2 # -*- coding: utf-8 -*-
4 """The MIT License (MIT)
6 Copyright (c) 2016-2020 Davide Zanotti
8 Modifications Copyright (c) 2021-2022 Scott Gasch
10 Permission is hereby granted, free of charge, to any person obtaining a copy
11 of this software and associated documentation files (the "Software"), to deal
12 in the Software without restriction, including without limitation the rights
13 to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
14 copies of the Software, and to permit persons to whom the Software is
15 furnished to do so, subject to the following conditions:
17 The above copyright notice and this permission notice shall be included in all
18 copies or substantial portions of the Software.
20 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
21 IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
22 FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
23 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
24 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
25 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
28 This class is based on:
29 https://github.com/daveoncode/python-string-utils. See `NOTICE
30 <https://wannabe.guru.org/gitweb/?p=pyutils.git;a=blob_plain;f=NOTICE;hb=HEAD>`__
31 in the root of this module for a detailed enumeration of what work is
32 Davide's and what work was added by Scott.
37 import contextlib # type: ignore
48 from itertools import zip_longest
60 from uuid import uuid4
62 from pyutils import list_utils
64 logger = logging.getLogger(__name__)
66 NUMBER_RE = re.compile(r"^([+\-]?)((\d+)(\.\d+)?([e|E]\d+)?|\.\d+)$")
68 HEX_NUMBER_RE = re.compile(r"^([+|-]?)0[x|X]([0-9A-Fa-f]+)$")
70 OCT_NUMBER_RE = re.compile(r"^([+|-]?)0[O|o]([0-7]+)$")
72 BIN_NUMBER_RE = re.compile(r"^([+|-]?)0[B|b]([0|1]+)$")
75 r"([a-z-]+://)" # scheme
76 r"([a-z_\d-]+:[a-z_\d-]+@)?" # user:password
78 r"((?<!\.)[a-z\d]+[a-z\d.-]+\.[a-z]{2,6}|\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}|localhost)" # domain
79 r"(:\d{2,})?" # port number
80 r"(/[a-z\d_%+-]*)*" # folders
81 r"(\.[a-z\d_%+-]+)*" # file extension
82 r"(\?[a-z\d_+%-=]*)?" # query string
86 URL_RE = re.compile(rf"^{URLS_RAW_STRING}$", re.IGNORECASE)
88 URLS_RE = re.compile(rf"({URLS_RAW_STRING})", re.IGNORECASE)
90 ESCAPED_AT_SIGN = re.compile(r'(?!"[^"]*)@+(?=[^"]*")|\\@')
93 r"[a-zA-Z\d._\+\-'`!%#$&*/=\?\^\{\}\|~\\]+@[a-z\d-]+\.?[a-z\d-]+\.[a-z]{2,4}"
96 EMAIL_RE = re.compile(rf"^{EMAILS_RAW_STRING}$")
98 EMAILS_RE = re.compile(rf"({EMAILS_RAW_STRING})")
100 CAMEL_CASE_TEST_RE = re.compile(r"^[a-zA-Z]*([a-z]+[A-Z]+|[A-Z]+[a-z]+)[a-zA-Z\d]*$")
102 CAMEL_CASE_REPLACE_RE = re.compile(r"([a-z]|[A-Z]+)(?=[A-Z])")
104 SNAKE_CASE_TEST_RE = re.compile(
105 r"^([a-z]+\d*_[a-z\d_]*|_+[a-z\d]+[a-z\d_]*)$", re.IGNORECASE
108 SNAKE_CASE_TEST_DASH_RE = re.compile(
109 r"([a-z]+\d*-[a-z\d-]*|-+[a-z\d]+[a-z\d-]*)$", re.IGNORECASE
112 SNAKE_CASE_REPLACE_RE = re.compile(r"(_)([a-z\d])")
114 SNAKE_CASE_REPLACE_DASH_RE = re.compile(r"(-)([a-z\d])")
117 "VISA": re.compile(r"^4\d{12}(?:\d{3})?$"),
118 "MASTERCARD": re.compile(r"^5[1-5]\d{14}$"),
119 "AMERICAN_EXPRESS": re.compile(r"^3[47]\d{13}$"),
120 "DINERS_CLUB": re.compile(r"^3(?:0[0-5]|[68]\d)\d{11}$"),
121 "DISCOVER": re.compile(r"^6(?:011|5\d{2})\d{12}$"),
122 "JCB": re.compile(r"^(?:2131|1800|35\d{3})\d{11}$"),
125 JSON_WRAPPER_RE = re.compile(r"^\s*[\[{]\s*(.*)\s*[\}\]]\s*$", re.MULTILINE | re.DOTALL)
127 UUID_RE = re.compile(
128 r"^[a-f\d]{8}-[a-f\d]{4}-[a-f\d]{4}-[a-f\d]{4}-[a-f\d]{12}$", re.IGNORECASE
131 UUID_HEX_OK_RE = re.compile(
132 r"^[a-f\d]{8}-?[a-f\d]{4}-?[a-f\d]{4}-?[a-f\d]{4}-?[a-f\d]{12}$",
136 SHALLOW_IP_V4_RE = re.compile(r"^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$")
138 ANYWHERE_IP_V4_RE = re.compile(r"\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}")
140 IP_V6_RE = re.compile(r"^([a-z\d]{0,4}:){7}[a-z\d]{0,4}$", re.IGNORECASE)
142 ANYWHERE_IP_V6_RE = re.compile(r"([a-z\d]{0,4}:){7}[a-z\d]{0,4}", re.IGNORECASE)
144 MAC_ADDRESS_RE = re.compile(r"^([0-9A-F]{2}[:-]){5}([0-9A-F]{2})$", re.IGNORECASE)
146 ANYWHERE_MAC_ADDRESS_RE = re.compile(
147 r"([0-9A-F]{2}[:-]){5}([0-9A-F]{2})", re.IGNORECASE
150 WORDS_COUNT_RE = re.compile(r"\W*[^\W_]+\W*", re.IGNORECASE | re.MULTILINE | re.UNICODE)
152 HTML_RE = re.compile(
153 r"((<([a-z]+:)?[a-z]+[^>]*/?>)(.*?(</([a-z]+:)?[a-z]+>))?|<!--.*-->|<!doctype.*>)",
154 re.IGNORECASE | re.MULTILINE | re.DOTALL,
157 HTML_TAG_ONLY_RE = re.compile(
158 r"(<([a-z]+:)?[a-z]+[^>]*/?>|</([a-z]+:)?[a-z]+>|<!--.*-->|<!doctype.*>)",
159 re.IGNORECASE | re.MULTILINE | re.DOTALL,
162 SPACES_RE = re.compile(r"\s")
164 NO_LETTERS_OR_NUMBERS_RE = re.compile(r"[^\w\d]+|_+", re.IGNORECASE | re.UNICODE)
166 MARGIN_RE = re.compile(r"^[^\S\r\n]+")
168 ESCAPE_SEQUENCE_RE = re.compile(r"\x1B\[[^A-Za-z]*[A-Za-z]")
229 NUM_WORDS["and"] = (1, 0)
230 for i, word in enumerate(UNIT_WORDS):
231 NUM_WORDS[word] = (1, i)
232 for i, word in enumerate(TENS_WORDS):
233 NUM_WORDS[word] = (1, i * 10)
234 for i, word in enumerate(MAGNITUDE_SCALES):
236 NUM_WORDS[word] = (100, 0)
238 NUM_WORDS[word] = (10 ** (i * 3), 0)
239 NUM_WORDS['score'] = (20, 0)
242 def is_none_or_empty(in_str: Optional[str]) -> bool:
245 in_str: the string to test
248 True if the input string is either None or an empty string,
251 See also :meth:`is_string` and :meth:`is_empty_string`.
253 >>> is_none_or_empty("")
255 >>> is_none_or_empty(None)
257 >>> is_none_or_empty(" \t ")
259 >>> is_none_or_empty('Test')
262 return in_str is None or len(in_str.strip()) == 0
265 def is_string(in_str: Any) -> bool:
268 in_str: the object to test
271 True if the object is a string and False otherwise.
273 See also :meth:`is_empty_string`, :meth:`is_none_or_empty`.
275 >>> is_string('test')
281 >>> is_string([1, 2, 3])
284 return isinstance(in_str, str)
287 def is_empty_string(in_str: Any) -> bool:
290 in_str: the string to test
293 True if the string is empty and False otherwise.
295 See also :meth:`is_none_or_empty`, :meth:`is_full_string`.
297 return is_empty(in_str)
300 def is_empty(in_str: Any) -> bool:
303 in_str: the string to test
306 True if the string is empty and false otherwise.
308 See also :meth:`is_none_or_empty`, :meth:`is_full_string`.
312 >>> is_empty(' \t\t ')
318 >>> is_empty([1, 2, 3])
321 return is_string(in_str) and in_str.strip() == ""
324 def is_full_string(in_str: Any) -> bool:
327 in_str: the object to test
330 True if the object is a string and is not empty ('') and
331 is not only composed of whitespace.
333 See also :meth:`is_string`, :meth:`is_empty_string`, :meth:`is_none_or_empty`.
335 >>> is_full_string('test!')
337 >>> is_full_string('')
339 >>> is_full_string(' ')
341 >>> is_full_string(100.999)
343 >>> is_full_string({"a": 1, "b": 2})
346 return is_string(in_str) and in_str.strip() != ""
349 def is_number(in_str: str) -> bool:
352 in_str: the string to test
355 True if the string contains a valid numberic value and
358 See also :meth:`is_integer_number`, :meth:`is_decimal_number`,
359 :meth:`is_hexidecimal_integer_number`, :meth:`is_octal_integer_number`,
363 Traceback (most recent call last):
366 >>> is_number("100.5")
368 >>> is_number("test")
372 >>> is_number([1, 2, 3])
373 Traceback (most recent call last):
375 ValueError: [1, 2, 3]
377 if not is_string(in_str):
378 raise ValueError(in_str)
379 return NUMBER_RE.match(in_str) is not None
382 def is_integer_number(in_str: str) -> bool:
385 in_str: the string to test
388 True if the string contains a valid (signed or unsigned,
389 decimal, hex, or octal, regular or scientific) integral
390 expression and False otherwise.
392 See also :meth:`is_number`, :meth:`is_decimal_number`,
393 :meth:`is_hexidecimal_integer_number`, :meth:`is_octal_integer_number`,
396 >>> is_integer_number('42')
398 >>> is_integer_number('42.0')
402 (is_number(in_str) and "." not in in_str)
403 or is_hexidecimal_integer_number(in_str)
404 or is_octal_integer_number(in_str)
405 or is_binary_integer_number(in_str)
409 def is_hexidecimal_integer_number(in_str: str) -> bool:
412 in_str: the string to test
415 True if the string is a hex integer number and False otherwise.
417 See also :meth:`is_integer_number`, :meth:`is_decimal_number`,
418 :meth:`is_octal_integer_number`, :meth:`is_binary_integer_number`, etc...
420 >>> is_hexidecimal_integer_number('0x12345')
422 >>> is_hexidecimal_integer_number('0x1A3E')
424 >>> is_hexidecimal_integer_number('1234') # Needs 0x
426 >>> is_hexidecimal_integer_number('-0xff')
428 >>> is_hexidecimal_integer_number('test')
430 >>> is_hexidecimal_integer_number(12345) # Not a string
431 Traceback (most recent call last):
434 >>> is_hexidecimal_integer_number(101.4)
435 Traceback (most recent call last):
438 >>> is_hexidecimal_integer_number(0x1A3E)
439 Traceback (most recent call last):
443 if not is_string(in_str):
444 raise ValueError(in_str)
445 return HEX_NUMBER_RE.match(in_str) is not None
448 def is_octal_integer_number(in_str: str) -> bool:
451 in_str: the string to test
454 True if the string is a valid octal integral number and False otherwise.
456 See also :meth:`is_integer_number`, :meth:`is_decimal_number`,
457 :meth:`is_hexidecimal_integer_number`, :meth:`is_binary_integer_number`,
460 >>> is_octal_integer_number('0o777')
462 >>> is_octal_integer_number('-0O115')
464 >>> is_octal_integer_number('0xFF') # Not octal, needs 0o
466 >>> is_octal_integer_number('7777') # Needs 0o
468 >>> is_octal_integer_number('test')
471 if not is_string(in_str):
472 raise ValueError(in_str)
473 return OCT_NUMBER_RE.match(in_str) is not None
476 def is_binary_integer_number(in_str: str) -> bool:
479 in_str: the string to test
482 True if the string contains a binary integral number and False otherwise.
484 See also :meth:`is_integer_number`, :meth:`is_decimal_number`,
485 :meth:`is_hexidecimal_integer_number`, :meth:`is_octal_integer_number`,
488 >>> is_binary_integer_number('0b10111')
490 >>> is_binary_integer_number('-0b111')
492 >>> is_binary_integer_number('0B10101')
494 >>> is_binary_integer_number('0b10102')
496 >>> is_binary_integer_number('0xFFF')
498 >>> is_binary_integer_number('test')
501 if not is_string(in_str):
502 raise ValueError(in_str)
503 return BIN_NUMBER_RE.match(in_str) is not None
506 def to_int(in_str: str) -> int:
509 in_str: the string to convert
512 The integral value of the string or raises on error.
514 See also :meth:`is_integer_number`, :meth:`is_decimal_number`,
515 :meth:`is_hexidecimal_integer_number`, :meth:`is_octal_integer_number`,
516 :meth:`is_binary_integer_number`, etc...
522 >>> to_int('0b01101')
527 Traceback (most recent call last):
529 ValueError: invalid literal for int() with base 10: 'test'
531 if not is_string(in_str):
532 raise ValueError(in_str)
533 if is_binary_integer_number(in_str):
534 return int(in_str, 2)
535 if is_octal_integer_number(in_str):
536 return int(in_str, 8)
537 if is_hexidecimal_integer_number(in_str):
538 return int(in_str, 16)
542 def number_string_to_integer(in_str: str) -> int:
543 """Convert a string containing a written-out number into an int.
546 in_str: the string containing the long-hand written out integer number
547 in English. See examples below.
550 The integer whose value was parsed from in_str.
552 See also :meth:`integer_to_number_string`.
555 This code only handles integers; it will not work with decimals / floats.
557 >>> number_string_to_integer("one hundred fifty two")
560 >>> number_string_to_integer("ten billion two hundred million fifty four thousand three")
563 >>> number_string_to_integer("four-score and 7")
566 >>> number_string_to_integer("fifty xyzzy three")
567 Traceback (most recent call last):
569 ValueError: Unknown word: xyzzy
571 if isinstance(in_str, int):
575 in_str = in_str.replace('-', ' ')
576 for w in in_str.split():
577 if w not in NUM_WORDS:
578 if is_integer_number(w):
582 raise ValueError("Unknown word: " + w)
583 scale, increment = NUM_WORDS[w]
584 current = current * scale + increment
588 return result + current
591 def integer_to_number_string(num: int) -> str:
593 Opposite of :meth:`number_string_to_integer`; converts a number to a written out
594 longhand format in English.
597 num: the integer number to convert
600 The long-hand written out English form of the number. See examples below.
602 See also :meth:`number_string_to_integer`.
605 This method does not handle decimals or floats, only ints.
607 >>> integer_to_number_string(9)
610 >>> integer_to_number_string(42)
613 >>> integer_to_number_string(123219982)
614 'one hundred twenty three million two hundred nineteen thousand nine hundred eighty two'
618 return UNIT_WORDS[num]
620 ret = TENS_WORDS[num // 10]
623 ret += ' ' + UNIT_WORDS[leftover]
626 # If num > 100 go find the highest chunk and convert that, then recursively
627 # convert the rest. NUM_WORDS contains items like 'thousand' -> (1000, 0).
628 # The second item in the tuple is an increment that can be ignored; the first
629 # is the numeric "scale" of the entry. So find the greatest entry in NUM_WORDS
630 # still less than num. For 123,456 it would be thousand. Then pull out the
631 # 123, convert it, and append "thousand". Then do the rest.
633 for name, val in NUM_WORDS.items():
635 scales[name] = val[0]
636 scale = max(scales.items(), key=lambda _: _[1])
638 # scale[1] = numeric magnitude (e.g. 1000)
639 # scale[0] = name (e.g. "thousand")
640 ret = integer_to_number_string(num // scale[1]) + ' ' + scale[0]
641 leftover = num % scale[1]
643 ret += ' ' + integer_to_number_string(leftover)
647 def is_decimal_number(in_str: str) -> bool:
650 in_str: the string to check
653 True if the given string represents a decimal or False
654 otherwise. A decimal may be signed or unsigned or use
655 a "scientific notation".
657 See also :meth:`is_integer_number`.
660 We do not consider integers without a decimal point
661 to be decimals; they return False (see example).
663 >>> is_decimal_number('42.0')
665 >>> is_decimal_number('42')
668 return is_number(in_str) and "." in in_str
671 def strip_escape_sequences(in_str: str) -> str:
674 in_str: the string to strip of escape sequences.
677 in_str with escape sequences removed.
679 See also: :mod:`pyutils.ansi`.
682 What is considered to be an "escape sequence" is defined
683 by a regular expression. While this gets common ones,
684 there may exist valid sequences that it doesn't match.
686 >>> strip_escape_sequences('\x1B[12;11;22mthis is a test!')
689 in_str = ESCAPE_SEQUENCE_RE.sub("", in_str)
693 def add_thousands_separator(
694 in_str: str, *, separator_char: str = ',', places: int = 3
698 in_str: string or number to which to add thousands separator(s)
699 separator_char: the separator character to add (defaults to comma)
700 places: add a separator every N places (defaults to three)
703 A numeric string with thousands separators added appropriately.
705 >>> add_thousands_separator('12345678')
707 >>> add_thousands_separator(12345678)
709 >>> add_thousands_separator(12345678.99)
711 >>> add_thousands_separator('test')
712 Traceback (most recent call last):
717 if isinstance(in_str, numbers.Number):
719 if is_number(in_str):
720 return _add_thousands_separator(
721 in_str, separator_char=separator_char, places=places
723 raise ValueError(in_str)
726 def _add_thousands_separator(in_str: str, *, separator_char=',', places=3) -> str:
727 """Internal helper"""
730 (in_str, decimal_part) = in_str.split('.')
731 tmp = [iter(in_str[::-1])] * places
732 ret = separator_char.join("".join(x) for x in zip_longest(*tmp, fillvalue=""))[::-1]
733 if len(decimal_part) > 0:
739 def is_url(in_str: Any, allowed_schemes: Optional[List[str]] = None) -> bool:
742 in_str: the string to test
743 allowed_schemes: an optional list of allowed schemes (e.g.
744 ['http', 'https', 'ftp']. If passed, only URLs that
745 begin with the one of the schemes passed will be considered
746 to be valid. Otherwise, any scheme:// will be considered
750 True if in_str contains a valid URL and False otherwise.
752 >>> is_url('http://www.mysite.com')
754 >>> is_url('https://mysite.com')
756 >>> is_url('.mysite.com')
758 >>> is_url('scheme://username:
[email protected]:8042/folder/subfolder/file.extension?param=value¶m2=value2#hash')
761 if not is_full_string(in_str):
764 valid = URL_RE.match(in_str) is not None
767 return valid and any([in_str.startswith(s) for s in allowed_schemes])
771 def is_email(in_str: Any) -> bool:
774 in_str: the email address to check
776 Returns: True if the in_str contains a valid email (as defined by
777 https://tools.ietf.org/html/rfc3696#section-3) or False
782 >>> is_email('@gmail.com')
785 if not is_full_string(in_str) or len(in_str) > 320 or in_str.startswith("."):
789 # we expect 2 tokens, one before "@" and one after, otherwise
790 # we have an exception and the email is not valid.
791 head, tail = in_str.split("@")
793 # head's size must be <= 64, tail <= 255, head must not start
794 # with a dot or contain multiple consecutive dots.
795 if len(head) > 64 or len(tail) > 255 or head.endswith(".") or (".." in head):
798 # removes escaped spaces, so that later on the test regex will
800 head = head.replace("\\ ", "")
801 if head.startswith('"') and head.endswith('"'):
802 head = head.replace(" ", "")[1:-1]
803 return EMAIL_RE.match(head + "@" + tail) is not None
806 # borderline case in which we have multiple "@" signs but the
807 # head part is correctly escaped.
808 if ESCAPED_AT_SIGN.search(in_str) is not None:
809 # replace "@" with "a" in the head
810 return is_email(ESCAPED_AT_SIGN.sub("a", in_str))
814 def suffix_string_to_number(in_str: str) -> Optional[int]:
815 """Takes a string like "33Gb" and converts it into a number (of bytes)
819 in_str: the string with a suffix to be interpreted and removed.
822 An integer number of bytes or None to indicate an error.
824 See also :meth:`number_to_suffix_string`.
826 >>> suffix_string_to_number('1Mb')
828 >>> suffix_string_to_number('13.1Gb')
832 def suffix_capitalize(s: str) -> str:
836 return f"{s[0].upper()}{s[1].lower()}"
837 return suffix_capitalize(s[0:1])
839 if is_string(in_str):
840 if is_integer_number(in_str):
841 return to_int(in_str)
842 suffixes = [in_str[-2:], in_str[-1:]]
843 rest = [in_str[:-2], in_str[:-1]]
844 for x in range(len(suffixes)):
846 s = suffix_capitalize(s)
847 multiplier = NUM_SUFFIXES.get(s, None)
848 if multiplier is not None:
850 if is_integer_number(r):
851 return to_int(r) * multiplier
852 if is_decimal_number(r):
853 return int(float(r) * multiplier)
857 def number_to_suffix_string(num: int) -> Optional[str]:
858 """Take a number (of bytes) and returns a string like "43.8Gb".
861 num: an integer number of bytes
864 A string with a suffix representing num bytes concisely or
865 None to indicate an error.
867 See also: :meth:`suffix_string_to_number`.
869 >>> number_to_suffix_string(14066017894)
871 >>> number_to_suffix_string(1024 * 1024)
876 for (sfx, size) in NUM_SUFFIXES.items():
881 if suffix is not None:
882 return f"{d:.1f}{suffix}"
887 def is_credit_card(in_str: Any, card_type: str = None) -> bool:
890 in_str: a string to check
891 card_type: if provided, contains the card type to validate
892 with. Otherwise, all known credit card number types will
895 Supported card types are the following:
905 True if in_str is a valid credit card number.
908 This code is not verifying the authenticity of the credit card (i.e.
909 not checking whether it's a real card that can be charged); rather
910 it's only checking that the number follows the "rules" for numbering
911 established by credit card issuers.
914 if not is_full_string(in_str):
917 if card_type is not None:
918 if card_type not in CREDIT_CARDS:
920 f'Invalid card type "{card_type}". Valid types are: {CREDIT_CARDS.keys()}'
922 return CREDIT_CARDS[card_type].match(in_str) is not None
923 for c in CREDIT_CARDS:
924 if CREDIT_CARDS[c].match(in_str) is not None:
929 def is_camel_case(in_str: Any) -> bool:
932 in_str: the string to test
935 True if the string is formatted as camel case and False otherwise.
936 A string is considered camel case when:
938 * it's composed only by letters ([a-zA-Z]) and optionally numbers ([0-9])
939 * it contains both lowercase and uppercase letters
940 * it does not start with a number
942 See also :meth:`is_snake_case`, :meth:`is_slug`, and :meth:`camel_case_to_snake_case`.
944 return is_full_string(in_str) and CAMEL_CASE_TEST_RE.match(in_str) is not None
947 def is_snake_case(in_str: Any, *, separator: str = "_") -> bool:
950 in_str: the string to test
951 separator: the snake case separator character to use
953 Returns: True if the string is snake case and False otherwise. A
954 string is considered snake case when:
956 * it's composed only by lowercase/uppercase letters and digits
957 * it contains at least one underscore (or provided separator)
958 * it does not start with a number
960 See also :meth:`is_camel_case`, :meth:`is_slug`, and :meth:`snake_case_to_camel_case`.
962 >>> is_snake_case('this_is_a_test')
964 >>> is_snake_case('___This_Is_A_Test_1_2_3___')
966 >>> is_snake_case('this-is-a-test')
968 >>> is_snake_case('this-is-a-test', separator='-')
971 if is_full_string(in_str):
972 re_map = {"_": SNAKE_CASE_TEST_RE, "-": SNAKE_CASE_TEST_DASH_RE}
973 re_template = r"([a-z]+\d*{sign}[a-z\d{sign}]*|{sign}+[a-z\d]+[a-z\d{sign}]*)"
976 re.compile(re_template.format(sign=re.escape(separator)), re.IGNORECASE),
978 return r.match(in_str) is not None
982 def is_json(in_str: Any) -> bool:
985 in_str: the string to test
988 True if the in_str contains valid JSON and False otherwise.
990 >>> is_json('{"name": "Peter"}')
992 >>> is_json('[1, 2, 3]')
994 >>> is_json('{nope}')
997 if is_full_string(in_str) and JSON_WRAPPER_RE.match(in_str) is not None:
999 return isinstance(json.loads(in_str), (dict, list))
1000 except (TypeError, ValueError, OverflowError):
1005 def is_uuid(in_str: Any, allow_hex: bool = False) -> bool:
1008 in_str: the string to test
1009 allow_hex: should we allow hexidecimal digits in valid uuids?
1012 True if the in_str contains a valid UUID and False otherwise.
1014 See also :meth:`generate_uuid`.
1016 >>> is_uuid('6f8aa2f9-686c-4ac3-8766-5712354a04cf')
1018 >>> is_uuid('6f8aa2f9686c4ac387665712354a04cf')
1020 >>> is_uuid('6f8aa2f9686c4ac387665712354a04cf', allow_hex=True)
1023 # string casting is used to allow UUID itself as input data type
1026 return UUID_HEX_OK_RE.match(s) is not None
1027 return UUID_RE.match(s) is not None
1030 def is_ip_v4(in_str: Any) -> bool:
1033 in_str: the string to test
1036 True if in_str contains a valid IPv4 address and False otherwise.
1038 See also :meth:`extract_ip_v4`, :meth:`is_ip_v6`, :meth:`extract_ip_v6`,
1041 >>> is_ip_v4('255.200.100.75')
1043 >>> is_ip_v4('nope')
1045 >>> is_ip_v4('255.200.100.999') # 999 out of range
1048 if not is_full_string(in_str) or SHALLOW_IP_V4_RE.match(in_str) is None:
1051 # checks that each entry in the ip is in the valid range (0 to 255)
1052 for token in in_str.split("."):
1053 if not 0 <= int(token) <= 255:
1058 def extract_ip_v4(in_str: Any) -> Optional[str]:
1061 in_str: the string to extract an IPv4 address from.
1064 The first extracted IPv4 address from in_str or None if
1065 none were found or an error occurred.
1067 See also :meth:`is_ip_v4`, :meth:`is_ip_v6`, :meth:`extract_ip_v6`,
1070 >>> extract_ip_v4(' The secret IP address: 127.0.0.1 (use it wisely) ')
1072 >>> extract_ip_v4('Your mom dresses you funny.')
1074 if not is_full_string(in_str):
1076 m = ANYWHERE_IP_V4_RE.search(in_str)
1082 def is_ip_v6(in_str: Any) -> bool:
1085 in_str: the string to test.
1088 True if in_str contains a valid IPv6 address and False otherwise.
1090 See also :meth:`is_ip_v4`, :meth:`extract_ip_v4`, :meth:`extract_ip_v6`,
1093 >>> is_ip_v6('2001:db8:85a3:0000:0000:8a2e:370:7334')
1095 >>> is_ip_v6('2001:db8:85a3:0000:0000:8a2e:370:?') # invalid "?"
1098 return is_full_string(in_str) and IP_V6_RE.match(in_str) is not None
1101 def extract_ip_v6(in_str: Any) -> Optional[str]:
1104 in_str: the string from which to extract an IPv6 address.
1107 The first IPv6 address found in in_str or None if no address
1108 was found or an error occurred.
1110 See also :meth:`is_ip_v4`, :meth:`is_ip_v6`, :meth:`extract_ip_v4`,
1113 >>> extract_ip_v6('IP: 2001:db8:85a3:0000:0000:8a2e:370:7334')
1114 '2001:db8:85a3:0000:0000:8a2e:370:7334'
1115 >>> extract_ip_v6("(and she's ugly too, btw)")
1117 if not is_full_string(in_str):
1119 m = ANYWHERE_IP_V6_RE.search(in_str)
1125 def is_ip(in_str: Any) -> bool:
1128 in_str: the string to test.
1131 True if in_str contains a valid IP address (either IPv4 or
1134 See also :meth:`is_ip_v4`, :meth:`is_ip_v6`, :meth:`extract_ip_v6`,
1135 and :meth:`extract_ip_v4`.
1137 >>> is_ip('255.200.100.75')
1139 >>> is_ip('2001:db8:85a3:0000:0000:8a2e:370:7334')
1143 >>> is_ip('1.2.3.999')
1146 return is_ip_v6(in_str) or is_ip_v4(in_str)
1149 def extract_ip(in_str: Any) -> Optional[str]:
1152 in_str: the string from which to extract in IP address.
1155 The first IP address (IPv4 or IPv6) found in in_str or
1156 None to indicate none found or an error condition.
1158 See also :meth:`is_ip_v4`, :meth:`is_ip_v6`, :meth:`extract_ip_v6`,
1159 and :meth:`extract_ip_v4`.
1161 >>> extract_ip('Attacker: 255.200.100.75')
1163 >>> extract_ip('Remote host: 2001:db8:85a3:0000:0000:8a2e:370:7334')
1164 '2001:db8:85a3:0000:0000:8a2e:370:7334'
1165 >>> extract_ip('1.2.3')
1167 ip = extract_ip_v4(in_str)
1169 ip = extract_ip_v6(in_str)
1173 def is_mac_address(in_str: Any) -> bool:
1176 in_str: the string to test
1179 True if in_str is a valid MAC address False otherwise.
1181 See also :meth:`extract_mac_address`, :meth:`is_ip`, etc...
1183 >>> is_mac_address("34:29:8F:12:0D:2F")
1185 >>> is_mac_address('34:29:8f:12:0d:2f')
1187 >>> is_mac_address('34-29-8F-12-0D-2F')
1189 >>> is_mac_address("test")
1192 return is_full_string(in_str) and MAC_ADDRESS_RE.match(in_str) is not None
1195 def extract_mac_address(in_str: Any, *, separator: str = ":") -> Optional[str]:
1198 in_str: the string from which to extract a MAC address.
1199 separator: the MAC address hex byte separator to use.
1202 The first MAC address found in in_str or None to indicate no
1205 See also :meth:`is_mac_address`, :meth:`is_ip`, and :meth:`extract_ip`.
1207 >>> extract_mac_address(' MAC Address: 34:29:8F:12:0D:2F')
1210 >>> extract_mac_address('? (10.0.0.30) at d8:5d:e2:34:54:86 on em0 expires in 1176 seconds [ethernet]')
1213 if not is_full_string(in_str):
1216 m = ANYWHERE_MAC_ADDRESS_RE.search(in_str)
1219 mac.replace(":", separator)
1220 mac.replace("-", separator)
1225 def is_slug(in_str: Any, separator: str = "-") -> bool:
1228 in_str: string to test
1229 separator: the slug character to use
1232 True if in_str is a slug string and False otherwise.
1234 See also :meth:`is_camel_case`, :meth:`is_snake_case`, and :meth:`slugify`.
1236 >>> is_slug('my-blog-post-title')
1238 >>> is_slug('My blog post title')
1241 if not is_full_string(in_str):
1243 rex = r"^([a-z\d]+" + re.escape(separator) + r"*?)*[a-z\d]$"
1244 return re.match(rex, in_str) is not None
1247 def contains_html(in_str: str) -> bool:
1250 in_str: the string to check for tags in
1253 True if the given string contains HTML/XML tags and False
1256 See also :meth:`strip_html`.
1259 By design, this function matches ANY type of tag, so don't expect
1260 to use it as an HTML validator. It's a quick sanity check at
1261 best. See something like BeautifulSoup for a more full-featuered
1264 >>> contains_html('my string is <strong>bold</strong>')
1266 >>> contains_html('my string is not bold')
1270 if not is_string(in_str):
1271 raise ValueError(in_str)
1272 return HTML_RE.search(in_str) is not None
1275 def words_count(in_str: str) -> int:
1278 in_str: the string to count words in
1281 The number of words contained in the given string.
1284 This method is "smart" in that it does consider only sequences
1285 of one or more letter and/or numbers to be "words". Thus a
1286 string like this: "! @ # % ... []" will return zero. Moreover
1287 it is aware of punctuation, so the count for a string like
1288 "one,two,three.stop" will be 4 not 1 (even if there are no spaces
1291 >>> words_count('hello world')
1293 >>> words_count('one,two,three.stop')
1296 if not is_string(in_str):
1297 raise ValueError(in_str)
1298 return len(WORDS_COUNT_RE.findall(in_str))
1301 def word_count(in_str: str) -> int:
1304 in_str: the string to count words in
1307 The number of words contained in the given string.
1310 This method is "smart" in that it does consider only sequences
1311 of one or more letter and/or numbers to be "words". Thus a
1312 string like this: "! @ # % ... []" will return zero. Moreover
1313 it is aware of punctuation, so the count for a string like
1314 "one,two,three.stop" will be 4 not 1 (even if there are no spaces
1317 >>> word_count('hello world')
1319 >>> word_count('one,two,three.stop')
1322 return words_count(in_str)
1325 def generate_uuid(omit_dashes: bool = False) -> str:
1328 omit_dashes: should we omit the dashes in the generated UUID?
1331 A generated UUID string (using `uuid.uuid4()`) with or without
1332 dashes per the omit_dashes arg.
1334 See also :meth:`is_uuid`, :meth:`generate_random_alphanumeric_string`.
1336 generate_uuid() # possible output: '97e3a716-6b33-4ab9-9bb1-8128cb24d76b'
1337 generate_uuid(omit_dashes=True) # possible output: '97e3a7166b334ab99bb18128cb24d76b'
1345 def generate_random_alphanumeric_string(size: int) -> str:
1348 size: number of characters to generate
1351 A string of the specified size containing random characters
1352 (uppercase/lowercase ascii letters and digits).
1354 See also :meth:`asciify`, :meth:`generate_uuid`.
1357 >>> generate_random_alphanumeric_string(9)
1361 raise ValueError("size must be >= 1")
1362 chars = string.ascii_letters + string.digits
1363 buffer = [random.choice(chars) for _ in range(size)]
1364 return from_char_list(buffer)
1367 def reverse(in_str: str) -> str:
1370 in_str: the string to reverse
1373 The reversed (chracter by character) string.
1378 if not is_string(in_str):
1379 raise ValueError(in_str)
1383 def camel_case_to_snake_case(in_str: str, *, separator: str = "_"):
1386 in_str: the camel case string to convert
1387 separator: the snake case separator character to use
1390 A snake case string equivalent to the camel case input or the
1391 original string if it is not a valid camel case string or some
1394 See also :meth:`is_camel_case`, :meth:`is_snake_case`, and :meth:`is_slug`.
1396 >>> camel_case_to_snake_case('MacAddressExtractorFactory')
1397 'mac_address_extractor_factory'
1398 >>> camel_case_to_snake_case('Luke Skywalker')
1401 if not is_string(in_str):
1402 raise ValueError(in_str)
1403 if not is_camel_case(in_str):
1405 return CAMEL_CASE_REPLACE_RE.sub(lambda m: m.group(1) + separator, in_str).lower()
1408 def snake_case_to_camel_case(
1409 in_str: str, *, upper_case_first: bool = True, separator: str = "_"
1413 in_str: the snake case string to convert
1414 upper_case_first: should we capitalize the first letter?
1415 separator: the separator character to use
1418 A camel case string that is equivalent to the snake case string
1419 provided or the original string back again if it is not valid
1420 snake case or another error occurs.
1422 See also :meth:`is_camel_case`, :meth:`is_snake_case`, and :meth:`is_slug`.
1424 >>> snake_case_to_camel_case('this_is_a_test')
1426 >>> snake_case_to_camel_case('Han Solo')
1429 if not is_string(in_str):
1430 raise ValueError(in_str)
1431 if not is_snake_case(in_str, separator=separator):
1433 tokens = [s.title() for s in in_str.split(separator) if is_full_string(s)]
1434 if not upper_case_first:
1435 tokens[0] = tokens[0].lower()
1436 return from_char_list(tokens)
1439 def to_char_list(in_str: str) -> List[str]:
1442 in_str: the string to split into a char list
1445 A list of strings of length one each.
1447 See also :meth:`from_char_list`.
1449 >>> to_char_list('test')
1450 ['t', 'e', 's', 't']
1452 if not is_string(in_str):
1457 def from_char_list(in_list: List[str]) -> str:
1460 in_list: A list of characters to convert into a string.
1463 The string resulting from gluing the characters in in_list
1466 See also :meth:`to_char_list`.
1468 >>> from_char_list(['t', 'e', 's', 't'])
1471 return "".join(in_list)
1474 def shuffle(in_str: str) -> Optional[str]:
1477 in_str: a string to shuffle randomly by character
1480 A new string containing same chars of the given one but in
1481 a randomized order. Note that in rare cases this could result
1482 in the same original string as no check is done. Returns
1483 None to indicate error conditions.
1486 >>> shuffle('awesome')
1489 if not is_string(in_str):
1491 chars = to_char_list(in_str)
1492 random.shuffle(chars)
1493 return from_char_list(chars)
1496 def scramble(in_str: str) -> Optional[str]:
1499 in_str: a string to shuffle randomly by character
1502 A new string containing same chars of the given one but in
1503 a randomized order. Note that in rare cases this could result
1504 in the same original string as no check is done. Returns
1505 None to indicate error conditions.
1507 See also :mod:`pyutils.unscrambler`.
1510 >>> scramble('awesome')
1513 return shuffle(in_str)
1516 def strip_html(in_str: str, keep_tag_content: bool = False) -> str:
1519 in_str: the string to strip tags from
1520 keep_tag_content: should we keep the inner contents of tags?
1523 A string with all HTML tags removed (optionally with tag contents
1526 See also :meth:`contains_html`.
1529 This method uses simple regular expressions to strip tags and is
1530 not a full fledged HTML parser by any means. Consider using
1531 something like BeautifulSoup if your needs are more than this
1532 simple code can fulfill.
1534 >>> strip_html('test: <a href="foo/bar">click here</a>')
1536 >>> strip_html('test: <a href="foo/bar">click here</a>', keep_tag_content=True)
1539 if not is_string(in_str):
1540 raise ValueError(in_str)
1541 r = HTML_TAG_ONLY_RE if keep_tag_content else HTML_RE
1542 return r.sub("", in_str)
1545 def asciify(in_str: str) -> str:
1548 in_str: the string to asciify.
1551 An output string roughly equivalent to the original string
1552 where all content to are ascii-only. This is accomplished
1553 by translating all non-ascii chars into their closest possible
1554 ASCII representation (eg: ó -> o, Ë -> E, ç -> c...).
1556 See also :meth:`to_ascii`, :meth:`generate_random_alphanumeric_string`.
1559 Some chars may be lost if impossible to translate.
1561 >>> asciify('èéùúòóäåëýñÅÀÁÇÌÍÑÓË')
1562 'eeuuooaaeynAAACIINOE'
1564 if not is_string(in_str):
1565 raise ValueError(in_str)
1567 # "NFKD" is the algorithm which is able to successfully translate
1568 # the most of non-ascii chars.
1569 normalized = unicodedata.normalize("NFKD", in_str)
1571 # encode string forcing ascii and ignore any errors
1572 # (unrepresentable chars will be stripped out)
1573 ascii_bytes = normalized.encode("ascii", "ignore")
1575 # turns encoded bytes into an utf-8 string
1576 return ascii_bytes.decode("utf-8")
1579 def slugify(in_str: str, *, separator: str = "-") -> str:
1582 in_str: the string to slugify
1583 separator: the character to use during sligification (default
1587 The converted string. The returned string has the following properties:
1590 * all letters are in lower case
1591 * all punctuation signs and non alphanumeric chars are removed
1592 * words are divided using provided separator
1593 * all chars are encoded as ascii (by using :meth:`asciify`)
1596 See also :meth:`is_slug` and :meth:`asciify`.
1598 >>> slugify('Top 10 Reasons To Love Dogs!!!')
1599 'top-10-reasons-to-love-dogs'
1600 >>> slugify('Mönstér Mägnët')
1603 if not is_string(in_str):
1604 raise ValueError(in_str)
1606 # replace any character that is NOT letter or number with spaces
1607 out = NO_LETTERS_OR_NUMBERS_RE.sub(" ", in_str.lower()).strip()
1609 # replace spaces with join sign
1610 out = SPACES_RE.sub(separator, out)
1612 # normalize joins (remove duplicates)
1613 out = re.sub(re.escape(separator) + r"+", separator, out)
1617 def to_bool(in_str: str) -> bool:
1620 in_str: the string to convert to boolean
1623 A boolean equivalent of the original string based on its contents.
1624 All conversion is case insensitive. A positive boolean (True) is
1625 returned if the string value is any of the following:
1634 Otherwise False is returned.
1636 See also :mod:`pyutils.argparse_utils`.
1656 if not is_string(in_str):
1657 raise ValueError(in_str)
1658 return in_str.lower() in set(["true", "1", "yes", "y", "t", "on"])
1661 def to_date(in_str: str) -> Optional[datetime.date]:
1664 in_str: the string to convert into a date
1667 The datetime.date the string contained or None to indicate
1668 an error. This parser is relatively clever; see
1669 :class:`datetimes.dateparse_utils` docs for details.
1671 See also: :mod:`pyutils.datetimes.dateparse_utils`, :meth:`extract_date`,
1672 :meth:`is_valid_date`, :meth:`to_datetime`, :meth:`valid_datetime`.
1674 >>> to_date('9/11/2001')
1675 datetime.date(2001, 9, 11)
1676 >>> to_date('xyzzy')
1678 import pyutils.datetimes.dateparse_utils as du
1681 d = du.DateParser() # type: ignore
1684 except du.ParseException: # type: ignore
1689 def extract_date(in_str: Any) -> Optional[datetime.datetime]:
1690 """Finds and extracts a date from the string, if possible.
1693 in_str: the string to extract a date from
1696 a datetime if date was found, otherwise None
1698 See also: :mod:`pyutils.datetimes.dateparse_utils`, :meth:`to_date`,
1699 :meth:`is_valid_date`, :meth:`to_datetime`, :meth:`valid_datetime`.
1701 >>> extract_date("filename.txt dec 13, 2022")
1702 datetime.datetime(2022, 12, 13, 0, 0)
1704 >>> extract_date("Dear Santa, please get me a pony.")
1709 import pyutils.datetimes.dateparse_utils as du
1711 d = du.DateParser() # type: ignore
1712 chunks = in_str.split()
1713 for ngram in itertools.chain(
1714 list_utils.ngrams(chunks, 5),
1715 list_utils.ngrams(chunks, 4),
1716 list_utils.ngrams(chunks, 3),
1717 list_utils.ngrams(chunks, 2),
1720 expr = " ".join(ngram)
1721 logger.debug("Trying %s", expr)
1723 return d.get_datetime()
1724 except du.ParseException: # type: ignore
1729 def is_valid_date(in_str: str) -> bool:
1732 in_str: the string to check
1735 True if the string represents a valid date that we can recognize
1736 and False otherwise. This parser is relatively clever; see
1737 :class:`datetimes.dateparse_utils` docs for details.
1739 See also: :mod:`pyutils.datetimes.dateparse_utils`, :meth:`to_date`,
1740 :meth:`extract_date`, :meth:`to_datetime`, :meth:`valid_datetime`.
1742 >>> is_valid_date('1/2/2022')
1744 >>> is_valid_date('christmas')
1746 >>> is_valid_date('next wednesday')
1748 >>> is_valid_date('xyzzy')
1751 import pyutils.datetimes.dateparse_utils as dp
1754 d = dp.DateParser() # type: ignore
1757 except dp.ParseException: # type: ignore
1762 def to_datetime(in_str: str) -> Optional[datetime.datetime]:
1765 in_str: string to parse into a datetime
1768 A python datetime parsed from in_str or None to indicate
1769 an error. This parser is relatively clever; see
1770 :class:`datetimes.dateparse_utils` docs for details.
1772 See also: :mod:`pyutils.datetimes.dateparse_utils`, :meth:`to_date`,
1773 :meth:`extract_date`, :meth:`valid_datetime`.
1775 >>> to_datetime('7/20/1969 02:56 GMT')
1776 datetime.datetime(1969, 7, 20, 2, 56, tzinfo=<StaticTzInfo 'GMT'>)
1778 import pyutils.datetimes.dateparse_utils as dp
1781 d = dp.DateParser() # type: ignore
1782 dt = d.parse(in_str)
1783 if isinstance(dt, datetime.datetime):
1790 def valid_datetime(in_str: str) -> bool:
1793 in_str: the string to check
1796 True if in_str contains a valid datetime and False otherwise.
1797 This parser is relatively clever; see
1798 :class:`datetimes.dateparse_utils` docs for details.
1800 >>> valid_datetime('next wednesday at noon')
1802 >>> valid_datetime('3 weeks ago at midnight')
1804 >>> valid_datetime('next easter at 5:00 am')
1806 >>> valid_datetime('sometime soon')
1809 _ = to_datetime(in_str)
1815 def squeeze(in_str: str, character_to_squeeze: str = ' ') -> str:
1818 in_str: the string to squeeze
1819 character_to_squeeze: the character to remove runs of
1820 more than one in a row (default = space)
1822 Returns: A "squeezed string" where runs of more than one
1823 character_to_squeeze into one.
1825 >>> squeeze(' this is a test ')
1828 >>> squeeze('one|!||!|two|!||!|three', character_to_squeeze='|!|')
1833 r'(' + re.escape(character_to_squeeze) + r')+',
1834 character_to_squeeze,
1839 def dedent(in_str: str) -> Optional[str]:
1842 in_str: the string to dedent
1845 A string with tab indentation removed or None on error.
1847 See also :meth:`indent`.
1849 >>> dedent('\t\ttest\\n\t\ting')
1852 if not is_string(in_str):
1854 line_separator = '\n'
1855 lines = [MARGIN_RE.sub('', line) for line in in_str.split(line_separator)]
1856 return line_separator.join(lines)
1859 def indent(in_str: str, amount: int) -> str:
1862 in_str: the string to indent
1863 amount: count of spaces to indent each line by
1866 An indented string created by prepending amount spaces.
1868 See also :meth:`dedent`.
1870 >>> indent('This is a test', 4)
1873 if not is_string(in_str):
1874 raise ValueError(in_str)
1875 line_separator = '\n'
1876 lines = [" " * amount + line for line in in_str.split(line_separator)]
1877 return line_separator.join(lines)
1880 def _sprintf(*args, **kwargs) -> str:
1881 """Internal helper."""
1884 sep = kwargs.pop("sep", None)
1886 if not isinstance(sep, str):
1887 raise TypeError("sep must be None or a string")
1889 end = kwargs.pop("end", None)
1891 if not isinstance(end, str):
1892 raise TypeError("end must be None or a string")
1895 raise TypeError("invalid keyword arguments to sprint()")
1901 for n, arg in enumerate(args):
1904 if isinstance(arg, str):
1912 def strip_ansi_sequences(in_str: str) -> str:
1915 in_str: the string to strip
1918 in_str with recognized ANSI escape sequences removed.
1920 See also :mod:`pyutils.ansi`.
1923 This method works by using a regular expression.
1924 It works for all ANSI escape sequences I've tested with but
1925 may miss some; caveat emptor.
1927 >>> import ansi as a
1928 >>> s = a.fg('blue') + 'blue!' + a.reset()
1929 >>> len(s) # '\x1b[38;5;21mblue!\x1b[m'
1931 >>> len(strip_ansi_sequences(s))
1933 >>> strip_ansi_sequences(s)
1937 return re.sub(r'\x1b\[[\d+;]*[a-z]', '', in_str)
1940 class SprintfStdout(contextlib.AbstractContextManager):
1942 A context manager that captures outputs to stdout to a buffer
1943 without printing them.
1945 >>> with SprintfStdout() as buf:
1947 ... print("1, 2, 3")
1949 >>> print(buf(), end='')
1954 def __init__(self) -> None:
1955 self.destination = io.StringIO()
1956 self.recorder: contextlib.redirect_stdout
1958 def __enter__(self) -> Callable[[], str]:
1959 self.recorder = contextlib.redirect_stdout(self.destination)
1960 self.recorder.__enter__()
1961 return lambda: self.destination.getvalue()
1963 def __exit__(self, *args) -> Literal[False]:
1964 self.recorder.__exit__(*args)
1965 self.destination.seek(0)
1969 def capitalize_first_letter(in_str: str) -> str:
1972 in_str: the string to capitalize
1975 in_str with the first character capitalized.
1977 >>> capitalize_first_letter('test')
1979 >>> capitalize_first_letter("ALREADY!")
1982 return in_str[0].upper() + in_str[1:]
1985 def it_they(n: int) -> str:
1988 n: how many of them are there?
1991 'it' if n is one or 'they' otherwize.
1993 See also :meth:`is_are`, :meth:`pluralize`, :meth:`make_contractions`,
1998 n = num_files_saved_to_tmp()
1999 print(f'Saved file{pluralize(n)} successfully.')
2000 print(f'{it_they(n)} {is_are(n)} located in /tmp.')
2012 def is_are(n: int) -> str:
2015 n: how many of them are there?
2018 'is' if n is one or 'are' otherwize.
2020 See also :meth:`it_they`, :meth:`pluralize`, :meth:`make_contractions`,
2025 n = num_files_saved_to_tmp()
2026 print(f'Saved file{pluralize(n)} successfully.')
2027 print(f'{it_they(n)} {is_are(n)} located in /tmp.')
2040 def pluralize(n: int) -> str:
2043 n: how many of them are there?
2046 's' if n is greater than one otherwize ''.
2048 See also :meth:`it_they`, :meth:`is_are`, :meth:`make_contractions`,
2053 n = num_files_saved_to_tmp()
2054 print(f'Saved file{pluralize(n)} successfully.')
2055 print(f'{it_they(n)} {is_are(n)} located in /tmp.')
2060 >>> print(f'There {is_are(count)} {count} file{pluralize(count)}.')
2063 >>> print(f'There {is_are(count)} {count} file{pluralize(count)}.')
2071 def make_contractions(txt: str) -> str:
2072 """This code glues words in txt together to form (English)
2076 txt: the input text to be contractionized.
2079 Output text identical to original input except for any
2080 recognized contractions are formed.
2082 See also :meth:`it_they`, :meth:`is_are`, :meth:`make_contractions`.
2085 The order in which we create contractions is defined by the
2086 implementation and what I thought made more sense when writing
2089 >>> make_contractions('It is nice today.')
2092 >>> make_contractions('I can not even...')
2095 >>> make_contractions('She could not see!')
2098 >>> make_contractions('But she will not go.')
2101 >>> make_contractions('Verily, I shall not.')
2104 >>> make_contractions('No you cannot.')
2107 >>> make_contractions('I said you can not go.')
2108 "I said you can't go."
2144 ['woul(d)', 'i(s)', 'a(re)', 'ha(s)', 'ha(ve)', 'ha(d)', 'wi(ll)'],
2148 # Special cases: can't, shan't and won't.
2149 txt = re.sub(r'\b(can)\s*no(t)\b', r"\1'\2", txt, count=0, flags=re.IGNORECASE)
2151 r'\b(sha)ll\s*(n)o(t)\b', r"\1\2'\3", txt, count=0, flags=re.IGNORECASE
2154 r'\b(w)ill\s*(n)(o)(t)\b',
2158 flags=re.IGNORECASE,
2161 for first_list, second_list in first_second:
2162 for first in first_list:
2163 for second in second_list:
2164 # Disallow there're/where're. They're valid English
2166 if (first in set(['there', 'where'])) and second == 'a(re)':
2169 pattern = fr'\b({first})\s+{second}\b'
2170 if second == '(n)o(t)':
2171 replacement = r"\1\2'\3"
2173 replacement = r"\1'\2"
2174 txt = re.sub(pattern, replacement, txt, count=0, flags=re.IGNORECASE)
2179 def thify(n: int) -> str:
2182 n: how many of them are there?
2185 The proper cardinal suffix for a number.
2187 See also :meth:`it_they`, :meth:`is_are`, :meth:`make_contractions`.
2196 print(f'The {attempt_count}{thify(attempt_count)} failed, trying again.')
2206 assert is_integer_number(digit)
2218 def ngrams(txt: str, n: int):
2221 txt: the string to create ngrams using
2222 n: how many words per ngram created?
2225 Generates the ngrams from the input string.
2227 See also :meth:`ngrams_presplit`, :meth:`bigrams`, :meth:`trigrams`.
2229 >>> [x for x in ngrams('This is a test', 2)]
2230 ['This is', 'is a', 'a test']
2233 for ngram in ngrams_presplit(words, n):
2240 def ngrams_presplit(words: Sequence[str], n: int):
2242 Same as :meth:`ngrams` but with the string pre-split.
2244 See also :meth:`ngrams`, :meth:`bigrams`, :meth:`trigrams`.
2246 return list_utils.ngrams(words, n)
2249 def bigrams(txt: str):
2250 """Generates the bigrams (n=2) of the given string.
2252 See also :meth:`ngrams`, :meth:`trigrams`.
2254 >>> [x for x in bigrams('this is a test')]
2255 ['this is', 'is a', 'a test']
2257 return ngrams(txt, 2)
2260 def trigrams(txt: str):
2261 """Generates the trigrams (n=3) of the given string.
2263 See also :meth:`ngrams`, :meth:`bigrams`.
2265 return ngrams(txt, 3)
2268 def shuffle_columns_into_list(
2269 input_lines: Sequence[str], column_specs: Iterable[Iterable[int]], delim: str = ''
2271 """Helper to shuffle / parse columnar data and return the results as a
2275 input_lines: A sequence of strings that represents text that
2276 has been broken into columns by the caller
2277 column_specs: an iterable collection of numeric sequences that
2278 indicate one or more column numbers to copy to form the Nth
2279 position in the output list. See example below.
2280 delim: for column_specs that indicate we should copy more than
2281 one column from the input into this position, use delim to
2282 separate source data. Defaults to ''.
2285 A list of string created by following the instructions set forth
2288 See also :meth:`shuffle_columns_into_dict`.
2290 >>> cols = '-rwxr-xr-x 1 scott wheel 3.1K Jul 9 11:34 acl_test.py'.split()
2291 >>> shuffle_columns_into_list(
2293 ... [ [8], [2, 3], [5, 6, 7] ],
2296 ['acl_test.py', 'scott!wheel', 'Jul!9!11:34']
2300 # Column specs map input lines' columns into outputs.
2302 for spec in column_specs:
2305 hunk = hunk + delim + input_lines[n]
2306 hunk = hunk.strip(delim)
2311 def shuffle_columns_into_dict(
2312 input_lines: Sequence[str],
2313 column_specs: Iterable[Tuple[str, Iterable[int]]],
2315 ) -> Dict[str, str]:
2316 """Helper to shuffle / parse columnar data and return the results
2320 input_lines: a sequence of strings that represents text that
2321 has been broken into columns by the caller
2322 column_specs: instructions for what dictionary keys to apply
2323 to individual or compound input column data. See example
2325 delim: when forming compound output data by gluing more than
2326 one input column together, use this character to separate
2327 the source data. Defaults to ''.
2330 A dict formed by applying the column_specs instructions.
2332 See also :meth:`shuffle_columns_into_list`, :meth:`interpolate_using_dict`.
2334 >>> cols = '-rwxr-xr-x 1 scott wheel 3.1K Jul 9 11:34 acl_test.py'.split()
2335 >>> shuffle_columns_into_dict(
2337 ... [ ('filename', [8]), ('owner', [2, 3]), ('mtime', [5, 6, 7]) ],
2340 {'filename': 'acl_test.py', 'owner': 'scott!wheel', 'mtime': 'Jul!9!11:34'}
2344 # Column specs map input lines' columns into outputs.
2345 # "key", [col1, col2...]
2346 for spec in column_specs:
2349 hunk = hunk + delim + input_lines[n]
2350 hunk = hunk.strip(delim)
2355 def interpolate_using_dict(txt: str, values: Dict[str, str]) -> str:
2357 Interpolate a string with data from a dict.
2360 txt: the mad libs template
2361 values: what you and your kids chose for each category.
2363 See also :meth:`shuffle_columns_into_list`, :meth:`shuffle_columns_into_dict`.
2365 >>> interpolate_using_dict('This is a {adjective} {noun}.',
2366 ... {'adjective': 'good', 'noun': 'example'})
2367 'This is a good example.'
2369 return _sprintf(txt.format(**values), end='')
2372 def to_ascii(txt: str):
2375 txt: the input data to encode
2378 txt encoded as an ASCII byte string.
2380 See also :meth:`to_base64`, :meth:`to_bitstring`, :meth:`to_bytes`,
2381 :meth:`generate_random_alphanumeric_string`, :meth:`asciify`.
2383 >>> to_ascii('test')
2386 >>> to_ascii(b'1, 2, 3')
2389 if isinstance(txt, str):
2390 return txt.encode('ascii')
2391 if isinstance(txt, bytes):
2393 raise Exception('to_ascii works with strings and bytes')
2397 txt: str, *, encoding: str = 'utf-8', errors: str = 'surrogatepass'
2401 txt: the input data to encode
2402 encoding: the encoding to use during conversion
2403 errors: how to handle encoding errors
2406 txt encoded with a 64-chracter alphabet. Similar to and compatible
2407 with uuencode/uudecode.
2409 See also :meth:`is_base64`, :meth:`to_ascii`, :meth:`to_bitstring`,
2410 :meth:`from_base64`.
2412 >>> to_base64('hello?')
2415 return base64.encodebytes(txt.encode(encoding, errors))
2418 def is_base64(txt: str) -> bool:
2421 txt: the string to check
2424 True if txt is a valid base64 encoded string. This assumes
2425 txt was encoded with Python's standard base64 alphabet which
2426 is the same as what uuencode/uudecode uses).
2428 See also :meth:`to_base64`, :meth:`from_base64`.
2430 >>> is_base64('test') # all letters in the b64 alphabet
2433 >>> is_base64('another test, how do you like this one?')
2436 >>> is_base64(b'aGVsbG8/\\n') # Ending newline is ok.
2440 a = string.ascii_uppercase + string.ascii_lowercase + string.digits + '+/'
2441 alphabet = set(a.encode('ascii'))
2442 for char in to_ascii(txt.strip()):
2443 if char not in alphabet:
2449 b64: bytes, encoding: str = 'utf-8', errors: str = 'surrogatepass'
2453 b64: bytestring of 64-bit encoded data to decode / convert.
2454 encoding: the encoding to use during conversion
2455 errors: how to handle encoding errors
2458 The decoded form of b64 as a normal python string. Similar to
2459 and compatible with uuencode / uudecode.
2461 See also :meth:`to_base64`, :meth:`is_base64`.
2463 >>> from_base64(b'aGVsbG8/\\n')
2466 return base64.decodebytes(b64).decode(encoding, errors)
2469 def chunk(txt: str, chunk_size: int):
2472 txt: a string to be chunked into evenly spaced pieces.
2473 chunk_size: the size of each chunk to make
2476 The original string chunked into evenly spaced pieces.
2478 >>> ' '.join(chunk('010011011100010110101010101010101001111110101000', 8))
2479 '01001101 11000101 10101010 10101010 10011111 10101000'
2481 if len(txt) % chunk_size != 0:
2482 msg = f'String to chunk\'s length ({len(txt)} is not an even multiple of chunk_size ({chunk_size})'
2484 warnings.warn(msg, stacklevel=2)
2485 for x in range(0, len(txt), chunk_size):
2486 yield txt[x : x + chunk_size]
2489 def to_bitstring(txt: str, *, delimiter: str = '') -> str:
2492 txt: the string to convert into a bitstring
2493 delimiter: character to insert between adjacent bytes. Note that
2494 only bitstrings with delimiter='' are interpretable by
2495 :meth:`from_bitstring`.
2498 txt converted to ascii/binary and then chopped into bytes.
2500 See also :meth:`to_base64`, :meth:`from_bitstring`, :meth:`is_bitstring`,
2503 >>> to_bitstring('hello?')
2504 '011010000110010101101100011011000110111100111111'
2506 >>> to_bitstring('test', delimiter=' ')
2507 '01110100 01100101 01110011 01110100'
2509 >>> to_bitstring(b'test')
2510 '01110100011001010111001101110100'
2512 etxt = to_ascii(txt)
2513 bits = bin(int.from_bytes(etxt, 'big'))
2515 return delimiter.join(chunk(bits.zfill(8 * ((len(bits) + 7) // 8)), 8))
2518 def is_bitstring(txt: str) -> bool:
2521 txt: the string to check
2524 True if txt is a recognized bitstring and False otherwise.
2525 Note that if delimiter is non empty this code will not
2526 recognize the bitstring.
2528 See also :meth:`to_base64`, :meth:`from_bitstring`, :meth:`to_bitstring`,
2531 >>> is_bitstring('011010000110010101101100011011000110111100111111')
2534 >>> is_bitstring('1234')
2537 return is_binary_integer_number(f'0b{txt}')
2541 bits: str, encoding: str = 'utf-8', errors: str = 'surrogatepass'
2545 bits: the bitstring to convert back into a python string
2546 encoding: the encoding to use during conversion
2547 errors: how to handle encoding errors
2550 The regular python string represented by bits. Note that this
2551 code does not work with to_bitstring when delimiter is non-empty.
2553 See also :meth:`to_base64`, :meth:`to_bitstring`, :meth:`is_bitstring`,
2556 >>> from_bitstring('011010000110010101101100011011000110111100111111')
2560 return n.to_bytes((n.bit_length() + 7) // 8, 'big').decode(encoding, errors) or '\0'
2563 def ip_v4_sort_key(txt: str) -> Optional[Tuple[int, ...]]:
2566 txt: an IP address to chunk up for sorting purposes
2569 A tuple of IP components arranged such that the sorting of
2570 IP addresses using a normal comparator will do something sane
2573 See also :meth:`is_ip_v4`.
2575 >>> ip_v4_sort_key('10.0.0.18')
2578 >>> ips = ['10.0.0.10', '100.0.0.1', '1.2.3.4', '10.0.0.9']
2579 >>> sorted(ips, key=lambda x: ip_v4_sort_key(x))
2580 ['1.2.3.4', '10.0.0.9', '10.0.0.10', '100.0.0.1']
2582 if not is_ip_v4(txt):
2583 print(f"not IP: {txt}")
2585 return tuple(int(x) for x in txt.split('.'))
2588 def path_ancestors_before_descendants_sort_key(volume: str) -> Tuple[str, ...]:
2591 volume: the string to chunk up for sorting purposes
2594 A tuple of volume's components such that the sorting of
2595 volumes using a normal comparator will do something sane
2598 See also :mod:`pyutils.files.file_utils`.
2600 >>> path_ancestors_before_descendants_sort_key('/usr/local/bin')
2601 ('usr', 'local', 'bin')
2603 >>> paths = ['/usr/local', '/usr/local/bin', '/usr']
2604 >>> sorted(paths, key=lambda x: path_ancestors_before_descendants_sort_key(x))
2605 ['/usr', '/usr/local', '/usr/local/bin']
2607 return tuple(x for x in volume.split('/') if len(x) > 0)
2610 def replace_all(in_str: str, replace_set: str, replacement: str) -> str:
2612 Execute several replace operations in a row.
2615 in_str: the string in which to replace characters
2616 replace_set: the set of target characters to replace
2617 replacement: the character to replace any member of replace_set
2620 See also :meth:`replace_nth`.
2623 The string with replacements executed.
2625 >>> s = 'this_is a-test!'
2626 >>> replace_all(s, ' _-!', '')
2629 for char in replace_set:
2630 in_str = in_str.replace(char, replacement)
2634 def replace_nth(in_str: str, source: str, target: str, nth: int):
2636 Replaces the nth occurrance of a substring within a string.
2639 in_str: the string in which to run the replacement
2640 source: the substring to replace
2641 target: the replacement text
2642 nth: which occurrance of source to replace?
2644 See also :meth:`replace_all`.
2646 >>> replace_nth('this is a test', ' ', '-', 3)
2649 where = [m.start() for m in re.finditer(source, in_str)][nth - 1]
2650 before = in_str[:where]
2651 after = in_str[where:]
2652 after = after.replace(source, target, 1)
2653 return before + after
2656 if __name__ == '__main__':