1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408
|
"""Google-style docstring parsing."""
import inspect
import re
import typing as T
from collections import OrderedDict, namedtuple
from enum import IntEnum
from .common import (
EXAMPLES_KEYWORDS,
PARAM_KEYWORDS,
RAISES_KEYWORDS,
RETURNS_KEYWORDS,
YIELDS_KEYWORDS,
Docstring,
DocstringExample,
DocstringMeta,
DocstringParam,
DocstringRaises,
DocstringReturns,
DocstringStyle,
ParseError,
RenderingStyle,
)
class SectionType(IntEnum):
"""Types of sections."""
SINGULAR = 0
"""For sections like examples."""
MULTIPLE = 1
"""For sections like params."""
SINGULAR_OR_MULTIPLE = 2
"""For sections like returns or yields."""
class Section(namedtuple("SectionBase", "title key type")):
"""A docstring section."""
GOOGLE_TYPED_ARG_REGEX = re.compile(r"\s*(.+?)\s*\(\s*(.*[^\s]+)\s*\)")
GOOGLE_ARG_DESC_REGEX = re.compile(r".*\. Defaults to (.+)\.")
MULTIPLE_PATTERN = re.compile(r"(\s*[^:\s]+:)|([^:]*\]:.*)")
DEFAULT_SECTIONS = [
Section("Arguments", "param", SectionType.MULTIPLE),
Section("Args", "param", SectionType.MULTIPLE),
Section("Parameters", "param", SectionType.MULTIPLE),
Section("Params", "param", SectionType.MULTIPLE),
Section("Raises", "raises", SectionType.MULTIPLE),
Section("Exceptions", "raises", SectionType.MULTIPLE),
Section("Except", "raises", SectionType.MULTIPLE),
Section("Attributes", "attribute", SectionType.MULTIPLE),
Section("Example", "examples", SectionType.SINGULAR),
Section("Examples", "examples", SectionType.SINGULAR),
Section("Returns", "returns", SectionType.SINGULAR_OR_MULTIPLE),
Section("Yields", "yields", SectionType.SINGULAR_OR_MULTIPLE),
]
class GoogleParser:
"""Parser for Google-style docstrings."""
def __init__(
self, sections: T.Optional[T.List[Section]] = None, title_colon=True
):
"""Setup sections.
:param sections: Recognized sections or None to defaults.
:param title_colon: require colon after section title.
"""
if not sections:
sections = DEFAULT_SECTIONS
self.sections = {s.title: s for s in sections}
self.title_colon = title_colon
self._setup()
def _setup(self):
if self.title_colon:
colon = ":"
else:
colon = ""
self.titles_re = re.compile(
"^("
+ "|".join(f"({t})" for t in self.sections)
+ ")"
+ colon
+ "[ \t\r\f\v]*$",
flags=re.M,
)
def _build_meta(self, text: str, title: str) -> DocstringMeta:
"""Build docstring element.
:param text: docstring element text
:param title: title of section containing element
:return:
"""
section = self.sections[title]
if (
section.type == SectionType.SINGULAR_OR_MULTIPLE
and not MULTIPLE_PATTERN.match(text)
) or section.type == SectionType.SINGULAR:
return self._build_single_meta(section, text)
if ":" not in text:
raise ParseError(f"Expected a colon in {text!r}.")
# Split spec and description
before, desc = text.split(":", 1)
if desc:
desc = desc[1:] if desc[0] == " " else desc
if "\n" in desc:
first_line, rest = desc.split("\n", 1)
desc = first_line + "\n" + inspect.cleandoc(rest)
desc = desc.strip("\n")
return self._build_multi_meta(section, before, desc)
@staticmethod
def _build_single_meta(section: Section, desc: str) -> DocstringMeta:
if section.key in RETURNS_KEYWORDS | YIELDS_KEYWORDS:
return DocstringReturns(
args=[section.key],
description=desc,
type_name=None,
is_generator=section.key in YIELDS_KEYWORDS,
)
if section.key in RAISES_KEYWORDS:
return DocstringRaises(
args=[section.key], description=desc, type_name=None
)
if section.key in EXAMPLES_KEYWORDS:
return DocstringExample(
args=[section.key], snippet=None, description=desc
)
if section.key in PARAM_KEYWORDS:
raise ParseError("Expected paramenter name.")
return DocstringMeta(args=[section.key], description=desc)
@staticmethod
def _build_multi_meta(
section: Section, before: str, desc: str
) -> DocstringMeta:
if section.key in PARAM_KEYWORDS:
match = GOOGLE_TYPED_ARG_REGEX.match(before)
if match:
arg_name, type_name = match.group(1, 2)
if type_name.endswith(", optional"):
is_optional = True
type_name = type_name[:-10]
elif type_name.endswith("?"):
is_optional = True
type_name = type_name[:-1]
else:
is_optional = False
else:
arg_name, type_name = before, None
is_optional = None
match = GOOGLE_ARG_DESC_REGEX.match(desc)
default = match.group(1) if match else None
return DocstringParam(
args=[section.key, before],
description=desc,
arg_name=arg_name,
type_name=type_name,
is_optional=is_optional,
default=default,
)
if section.key in RETURNS_KEYWORDS | YIELDS_KEYWORDS:
return DocstringReturns(
args=[section.key, before],
description=desc,
type_name=before,
is_generator=section.key in YIELDS_KEYWORDS,
)
if section.key in RAISES_KEYWORDS:
return DocstringRaises(
args=[section.key, before], description=desc, type_name=before
)
return DocstringMeta(args=[section.key, before], description=desc)
def add_section(self, section: Section):
"""Add or replace a section.
:param section: The new section.
"""
self.sections[section.title] = section
self._setup()
def parse(self, text: str) -> Docstring:
"""Parse the Google-style docstring into its components.
:returns: parsed docstring
"""
ret = Docstring(style=DocstringStyle.GOOGLE)
if not text:
return ret
# Clean according to PEP-0257
text = inspect.cleandoc(text)
# Find first title and split on its position
match = self.titles_re.search(text)
if match:
desc_chunk = text[: match.start()]
meta_chunk = text[match.start() :]
else:
desc_chunk = text
meta_chunk = ""
# Break description into short and long parts
parts = desc_chunk.split("\n", 1)
ret.short_description = parts[0] or None
if len(parts) > 1:
long_desc_chunk = parts[1] or ""
ret.blank_after_short_description = long_desc_chunk.startswith(
"\n"
)
ret.blank_after_long_description = long_desc_chunk.endswith("\n\n")
ret.long_description = long_desc_chunk.strip() or None
# Split by sections determined by titles
matches = list(self.titles_re.finditer(meta_chunk))
if not matches:
return ret
splits = []
for j in range(len(matches) - 1):
splits.append((matches[j].end(), matches[j + 1].start()))
splits.append((matches[-1].end(), len(meta_chunk)))
chunks = OrderedDict() # type: T.Mapping[str,str]
for j, (start, end) in enumerate(splits):
title = matches[j].group(1)
if title not in self.sections:
continue
# Clear Any Unknown Meta
# Ref: https://github.com/rr-/docstring_parser/issues/29
meta_details = meta_chunk[start:end]
unknown_meta = re.search(r"\n\S", meta_details)
if unknown_meta is not None:
meta_details = meta_details[: unknown_meta.start()]
chunks[title] = meta_details.strip("\n")
if not chunks:
return ret
# Add elements from each chunk
for title, chunk in chunks.items():
# Determine indent
indent_match = re.search(r"^\s*", chunk)
if not indent_match:
raise ParseError(f'Can\'t infer indent from "{chunk}"')
indent = indent_match.group()
# Check for singular elements
if self.sections[title].type in [
SectionType.SINGULAR,
SectionType.SINGULAR_OR_MULTIPLE,
]:
part = inspect.cleandoc(chunk)
ret.meta.append(self._build_meta(part, title))
continue
# Split based on lines which have exactly that indent
_re = "^" + indent + r"(?=\S)"
c_matches = list(re.finditer(_re, chunk, flags=re.M))
if not c_matches:
raise ParseError(f'No specification for "{title}": "{chunk}"')
c_splits = []
for j in range(len(c_matches) - 1):
c_splits.append((c_matches[j].end(), c_matches[j + 1].start()))
c_splits.append((c_matches[-1].end(), len(chunk)))
for j, (start, end) in enumerate(c_splits):
part = chunk[start:end].strip("\n")
ret.meta.append(self._build_meta(part, title))
return ret
def parse(text: str) -> Docstring:
"""Parse the Google-style docstring into its components.
:returns: parsed docstring
"""
return GoogleParser().parse(text)
def compose(
docstring: Docstring,
rendering_style: RenderingStyle = RenderingStyle.COMPACT,
indent: str = " ",
) -> str:
"""Render a parsed docstring into docstring text.
:param docstring: parsed docstring representation
:param rendering_style: the style to render docstrings
:param indent: the characters used as indentation in the docstring string
:returns: docstring text
"""
def process_one(
one: T.Union[DocstringParam, DocstringReturns, DocstringRaises]
):
head = ""
if isinstance(one, DocstringParam):
head += one.arg_name or ""
elif isinstance(one, DocstringReturns):
head += one.return_name or ""
if isinstance(one, DocstringParam) and one.is_optional:
optional = (
"?"
if rendering_style == RenderingStyle.COMPACT
else ", optional"
)
else:
optional = ""
if one.type_name and head:
head += f" ({one.type_name}{optional}):"
elif one.type_name:
head += f"{one.type_name}{optional}:"
else:
head += ":"
head = indent + head
if one.description and rendering_style == RenderingStyle.EXPANDED:
body = f"\n{indent}{indent}".join(
[head] + one.description.splitlines()
)
parts.append(body)
elif one.description:
(first, *rest) = one.description.splitlines()
body = f"\n{indent}{indent}".join([head + " " + first] + rest)
parts.append(body)
else:
parts.append(head)
def process_sect(name: str, args: T.List[T.Any]):
if args:
parts.append(name)
for arg in args:
process_one(arg)
parts.append("")
parts: T.List[str] = []
if docstring.short_description:
parts.append(docstring.short_description)
if docstring.blank_after_short_description:
parts.append("")
if docstring.long_description:
parts.append(docstring.long_description)
if docstring.blank_after_long_description:
parts.append("")
process_sect(
"Args:", [p for p in docstring.params or [] if p.args[0] == "param"]
)
process_sect(
"Attributes:",
[p for p in docstring.params or [] if p.args[0] == "attribute"],
)
process_sect(
"Returns:",
[p for p in docstring.many_returns or [] if not p.is_generator],
)
process_sect(
"Yields:", [p for p in docstring.many_returns or [] if p.is_generator]
)
process_sect("Raises:", docstring.raises or [])
if docstring.returns and not docstring.many_returns:
ret = docstring.returns
parts.append("Yields:" if ret else "Returns:")
parts.append("-" * len(parts[-1]))
process_one(ret)
for meta in docstring.meta:
if isinstance(
meta, (DocstringParam, DocstringReturns, DocstringRaises)
):
continue # Already handled
parts.append(meta.args[0].replace("_", "").title() + ":")
if meta.description:
lines = [indent + l for l in meta.description.splitlines()]
parts.append("\n".join(lines))
parts.append("")
while parts and not parts[-1]:
parts.pop()
return "\n".join(parts)
|