File size: 40,615 Bytes
7885a28 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 |
"""Common IO api utilities"""
from __future__ import annotations
from abc import (
ABC,
abstractmethod,
)
import codecs
from collections import defaultdict
from collections.abc import (
Hashable,
Mapping,
Sequence,
)
import dataclasses
import functools
import gzip
from io import (
BufferedIOBase,
BytesIO,
RawIOBase,
StringIO,
TextIOBase,
TextIOWrapper,
)
import mmap
import os
from pathlib import Path
import re
import tarfile
from typing import (
IO,
TYPE_CHECKING,
Any,
AnyStr,
DefaultDict,
Generic,
Literal,
TypeVar,
cast,
overload,
)
from urllib.parse import (
urljoin,
urlparse as parse_url,
uses_netloc,
uses_params,
uses_relative,
)
import warnings
import zipfile
from pandas._typing import (
BaseBuffer,
ReadCsvBuffer,
)
from pandas.compat import (
get_bz2_file,
get_lzma_file,
)
from pandas.compat._optional import import_optional_dependency
from pandas.util._decorators import doc
from pandas.util._exceptions import find_stack_level
from pandas.core.dtypes.common import (
is_bool,
is_file_like,
is_integer,
is_list_like,
)
from pandas.core.dtypes.generic import ABCMultiIndex
from pandas.core.shared_docs import _shared_docs
_VALID_URLS = set(uses_relative + uses_netloc + uses_params)
_VALID_URLS.discard("")
_RFC_3986_PATTERN = re.compile(r"^[A-Za-z][A-Za-z0-9+\-+.]*://")
BaseBufferT = TypeVar("BaseBufferT", bound=BaseBuffer)
if TYPE_CHECKING:
from types import TracebackType
from pandas._typing import (
CompressionDict,
CompressionOptions,
FilePath,
ReadBuffer,
StorageOptions,
WriteBuffer,
)
from pandas import MultiIndex
@dataclasses.dataclass
class IOArgs:
"""
Return value of io/common.py:_get_filepath_or_buffer.
"""
filepath_or_buffer: str | BaseBuffer
encoding: str
mode: str
compression: CompressionDict
should_close: bool = False
@dataclasses.dataclass
class IOHandles(Generic[AnyStr]):
"""
Return value of io/common.py:get_handle
Can be used as a context manager.
This is used to easily close created buffers and to handle corner cases when
TextIOWrapper is inserted.
handle: The file handle to be used.
created_handles: All file handles that are created by get_handle
is_wrapped: Whether a TextIOWrapper needs to be detached.
"""
# handle might not implement the IO-interface
handle: IO[AnyStr]
compression: CompressionDict
created_handles: list[IO[bytes] | IO[str]] = dataclasses.field(default_factory=list)
is_wrapped: bool = False
def close(self) -> None:
"""
Close all created buffers.
Note: If a TextIOWrapper was inserted, it is flushed and detached to
avoid closing the potentially user-created buffer.
"""
if self.is_wrapped:
assert isinstance(self.handle, TextIOWrapper)
self.handle.flush()
self.handle.detach()
self.created_handles.remove(self.handle)
for handle in self.created_handles:
handle.close()
self.created_handles = []
self.is_wrapped = False
def __enter__(self) -> IOHandles[AnyStr]:
return self
def __exit__(
self,
exc_type: type[BaseException] | None,
exc_value: BaseException | None,
traceback: TracebackType | None,
) -> None:
self.close()
def is_url(url: object) -> bool:
"""
Check to see if a URL has a valid protocol.
Parameters
----------
url : str or unicode
Returns
-------
isurl : bool
If `url` has a valid protocol return True otherwise False.
"""
if not isinstance(url, str):
return False
return parse_url(url).scheme in _VALID_URLS
@overload
def _expand_user(filepath_or_buffer: str) -> str:
...
@overload
def _expand_user(filepath_or_buffer: BaseBufferT) -> BaseBufferT:
...
def _expand_user(filepath_or_buffer: str | BaseBufferT) -> str | BaseBufferT:
"""
Return the argument with an initial component of ~ or ~user
replaced by that user's home directory.
Parameters
----------
filepath_or_buffer : object to be converted if possible
Returns
-------
expanded_filepath_or_buffer : an expanded filepath or the
input if not expandable
"""
if isinstance(filepath_or_buffer, str):
return os.path.expanduser(filepath_or_buffer)
return filepath_or_buffer
def validate_header_arg(header: object) -> None:
if header is None:
return
if is_integer(header):
header = cast(int, header)
if header < 0:
# GH 27779
raise ValueError(
"Passing negative integer to header is invalid. "
"For no header, use header=None instead"
)
return
if is_list_like(header, allow_sets=False):
header = cast(Sequence, header)
if not all(map(is_integer, header)):
raise ValueError("header must be integer or list of integers")
if any(i < 0 for i in header):
raise ValueError("cannot specify multi-index header with negative integers")
return
if is_bool(header):
raise TypeError(
"Passing a bool to header is invalid. Use header=None for no header or "
"header=int or list-like of ints to specify "
"the row(s) making up the column names"
)
# GH 16338
raise ValueError("header must be integer or list of integers")
@overload
def stringify_path(filepath_or_buffer: FilePath, convert_file_like: bool = ...) -> str:
...
@overload
def stringify_path(
filepath_or_buffer: BaseBufferT, convert_file_like: bool = ...
) -> BaseBufferT:
...
def stringify_path(
filepath_or_buffer: FilePath | BaseBufferT,
convert_file_like: bool = False,
) -> str | BaseBufferT:
"""
Attempt to convert a path-like object to a string.
Parameters
----------
filepath_or_buffer : object to be converted
Returns
-------
str_filepath_or_buffer : maybe a string version of the object
Notes
-----
Objects supporting the fspath protocol are coerced
according to its __fspath__ method.
Any other object is passed through unchanged, which includes bytes,
strings, buffers, or anything else that's not even path-like.
"""
if not convert_file_like and is_file_like(filepath_or_buffer):
# GH 38125: some fsspec objects implement os.PathLike but have already opened a
# file. This prevents opening the file a second time. infer_compression calls
# this function with convert_file_like=True to infer the compression.
return cast(BaseBufferT, filepath_or_buffer)
if isinstance(filepath_or_buffer, os.PathLike):
filepath_or_buffer = filepath_or_buffer.__fspath__()
return _expand_user(filepath_or_buffer)
def urlopen(*args, **kwargs):
"""
Lazy-import wrapper for stdlib urlopen, as that imports a big chunk of
the stdlib.
"""
import urllib.request
return urllib.request.urlopen(*args, **kwargs)
def is_fsspec_url(url: FilePath | BaseBuffer) -> bool:
"""
Returns true if the given URL looks like
something fsspec can handle
"""
return (
isinstance(url, str)
and bool(_RFC_3986_PATTERN.match(url))
and not url.startswith(("http://", "https://"))
)
@doc(
storage_options=_shared_docs["storage_options"],
compression_options=_shared_docs["compression_options"] % "filepath_or_buffer",
)
def _get_filepath_or_buffer(
filepath_or_buffer: FilePath | BaseBuffer,
encoding: str = "utf-8",
compression: CompressionOptions | None = None,
mode: str = "r",
storage_options: StorageOptions | None = None,
) -> IOArgs:
"""
If the filepath_or_buffer is a url, translate and return the buffer.
Otherwise passthrough.
Parameters
----------
filepath_or_buffer : a url, filepath (str, py.path.local or pathlib.Path),
or buffer
{compression_options}
.. versionchanged:: 1.4.0 Zstandard support.
encoding : the encoding to use to decode bytes, default is 'utf-8'
mode : str, optional
{storage_options}
Returns the dataclass IOArgs.
"""
filepath_or_buffer = stringify_path(filepath_or_buffer)
# handle compression dict
compression_method, compression = get_compression_method(compression)
compression_method = infer_compression(filepath_or_buffer, compression_method)
# GH21227 internal compression is not used for non-binary handles.
if compression_method and hasattr(filepath_or_buffer, "write") and "b" not in mode:
warnings.warn(
"compression has no effect when passing a non-binary object as input.",
RuntimeWarning,
stacklevel=find_stack_level(),
)
compression_method = None
compression = dict(compression, method=compression_method)
# bz2 and xz do not write the byte order mark for utf-16 and utf-32
# print a warning when writing such files
if (
"w" in mode
and compression_method in ["bz2", "xz"]
and encoding in ["utf-16", "utf-32"]
):
warnings.warn(
f"{compression} will not write the byte order mark for {encoding}",
UnicodeWarning,
stacklevel=find_stack_level(),
)
# Use binary mode when converting path-like objects to file-like objects (fsspec)
# except when text mode is explicitly requested. The original mode is returned if
# fsspec is not used.
fsspec_mode = mode
if "t" not in fsspec_mode and "b" not in fsspec_mode:
fsspec_mode += "b"
if isinstance(filepath_or_buffer, str) and is_url(filepath_or_buffer):
# TODO: fsspec can also handle HTTP via requests, but leaving this
# unchanged. using fsspec appears to break the ability to infer if the
# server responded with gzipped data
storage_options = storage_options or {}
# waiting until now for importing to match intended lazy logic of
# urlopen function defined elsewhere in this module
import urllib.request
# assuming storage_options is to be interpreted as headers
req_info = urllib.request.Request(filepath_or_buffer, headers=storage_options)
with urlopen(req_info) as req:
content_encoding = req.headers.get("Content-Encoding", None)
if content_encoding == "gzip":
# Override compression based on Content-Encoding header
compression = {"method": "gzip"}
reader = BytesIO(req.read())
return IOArgs(
filepath_or_buffer=reader,
encoding=encoding,
compression=compression,
should_close=True,
mode=fsspec_mode,
)
if is_fsspec_url(filepath_or_buffer):
assert isinstance(
filepath_or_buffer, str
) # just to appease mypy for this branch
# two special-case s3-like protocols; these have special meaning in Hadoop,
# but are equivalent to just "s3" from fsspec's point of view
# cc #11071
if filepath_or_buffer.startswith("s3a://"):
filepath_or_buffer = filepath_or_buffer.replace("s3a://", "s3://")
if filepath_or_buffer.startswith("s3n://"):
filepath_or_buffer = filepath_or_buffer.replace("s3n://", "s3://")
fsspec = import_optional_dependency("fsspec")
# If botocore is installed we fallback to reading with anon=True
# to allow reads from public buckets
err_types_to_retry_with_anon: list[Any] = []
try:
import_optional_dependency("botocore")
from botocore.exceptions import (
ClientError,
NoCredentialsError,
)
err_types_to_retry_with_anon = [
ClientError,
NoCredentialsError,
PermissionError,
]
except ImportError:
pass
try:
file_obj = fsspec.open(
filepath_or_buffer, mode=fsspec_mode, **(storage_options or {})
).open()
# GH 34626 Reads from Public Buckets without Credentials needs anon=True
except tuple(err_types_to_retry_with_anon):
if storage_options is None:
storage_options = {"anon": True}
else:
# don't mutate user input.
storage_options = dict(storage_options)
storage_options["anon"] = True
file_obj = fsspec.open(
filepath_or_buffer, mode=fsspec_mode, **(storage_options or {})
).open()
return IOArgs(
filepath_or_buffer=file_obj,
encoding=encoding,
compression=compression,
should_close=True,
mode=fsspec_mode,
)
elif storage_options:
raise ValueError(
"storage_options passed with file object or non-fsspec file path"
)
if isinstance(filepath_or_buffer, (str, bytes, mmap.mmap)):
return IOArgs(
filepath_or_buffer=_expand_user(filepath_or_buffer),
encoding=encoding,
compression=compression,
should_close=False,
mode=mode,
)
# is_file_like requires (read | write) & __iter__ but __iter__ is only
# needed for read_csv(engine=python)
if not (
hasattr(filepath_or_buffer, "read") or hasattr(filepath_or_buffer, "write")
):
msg = f"Invalid file path or buffer object type: {type(filepath_or_buffer)}"
raise ValueError(msg)
return IOArgs(
filepath_or_buffer=filepath_or_buffer,
encoding=encoding,
compression=compression,
should_close=False,
mode=mode,
)
def file_path_to_url(path: str) -> str:
"""
converts an absolute native path to a FILE URL.
Parameters
----------
path : a path in native format
Returns
-------
a valid FILE URL
"""
# lazify expensive import (~30ms)
from urllib.request import pathname2url
return urljoin("file:", pathname2url(path))
extension_to_compression = {
".tar": "tar",
".tar.gz": "tar",
".tar.bz2": "tar",
".tar.xz": "tar",
".gz": "gzip",
".bz2": "bz2",
".zip": "zip",
".xz": "xz",
".zst": "zstd",
}
_supported_compressions = set(extension_to_compression.values())
def get_compression_method(
compression: CompressionOptions,
) -> tuple[str | None, CompressionDict]:
"""
Simplifies a compression argument to a compression method string and
a mapping containing additional arguments.
Parameters
----------
compression : str or mapping
If string, specifies the compression method. If mapping, value at key
'method' specifies compression method.
Returns
-------
tuple of ({compression method}, Optional[str]
{compression arguments}, Dict[str, Any])
Raises
------
ValueError on mapping missing 'method' key
"""
compression_method: str | None
if isinstance(compression, Mapping):
compression_args = dict(compression)
try:
compression_method = compression_args.pop("method")
except KeyError as err:
raise ValueError("If mapping, compression must have key 'method'") from err
else:
compression_args = {}
compression_method = compression
return compression_method, compression_args
@doc(compression_options=_shared_docs["compression_options"] % "filepath_or_buffer")
def infer_compression(
filepath_or_buffer: FilePath | BaseBuffer, compression: str | None
) -> str | None:
"""
Get the compression method for filepath_or_buffer. If compression='infer',
the inferred compression method is returned. Otherwise, the input
compression method is returned unchanged, unless it's invalid, in which
case an error is raised.
Parameters
----------
filepath_or_buffer : str or file handle
File path or object.
{compression_options}
.. versionchanged:: 1.4.0 Zstandard support.
Returns
-------
string or None
Raises
------
ValueError on invalid compression specified.
"""
if compression is None:
return None
# Infer compression
if compression == "infer":
# Convert all path types (e.g. pathlib.Path) to strings
filepath_or_buffer = stringify_path(filepath_or_buffer, convert_file_like=True)
if not isinstance(filepath_or_buffer, str):
# Cannot infer compression of a buffer, assume no compression
return None
# Infer compression from the filename/URL extension
for extension, compression in extension_to_compression.items():
if filepath_or_buffer.lower().endswith(extension):
return compression
return None
# Compression has been specified. Check that it's valid
if compression in _supported_compressions:
return compression
valid = ["infer", None] + sorted(_supported_compressions)
msg = (
f"Unrecognized compression type: {compression}\n"
f"Valid compression types are {valid}"
)
raise ValueError(msg)
def check_parent_directory(path: Path | str) -> None:
"""
Check if parent directory of a file exists, raise OSError if it does not
Parameters
----------
path: Path or str
Path to check parent directory of
"""
parent = Path(path).parent
if not parent.is_dir():
raise OSError(rf"Cannot save file into a non-existent directory: '{parent}'")
@overload
def get_handle(
path_or_buf: FilePath | BaseBuffer,
mode: str,
*,
encoding: str | None = ...,
compression: CompressionOptions = ...,
memory_map: bool = ...,
is_text: Literal[False],
errors: str | None = ...,
storage_options: StorageOptions = ...,
) -> IOHandles[bytes]:
...
@overload
def get_handle(
path_or_buf: FilePath | BaseBuffer,
mode: str,
*,
encoding: str | None = ...,
compression: CompressionOptions = ...,
memory_map: bool = ...,
is_text: Literal[True] = ...,
errors: str | None = ...,
storage_options: StorageOptions = ...,
) -> IOHandles[str]:
...
@overload
def get_handle(
path_or_buf: FilePath | BaseBuffer,
mode: str,
*,
encoding: str | None = ...,
compression: CompressionOptions = ...,
memory_map: bool = ...,
is_text: bool = ...,
errors: str | None = ...,
storage_options: StorageOptions = ...,
) -> IOHandles[str] | IOHandles[bytes]:
...
@doc(compression_options=_shared_docs["compression_options"] % "path_or_buf")
def get_handle(
path_or_buf: FilePath | BaseBuffer,
mode: str,
*,
encoding: str | None = None,
compression: CompressionOptions | None = None,
memory_map: bool = False,
is_text: bool = True,
errors: str | None = None,
storage_options: StorageOptions | None = None,
) -> IOHandles[str] | IOHandles[bytes]:
"""
Get file handle for given path/buffer and mode.
Parameters
----------
path_or_buf : str or file handle
File path or object.
mode : str
Mode to open path_or_buf with.
encoding : str or None
Encoding to use.
{compression_options}
May be a dict with key 'method' as compression mode
and other keys as compression options if compression
mode is 'zip'.
Passing compression options as keys in dict is
supported for compression modes 'gzip', 'bz2', 'zstd' and 'zip'.
.. versionchanged:: 1.4.0 Zstandard support.
memory_map : bool, default False
See parsers._parser_params for more information. Only used by read_csv.
is_text : bool, default True
Whether the type of the content passed to the file/buffer is string or
bytes. This is not the same as `"b" not in mode`. If a string content is
passed to a binary file/buffer, a wrapper is inserted.
errors : str, default 'strict'
Specifies how encoding and decoding errors are to be handled.
See the errors argument for :func:`open` for a full list
of options.
storage_options: StorageOptions = None
Passed to _get_filepath_or_buffer
Returns the dataclass IOHandles
"""
# Windows does not default to utf-8. Set to utf-8 for a consistent behavior
encoding = encoding or "utf-8"
errors = errors or "strict"
# read_csv does not know whether the buffer is opened in binary/text mode
if _is_binary_mode(path_or_buf, mode) and "b" not in mode:
mode += "b"
# validate encoding and errors
codecs.lookup(encoding)
if isinstance(errors, str):
codecs.lookup_error(errors)
# open URLs
ioargs = _get_filepath_or_buffer(
path_or_buf,
encoding=encoding,
compression=compression,
mode=mode,
storage_options=storage_options,
)
handle = ioargs.filepath_or_buffer
handles: list[BaseBuffer]
# memory mapping needs to be the first step
# only used for read_csv
handle, memory_map, handles = _maybe_memory_map(handle, memory_map)
is_path = isinstance(handle, str)
compression_args = dict(ioargs.compression)
compression = compression_args.pop("method")
# Only for write methods
if "r" not in mode and is_path:
check_parent_directory(str(handle))
if compression:
if compression != "zstd":
# compression libraries do not like an explicit text-mode
ioargs.mode = ioargs.mode.replace("t", "")
elif compression == "zstd" and "b" not in ioargs.mode:
# python-zstandard defaults to text mode, but we always expect
# compression libraries to use binary mode.
ioargs.mode += "b"
# GZ Compression
if compression == "gzip":
if isinstance(handle, str):
# error: Incompatible types in assignment (expression has type
# "GzipFile", variable has type "Union[str, BaseBuffer]")
handle = gzip.GzipFile( # type: ignore[assignment]
filename=handle,
mode=ioargs.mode,
**compression_args,
)
else:
handle = gzip.GzipFile(
# No overload variant of "GzipFile" matches argument types
# "Union[str, BaseBuffer]", "str", "Dict[str, Any]"
fileobj=handle, # type: ignore[call-overload]
mode=ioargs.mode,
**compression_args,
)
# BZ Compression
elif compression == "bz2":
# Overload of "BZ2File" to handle pickle protocol 5
# "Union[str, BaseBuffer]", "str", "Dict[str, Any]"
handle = get_bz2_file()( # type: ignore[call-overload]
handle,
mode=ioargs.mode,
**compression_args,
)
# ZIP Compression
elif compression == "zip":
# error: Argument 1 to "_BytesZipFile" has incompatible type
# "Union[str, BaseBuffer]"; expected "Union[Union[str, PathLike[str]],
# ReadBuffer[bytes], WriteBuffer[bytes]]"
handle = _BytesZipFile(
handle, ioargs.mode, **compression_args # type: ignore[arg-type]
)
if handle.buffer.mode == "r":
handles.append(handle)
zip_names = handle.buffer.namelist()
if len(zip_names) == 1:
handle = handle.buffer.open(zip_names.pop())
elif not zip_names:
raise ValueError(f"Zero files found in ZIP file {path_or_buf}")
else:
raise ValueError(
"Multiple files found in ZIP file. "
f"Only one file per ZIP: {zip_names}"
)
# TAR Encoding
elif compression == "tar":
compression_args.setdefault("mode", ioargs.mode)
if isinstance(handle, str):
handle = _BytesTarFile(name=handle, **compression_args)
else:
# error: Argument "fileobj" to "_BytesTarFile" has incompatible
# type "BaseBuffer"; expected "Union[ReadBuffer[bytes],
# WriteBuffer[bytes], None]"
handle = _BytesTarFile(
fileobj=handle, **compression_args # type: ignore[arg-type]
)
assert isinstance(handle, _BytesTarFile)
if "r" in handle.buffer.mode:
handles.append(handle)
files = handle.buffer.getnames()
if len(files) == 1:
file = handle.buffer.extractfile(files[0])
assert file is not None
handle = file
elif not files:
raise ValueError(f"Zero files found in TAR archive {path_or_buf}")
else:
raise ValueError(
"Multiple files found in TAR archive. "
f"Only one file per TAR archive: {files}"
)
# XZ Compression
elif compression == "xz":
# error: Argument 1 to "LZMAFile" has incompatible type "Union[str,
# BaseBuffer]"; expected "Optional[Union[Union[str, bytes, PathLike[str],
# PathLike[bytes]], IO[bytes]], None]"
handle = get_lzma_file()(
handle, ioargs.mode, **compression_args # type: ignore[arg-type]
)
# Zstd Compression
elif compression == "zstd":
zstd = import_optional_dependency("zstandard")
if "r" in ioargs.mode:
open_args = {"dctx": zstd.ZstdDecompressor(**compression_args)}
else:
open_args = {"cctx": zstd.ZstdCompressor(**compression_args)}
handle = zstd.open(
handle,
mode=ioargs.mode,
**open_args,
)
# Unrecognized Compression
else:
msg = f"Unrecognized compression type: {compression}"
raise ValueError(msg)
assert not isinstance(handle, str)
handles.append(handle)
elif isinstance(handle, str):
# Check whether the filename is to be opened in binary mode.
# Binary mode does not support 'encoding' and 'newline'.
if ioargs.encoding and "b" not in ioargs.mode:
# Encoding
handle = open(
handle,
ioargs.mode,
encoding=ioargs.encoding,
errors=errors,
newline="",
)
else:
# Binary mode
handle = open(handle, ioargs.mode)
handles.append(handle)
# Convert BytesIO or file objects passed with an encoding
is_wrapped = False
if not is_text and ioargs.mode == "rb" and isinstance(handle, TextIOBase):
# not added to handles as it does not open/buffer resources
handle = _BytesIOWrapper(
handle,
encoding=ioargs.encoding,
)
elif is_text and (
compression or memory_map or _is_binary_mode(handle, ioargs.mode)
):
if (
not hasattr(handle, "readable")
or not hasattr(handle, "writable")
or not hasattr(handle, "seekable")
):
handle = _IOWrapper(handle)
# error: Argument 1 to "TextIOWrapper" has incompatible type
# "_IOWrapper"; expected "IO[bytes]"
handle = TextIOWrapper(
handle, # type: ignore[arg-type]
encoding=ioargs.encoding,
errors=errors,
newline="",
)
handles.append(handle)
# only marked as wrapped when the caller provided a handle
is_wrapped = not (
isinstance(ioargs.filepath_or_buffer, str) or ioargs.should_close
)
if "r" in ioargs.mode and not hasattr(handle, "read"):
raise TypeError(
"Expected file path name or file-like object, "
f"got {type(ioargs.filepath_or_buffer)} type"
)
handles.reverse() # close the most recently added buffer first
if ioargs.should_close:
assert not isinstance(ioargs.filepath_or_buffer, str)
handles.append(ioargs.filepath_or_buffer)
return IOHandles(
# error: Argument "handle" to "IOHandles" has incompatible type
# "Union[TextIOWrapper, GzipFile, BaseBuffer, typing.IO[bytes],
# typing.IO[Any]]"; expected "pandas._typing.IO[Any]"
handle=handle, # type: ignore[arg-type]
# error: Argument "created_handles" to "IOHandles" has incompatible type
# "List[BaseBuffer]"; expected "List[Union[IO[bytes], IO[str]]]"
created_handles=handles, # type: ignore[arg-type]
is_wrapped=is_wrapped,
compression=ioargs.compression,
)
# error: Definition of "__enter__" in base class "IOBase" is incompatible
# with definition in base class "BinaryIO"
class _BufferedWriter(BytesIO, ABC): # type: ignore[misc]
"""
Some objects do not support multiple .write() calls (TarFile and ZipFile).
This wrapper writes to the underlying buffer on close.
"""
buffer = BytesIO()
@abstractmethod
def write_to_buffer(self) -> None:
...
def close(self) -> None:
if self.closed:
# already closed
return
if self.getbuffer().nbytes:
# write to buffer
self.seek(0)
with self.buffer:
self.write_to_buffer()
else:
self.buffer.close()
super().close()
class _BytesTarFile(_BufferedWriter):
def __init__(
self,
name: str | None = None,
mode: Literal["r", "a", "w", "x"] = "r",
fileobj: ReadBuffer[bytes] | WriteBuffer[bytes] | None = None,
archive_name: str | None = None,
**kwargs,
) -> None:
super().__init__()
self.archive_name = archive_name
self.name = name
# error: Incompatible types in assignment (expression has type "TarFile",
# base class "_BufferedWriter" defined the type as "BytesIO")
self.buffer: tarfile.TarFile = tarfile.TarFile.open( # type: ignore[assignment]
name=name,
mode=self.extend_mode(mode),
fileobj=fileobj,
**kwargs,
)
def extend_mode(self, mode: str) -> str:
mode = mode.replace("b", "")
if mode != "w":
return mode
if self.name is not None:
suffix = Path(self.name).suffix
if suffix in (".gz", ".xz", ".bz2"):
mode = f"{mode}:{suffix[1:]}"
return mode
def infer_filename(self) -> str | None:
"""
If an explicit archive_name is not given, we still want the file inside the zip
file not to be named something.tar, because that causes confusion (GH39465).
"""
if self.name is None:
return None
filename = Path(self.name)
if filename.suffix == ".tar":
return filename.with_suffix("").name
elif filename.suffix in (".tar.gz", ".tar.bz2", ".tar.xz"):
return filename.with_suffix("").with_suffix("").name
return filename.name
def write_to_buffer(self) -> None:
# TarFile needs a non-empty string
archive_name = self.archive_name or self.infer_filename() or "tar"
tarinfo = tarfile.TarInfo(name=archive_name)
tarinfo.size = len(self.getvalue())
self.buffer.addfile(tarinfo, self)
class _BytesZipFile(_BufferedWriter):
def __init__(
self,
file: FilePath | ReadBuffer[bytes] | WriteBuffer[bytes],
mode: str,
archive_name: str | None = None,
**kwargs,
) -> None:
super().__init__()
mode = mode.replace("b", "")
self.archive_name = archive_name
kwargs.setdefault("compression", zipfile.ZIP_DEFLATED)
# error: Incompatible types in assignment (expression has type "ZipFile",
# base class "_BufferedWriter" defined the type as "BytesIO")
self.buffer: zipfile.ZipFile = zipfile.ZipFile( # type: ignore[assignment]
file, mode, **kwargs
)
def infer_filename(self) -> str | None:
"""
If an explicit archive_name is not given, we still want the file inside the zip
file not to be named something.zip, because that causes confusion (GH39465).
"""
if isinstance(self.buffer.filename, (os.PathLike, str)):
filename = Path(self.buffer.filename)
if filename.suffix == ".zip":
return filename.with_suffix("").name
return filename.name
return None
def write_to_buffer(self) -> None:
# ZipFile needs a non-empty string
archive_name = self.archive_name or self.infer_filename() or "zip"
self.buffer.writestr(archive_name, self.getvalue())
class _IOWrapper:
# TextIOWrapper is overly strict: it request that the buffer has seekable, readable,
# and writable. If we have a read-only buffer, we shouldn't need writable and vice
# versa. Some buffers, are seek/read/writ-able but they do not have the "-able"
# methods, e.g., tempfile.SpooledTemporaryFile.
# If a buffer does not have the above "-able" methods, we simple assume they are
# seek/read/writ-able.
def __init__(self, buffer: BaseBuffer) -> None:
self.buffer = buffer
def __getattr__(self, name: str):
return getattr(self.buffer, name)
def readable(self) -> bool:
if hasattr(self.buffer, "readable"):
return self.buffer.readable()
return True
def seekable(self) -> bool:
if hasattr(self.buffer, "seekable"):
return self.buffer.seekable()
return True
def writable(self) -> bool:
if hasattr(self.buffer, "writable"):
return self.buffer.writable()
return True
class _BytesIOWrapper:
# Wrapper that wraps a StringIO buffer and reads bytes from it
# Created for compat with pyarrow read_csv
def __init__(self, buffer: StringIO | TextIOBase, encoding: str = "utf-8") -> None:
self.buffer = buffer
self.encoding = encoding
# Because a character can be represented by more than 1 byte,
# it is possible that reading will produce more bytes than n
# We store the extra bytes in this overflow variable, and append the
# overflow to the front of the bytestring the next time reading is performed
self.overflow = b""
def __getattr__(self, attr: str):
return getattr(self.buffer, attr)
def read(self, n: int | None = -1) -> bytes:
assert self.buffer is not None
bytestring = self.buffer.read(n).encode(self.encoding)
# When n=-1/n greater than remaining bytes: Read entire file/rest of file
combined_bytestring = self.overflow + bytestring
if n is None or n < 0 or n >= len(combined_bytestring):
self.overflow = b""
return combined_bytestring
else:
to_return = combined_bytestring[:n]
self.overflow = combined_bytestring[n:]
return to_return
def _maybe_memory_map(
handle: str | BaseBuffer, memory_map: bool
) -> tuple[str | BaseBuffer, bool, list[BaseBuffer]]:
"""Try to memory map file/buffer."""
handles: list[BaseBuffer] = []
memory_map &= hasattr(handle, "fileno") or isinstance(handle, str)
if not memory_map:
return handle, memory_map, handles
# mmap used by only read_csv
handle = cast(ReadCsvBuffer, handle)
# need to open the file first
if isinstance(handle, str):
handle = open(handle, "rb")
handles.append(handle)
try:
# open mmap and adds *-able
# error: Argument 1 to "_IOWrapper" has incompatible type "mmap";
# expected "BaseBuffer"
wrapped = _IOWrapper(
mmap.mmap(
handle.fileno(), 0, access=mmap.ACCESS_READ # type: ignore[arg-type]
)
)
finally:
for handle in reversed(handles):
# error: "BaseBuffer" has no attribute "close"
handle.close() # type: ignore[attr-defined]
return wrapped, memory_map, [wrapped]
def file_exists(filepath_or_buffer: FilePath | BaseBuffer) -> bool:
"""Test whether file exists."""
exists = False
filepath_or_buffer = stringify_path(filepath_or_buffer)
if not isinstance(filepath_or_buffer, str):
return exists
try:
exists = os.path.exists(filepath_or_buffer)
# gh-5874: if the filepath is too long will raise here
except (TypeError, ValueError):
pass
return exists
def _is_binary_mode(handle: FilePath | BaseBuffer, mode: str) -> bool:
"""Whether the handle is opened in binary mode"""
# specified by user
if "t" in mode or "b" in mode:
return "b" in mode
# exceptions
text_classes = (
# classes that expect string but have 'b' in mode
codecs.StreamWriter,
codecs.StreamReader,
codecs.StreamReaderWriter,
)
if issubclass(type(handle), text_classes):
return False
return isinstance(handle, _get_binary_io_classes()) or "b" in getattr(
handle, "mode", mode
)
@functools.lru_cache
def _get_binary_io_classes() -> tuple[type, ...]:
"""IO classes that that expect bytes"""
binary_classes: tuple[type, ...] = (BufferedIOBase, RawIOBase)
# python-zstandard doesn't use any of the builtin base classes; instead we
# have to use the `zstd.ZstdDecompressionReader` class for isinstance checks.
# Unfortunately `zstd.ZstdDecompressionReader` isn't exposed by python-zstandard
# so we have to get it from a `zstd.ZstdDecompressor` instance.
# See also https://github.com/indygreg/python-zstandard/pull/165.
zstd = import_optional_dependency("zstandard", errors="ignore")
if zstd is not None:
with zstd.ZstdDecompressor().stream_reader(b"") as reader:
binary_classes += (type(reader),)
return binary_classes
def is_potential_multi_index(
columns: Sequence[Hashable] | MultiIndex,
index_col: bool | Sequence[int] | None = None,
) -> bool:
"""
Check whether or not the `columns` parameter
could be converted into a MultiIndex.
Parameters
----------
columns : array-like
Object which may or may not be convertible into a MultiIndex
index_col : None, bool or list, optional
Column or columns to use as the (possibly hierarchical) index
Returns
-------
bool : Whether or not columns could become a MultiIndex
"""
if index_col is None or isinstance(index_col, bool):
index_col = []
return bool(
len(columns)
and not isinstance(columns, ABCMultiIndex)
and all(isinstance(c, tuple) for c in columns if c not in list(index_col))
)
def dedup_names(
names: Sequence[Hashable], is_potential_multiindex: bool
) -> Sequence[Hashable]:
"""
Rename column names if duplicates exist.
Currently the renaming is done by appending a period and an autonumeric,
but a custom pattern may be supported in the future.
Examples
--------
>>> dedup_names(["x", "y", "x", "x"], is_potential_multiindex=False)
['x', 'y', 'x.1', 'x.2']
"""
names = list(names) # so we can index
counts: DefaultDict[Hashable, int] = defaultdict(int)
for i, col in enumerate(names):
cur_count = counts[col]
while cur_count > 0:
counts[col] = cur_count + 1
if is_potential_multiindex:
# for mypy
assert isinstance(col, tuple)
col = col[:-1] + (f"{col[-1]}.{cur_count}",)
else:
col = f"{col}.{cur_count}"
cur_count = counts[col]
names[i] = col
counts[col] = cur_count + 1
return names
|