?P<path>.*):(?P<line>[0-9]+): Locate: (?P<error>.*) \| Try: .*$",
|
|
|
|
|
line,
|
|
|
|
|
)
|
2018-07-23 17:24:10 +00:00
|
|
|
|
if match:
|
2018-07-23 22:28:13 +00:00
|
|
|
|
errors += 1
|
2018-07-31 16:55:58 +00:00
|
|
|
|
print(po_file, match.group("line"), match.group("error"), sep=":")
|
2018-12-24 14:49:54 +00:00
|
|
|
|
return errors
|
|
|
|
|
|
|
|
|
|
|
2019-12-10 14:10:17 +00:00
|
|
|
|
def gracefull_handling_of_missing_dicts(language):
|
|
|
|
|
"""Check if hunspell dictionary for given language is installed.
|
|
|
|
|
"""
|
|
|
|
|
hunspell_dash_d = subprocess.check_output(
|
|
|
|
|
["hunspell", "-D"], universal_newlines=True, stderr=subprocess.STDOUT
|
|
|
|
|
)
|
|
|
|
|
languages = {Path(line).name for line in hunspell_dash_d}
|
|
|
|
|
|
|
|
|
|
def error(*args, file=sys.stderr, **kwargs):
|
|
|
|
|
print(*args, file=file, **kwargs)
|
|
|
|
|
|
|
|
|
|
if language in languages:
|
|
|
|
|
return
|
|
|
|
|
error(
|
|
|
|
|
"The hunspell dictionary for your language is missing, please install it.",
|
|
|
|
|
end="\n\n",
|
|
|
|
|
)
|
|
|
|
|
if which("apt"):
|
|
|
|
|
error("Maybe try something like:")
|
|
|
|
|
error(" sudo apt install hunspell-{}".format(language))
|
|
|
|
|
else:
|
|
|
|
|
error(
|
|
|
|
|
"""I don't know your environment, but I bet the package name looks like:
|
|
|
|
|
|
|
|
|
|
hunspell-{language}
|
|
|
|
|
|
|
|
|
|
If you find it, please tell me (by opening an issue or a PR on
|
|
|
|
|
https://github.com/JulienPalard/pospell/) so I can enhance this error message.
|
|
|
|
|
""".format(
|
|
|
|
|
language=language
|
|
|
|
|
)
|
|
|
|
|
)
|
|
|
|
|
exit(1)
|
|
|
|
|
|
|
|
|
|
|
2018-12-24 14:49:54 +00:00
|
|
|
|
def main():
|
|
|
|
|
"""Module entry point.
|
|
|
|
|
"""
|
|
|
|
|
args = parse_args()
|
|
|
|
|
logging.basicConfig(level=50 - 10 * args.verbose)
|
2019-10-16 14:55:46 +00:00
|
|
|
|
default_drop_capitalized = DEFAULT_DROP_CAPITALIZED.get(args.language, False)
|
|
|
|
|
if args.drop_capitalized:
|
|
|
|
|
drop_capitalized = True
|
|
|
|
|
elif args.no_drop_capitalized:
|
|
|
|
|
drop_capitalized = False
|
|
|
|
|
else:
|
|
|
|
|
drop_capitalized = default_drop_capitalized
|
2018-12-24 14:49:54 +00:00
|
|
|
|
args.po_file = list(
|
|
|
|
|
chain(Path(".").glob(args.glob) if args.glob else [], args.po_file)
|
|
|
|
|
)
|
|
|
|
|
if args.modified:
|
|
|
|
|
git_status = subprocess.check_output(
|
|
|
|
|
["git", "status", "--porcelain"], encoding="utf-8"
|
|
|
|
|
)
|
|
|
|
|
git_status_lines = [
|
|
|
|
|
line.split(maxsplit=2) for line in git_status.split("\n") if line
|
|
|
|
|
]
|
|
|
|
|
args.po_file.extend(
|
|
|
|
|
Path(filename)
|
|
|
|
|
for status, filename in git_status_lines
|
|
|
|
|
if filename.endswith(".po")
|
|
|
|
|
)
|
2019-10-09 11:07:09 +00:00
|
|
|
|
errors = spell_check(
|
2019-10-16 14:55:46 +00:00
|
|
|
|
args.po_file, args.personal_dict, args.language, drop_capitalized, args.debug
|
2019-10-09 11:07:09 +00:00
|
|
|
|
)
|
2019-12-10 14:10:17 +00:00
|
|
|
|
if errors == -1:
|
|
|
|
|
gracefull_handling_of_missing_dicts(args.language)
|
2018-07-23 22:28:13 +00:00
|
|
|
|
exit(0 if errors == 0 else -1)
|
2018-07-23 15:37:50 +00:00
|
|
|
|
|
|
|
|
|
|
2018-07-27 13:54:10 +00:00
|
|
|
|
if __name__ == "__main__":
|
2018-07-23 15:37:50 +00:00
|
|
|
|
main()
|