django/django/utils/text.py

import builtins
import gzip
import re
import secrets
import unicodedata
from collections import deque
from gzip import GzipFile
from gzip import compress as gzip_compress
from html import escape
from html.parser import HTMLParser
from io import BytesIO
from itertools import islice

from django.core.exceptions import SuspiciousFileOperation
from django.utils.functional import (
    SimpleLazyObject,
    cached_property,
    keep_lazy_text,
    lazy,
)
from django.utils.regex_helper import _lazy_re_compile
from django.utils.translation import gettext as _
from django.utils.translation import gettext_lazy, pgettext


@keep_lazy_text
def capfirst(x):
    """Capitalize the first letter of a string."""
    if not x:
        return x
    if not isinstance(x, str):
        x = str(x)
    return x[0].upper() + x[1:]


# Set up regular expressions
re_newlines = _lazy_re_compile(r"\r\n|\r")  # Used in normalize_newlines
re_camel_case = _lazy_re_compile(r"(((?<=[a-z])[A-Z])|([A-Z](?![A-Z]|$)))")


@keep_lazy_text
def wrap(text, width):
    """
    A word-wrap function that preserves existing line breaks. Expects that
    existing line breaks are posix newlines.

    Preserve all white space except added line breaks consume the space on
    which they break the line.

    Don't wrap long words, thus the output text may have lines longer than
    ``width``.
    """

    def _generator():
        for line in text.splitlines(True):  # True keeps trailing linebreaks
            max_width = min((line.endswith("\n") and width + 1 or width), width)
            while len(line) > max_width:
                space = line[: max_width + 1].rfind(" ") + 1
                if space == 0:
                    space = line.find(" ") + 1
                    if space == 0:
                        yield line
                        line = ""
                        break
                yield "%s\n" % line[: space - 1]
                line = line[space:]
                max_width = min((line.endswith("\n") and width + 1 or width), width)
            if line:
                yield line

    return "".join(_generator())


def add_truncation_text(text, truncate=None):
    if truncate is None:
        truncate = pgettext(
            "String to return when truncating text", "%(truncated_text)s…"
        )
    if "%(truncated_text)s" in truncate:
        return truncate % {"truncated_text": text}
    # The truncation text didn't contain the %(truncated_text)s string
    # replacement argument so just append it to the text.
    if text.endswith(truncate):
        # But don't append the truncation text if the current text already ends
        # in this.
        return text
    return f"{text}{truncate}"


def calculate_truncate_chars_length(length, replacement):
    truncate_len = length
    for char in add_truncation_text("", replacement):
        if not unicodedata.combining(char):
            truncate_len -= 1
            if truncate_len == 0:
                break
    return truncate_len


class TruncateHTMLParser(HTMLParser):
    class TruncationCompleted(Exception):
        pass

    def __init__(self, *, length, replacement, convert_charrefs=True):
        super().__init__(convert_charrefs=convert_charrefs)
        self.tags = deque()
        self.output = ""
        self.remaining = length
        self.replacement = replacement

    @cached_property
    def void_elements(self):
        from django.utils.html import VOID_ELEMENTS

        return VOID_ELEMENTS

    def handle_startendtag(self, tag, attrs):
        self.handle_starttag(tag, attrs)
        if tag not in self.void_elements:
            self.handle_endtag(tag)

    def handle_starttag(self, tag, attrs):
        self.output += self.get_starttag_text()
        if tag not in self.void_elements:
            self.tags.appendleft(tag)

    def handle_endtag(self, tag):
        if tag not in self.void_elements:
            self.output += f"</{tag}>"
            try:
                self.tags.remove(tag)
            except ValueError:
                pass

    def handle_data(self, data):
        data, output = self.process(data)
        data_len = len(data)
        if self.remaining < data_len:
            self.remaining = 0
            self.output += add_truncation_text(output, self.replacement)
            raise self.TruncationCompleted
        self.remaining -= data_len
        self.output += output

    def feed(self, data):
        try:
            super().feed(data)
        except self.TruncationCompleted:
            self.output += "".join([f"</{tag}>" for tag in self.tags])
            self.tags.clear()
            self.reset()
        else:
            # No data was handled.
            self.reset()


class TruncateCharsHTMLParser(TruncateHTMLParser):
    def __init__(self, *, length, replacement, convert_charrefs=True):
        self.length = length
        self.processed_chars = 0
        super().__init__(
            length=calculate_truncate_chars_length(length, replacement),
            replacement=replacement,
            convert_charrefs=convert_charrefs,
        )

    def process(self, data):
        self.processed_chars += len(data)
        if (self.processed_chars == self.length) and (
            len(self.output) + len(data) == len(self.rawdata)
        ):
            self.output += data
            raise self.TruncationCompleted
        output = escape("".join(data[: self.remaining]))
        return data, output


class TruncateWordsHTMLParser(TruncateHTMLParser):
    def process(self, data):
        data = re.split(r"(?<=\S)\s+(?=\S)", data)
        output = escape(" ".join(data[: self.remaining]))
        return data, output


class Truncator(SimpleLazyObject):
    """
    An object used to truncate text, either by characters or words.

    When truncating HTML text (either chars or words), input will be limited to
    at most `MAX_LENGTH_HTML` characters.
    """

    # 5 million characters are approximately 4000 text pages or 3 web pages.
    MAX_LENGTH_HTML = 5_000_000

    def __init__(self, text):
        super().__init__(lambda: str(text))

    def chars(self, num, truncate=None, html=False):
        """
        Return the text truncated to be no longer than the specified number
        of characters.

        `truncate` specifies what should be used to notify that the string has
        been truncated, defaulting to a translatable string of an ellipsis.
        """
        self._setup()
        length = int(num)
        if length <= 0:
            return ""
        text = unicodedata.normalize("NFC", self._wrapped)

        if html:
            parser = TruncateCharsHTMLParser(length=length, replacement=truncate)
            parser.feed(text)
            parser.close()
            return parser.output
        return self._text_chars(length, truncate, text)

    def _text_chars(self, length, truncate, text):
        """Truncate a string after a certain number of chars."""
        truncate_len = calculate_truncate_chars_length(length, truncate)
        s_len = 0
        end_index = None
        for i, char in enumerate(text):
            if unicodedata.combining(char):
                # Don't consider combining characters
                # as adding to the string length
                continue
            s_len += 1
            if end_index is None and s_len > truncate_len:
                end_index = i
            if s_len > length:
                # Return the truncated string
                return add_truncation_text(text[: end_index or 0], truncate)

        # Return the original string since no truncation was necessary
        return text

    def words(self, num, truncate=None, html=False):
        """
        Truncate a string after a certain number of words. `truncate` specifies
        what should be used to notify that the string has been truncated,
        defaulting to ellipsis.
        """
        self._setup()
        length = int(num)
        if length <= 0:
            return ""
        if html:
            parser = TruncateWordsHTMLParser(length=length, replacement=truncate)
            parser.feed(self._wrapped)
            parser.close()
            return parser.output
        return self._text_words(length, truncate)

    def _text_words(self, length, truncate):
        """
        Truncate a string after a certain number of words.

        Strip newlines in the string.
        """
        words = self._wrapped.split()
        if len(words) > length:
            words = words[:length]
            return add_truncation_text(" ".join(words), truncate)
        return " ".join(words)


@keep_lazy_text
def get_valid_filename(name):
    """
    Return the given string converted to a string that can be used for a clean
    filename. Remove leading and trailing spaces; convert other spaces to
    underscores; and remove anything that is not an alphanumeric, dash,
    underscore, or dot.
    >>> get_valid_filename("john's portrait in 2004.jpg")
    'johns_portrait_in_2004.jpg'
    """
    s = str(name).strip().replace(" ", "_")
    s = re.sub(r"(?u)[^-\w.]", "", s)
    if s in {"", ".", ".."}:
        raise SuspiciousFileOperation("Could not derive file name from '%s'" % name)
    return s


@keep_lazy_text
def get_text_list(list_, last_word=gettext_lazy("or")):
    """
    >>> get_text_list(['a', 'b', 'c', 'd'])
    'a, b, c or d'
    >>> get_text_list(['a', 'b', 'c'], 'and')
    'a, b and c'
    >>> get_text_list(['a', 'b'], 'and')
    'a and b'
    >>> get_text_list(['a'])
    'a'
    >>> get_text_list([])
    ''
    """
    if not list_:
        return ""
    if len(list_) == 1:
        return str(list_[0])
    return "%s %s %s" % (
        # Translators: This string is used as a separator between list elements
        _(", ").join(str(i) for i in list_[:-1]),
        str(last_word),
        str(list_[-1]),
    )


@keep_lazy_text
def normalize_newlines(text):
    """Normalize CRLF and CR newlines to just LF."""
    return re_newlines.sub("\n", str(text))


@keep_lazy_text
def phone2numeric(phone):
    """Convert a phone number with letters into its numeric equivalent."""
    char2number = {
        "a": "2",
        "b": "2",
        "c": "2",
        "d": "3",
        "e": "3",
        "f": "3",
        "g": "4",
        "h": "4",
        "i": "4",
        "j": "5",
        "k": "5",
        "l": "5",
        "m": "6",
        "n": "6",
        "o": "6",
        "p": "7",
        "q": "7",
        "r": "7",
        "s": "7",
        "t": "8",
        "u": "8",
        "v": "8",
        "w": "9",
        "x": "9",
        "y": "9",
        "z": "9",
    }
    return "".join(char2number.get(c, c) for c in phone.lower())


def _get_random_filename(max_random_bytes):
    return b"a" * secrets.randbelow(max_random_bytes)


def compress_string(s, *, max_random_bytes=None):
    compressed_data = gzip_compress(s, compresslevel=6, mtime=0)

    if not max_random_bytes:
        return compressed_data

    compressed_view = memoryview(compressed_data)
    header = bytearray(compressed_view[:10])
    header[3] = gzip.FNAME

    filename = _get_random_filename(max_random_bytes) + b"\x00"

    return bytes(header) + filename + compressed_view[10:]


class StreamingBuffer(BytesIO):
    def read(self):
        ret = self.getvalue()
        self.seek(0)
        self.truncate()
        return ret


# Like compress_string, but for iterators of strings.
def compress_sequence(sequence, *, max_random_bytes=None):
    buf = StreamingBuffer()
    filename = _get_random_filename(max_random_bytes) if max_random_bytes else None
    with GzipFile(
        filename=filename, mode="wb", compresslevel=6, fileobj=buf, mtime=0
    ) as zfile:
        # Output headers...
        yield buf.read()
        for item in sequence:
            zfile.write(item)
            data = buf.read()
            if data:
                yield data
    yield buf.read()


# Expression to match some_token and some_token="with spaces" (and similarly
# for single-quoted strings).
smart_split_re = _lazy_re_compile(
    r"""
    ((?:
        [^\s'"]*
        (?:
            (?:"(?:[^"\\]|\\.)*" | '(?:[^'\\]|\\.)*')
            [^\s'"]*
        )+
    ) | \S+)
""",
    re.VERBOSE,
)


def smart_split(text):
    r"""
    Generator that splits a string by spaces, leaving quoted phrases together.
    Supports both single and double quotes, and supports escaping quotes with
    backslashes. In the output, strings will keep their initial and trailing
    quote marks and escaped quotes will remain escaped (the results can then
    be further processed with unescape_string_literal()).

    >>> list(smart_split(r'This is "a person\'s" test.'))
    ['This', 'is', '"a person\\\'s"', 'test.']
    >>> list(smart_split(r"Another 'person\'s' test."))
    ['Another', "'person\\'s'", 'test.']
    >>> list(smart_split(r'A "\"funky\" style" test.'))
    ['A', '"\\"funky\\" style"', 'test.']
    """
    for bit in smart_split_re.finditer(str(text)):
        yield bit[0]


@keep_lazy_text
def unescape_string_literal(s):
    r"""
    Convert quoted string literals to unquoted strings with escaped quotes and
    backslashes unquoted::

        >>> unescape_string_literal('"abc"')
        'abc'
        >>> unescape_string_literal("'abc'")
        'abc'
        >>> unescape_string_literal('"a \"bc\""')
        'a "bc"'
        >>> unescape_string_literal("'\'ab\' c'")
        "'ab' c"
    """
    if not s or s[0] not in "\"'" or s[-1] != s[0]:
        raise ValueError("Not a string literal: %r" % s)
    quote = s[0]
    return s[1:-1].replace(r"\%s" % quote, quote).replace(r"\\", "\\")


@keep_lazy_text
def slugify(value, allow_unicode=False):
    """
    Convert to ASCII if 'allow_unicode' is False. Convert spaces or repeated
    dashes to single dashes. Remove characters that aren't alphanumerics,
    underscores, or hyphens. Convert to lowercase. Also strip leading and
    trailing whitespace, dashes, and underscores.
    """
    value = str(value)
    if allow_unicode:
        value = unicodedata.normalize("NFKC", value)
    else:
        value = (
            unicodedata.normalize("NFKD", value)
            .encode("ascii", "ignore")
            .decode("ascii")
        )
    value = re.sub(r"[^\w\s-]", "", value.lower())
    return re.sub(r"[-\s]+", "-", value).strip("-_")


def camel_case_to_spaces(value):
    """
    Split CamelCase and convert to lowercase. Strip surrounding whitespace.
    """
    return re_camel_case.sub(r" \1", value).strip().lower()


def _format_lazy(format_string, *args, **kwargs):
    """
    Apply str.format() on 'format_string' where format_string, args,
    and/or kwargs might be lazy.
    """
    return format_string.format(*args, **kwargs)


format_lazy = lazy(_format_lazy, str)


class DebugRepr:
    """
    Modified Reprlib.Repr from Python 3.12 that includes fillvalue customization.
    """

    def __init__(self, limit):
        self.indent = 0
        self.maxlevel = 2
        self.limit = limit

    def repr(self, x):
        return self.repr1(x, self.maxlevel)

    def repr1(self, x, level):
        typename = type(x).__name__
        if " " in typename:
            parts = typename.split()
            typename = "_".join(parts)
        if hasattr(self, "repr_" + typename):
            return getattr(self, "repr_" + typename)(x, level)
        else:
            return self.repr_instance(x, level)

    def _join(self, pieces, level):
        if self.indent is None:
            return ", ".join(pieces)
        if not pieces:
            return ""
        indent = self.indent
        if isinstance(indent, int):
            if indent < 0:
                raise ValueError(f"Repr.indent cannot be negative int (was {indent!r})")
            indent *= " "
        try:
            sep = ", " + (self.maxlevel - level + 1) * indent
        except TypeError as error:
            raise TypeError(
                f"Repr.indent must be a str, int or None, not {type(indent)}"
            ) from error
        return sep.join(("", *pieces))[1 : -len(indent) or None]

    def _repr_iterable(self, x, level, left, right, maxiter, trail=""):
        n = len(x)
        fillvalue = self.gen_trim_msg(n)
        if level <= 0 and n:
            s = fillvalue
        else:
            newlevel = level - 1
            repr1 = self.repr1
            pieces = [repr1(elem, newlevel) for elem in islice(x, maxiter)]
            if n > maxiter:
                pieces.append(fillvalue)
            s = self._join(pieces, level)
            if n == 1 and trail and self.indent is None:
                right = trail + right
        return "%s%s%s" % (left, s, right)

    def repr_tuple(self, x, level):
        return self._repr_iterable(x, level, "(", ")", self.limit, ",")

    def repr_list(self, x, level):
        return self._repr_iterable(x, level, "[", "]", self.limit)

    def repr_array(self, x, level):
        if not x:
            return "array('%s')" % x.typecode
        header = "array('%s', [" % x.typecode
        return self._repr_iterable(x, level, header, "])", self.limit)

    def repr_set(self, x, level):
        if not x:
            return "set()"
        x = _possibly_sorted(x)
        return self._repr_iterable(x, level, "{", "}", self.limit)

    def repr_frozenset(self, x, level):
        if not x:
            return "frozenset()"
        x = _possibly_sorted(x)
        return self._repr_iterable(x, level, "frozenset({", "})", self.limit)

    def repr_deque(self, x, level):
        return self._repr_iterable(x, level, "deque([", "])", self.limit)

    def repr_dict(self, x, level):
        n = len(x)
        if n == 0:
            return "{}"
        fillvalue = self.gen_trim_msg(n)
        if level <= 0:
            return "{" + fillvalue + "}"
        newlevel = level - 1
        repr1 = self.repr1
        pieces = []
        for key in islice(_possibly_sorted(x), self.limit):
            keyrepr = repr1(key, newlevel)
            valrepr = repr1(x[key], newlevel)
            pieces.append("%s: %s" % (keyrepr, valrepr))
        if n > self.limit:
            pieces.append(fillvalue)
        s = self._join(pieces, level)
        return "{%s}" % (s,)

    def repr_int(self, x, level):
        s = builtins.repr(x)
        if len(s) > self.limit:
            i = max(0, (self.limit - 3) // 2)
            j = max(0, self.limit - 3 - i)
            s = s[:i] + self.limit + s[len(s) - j :]
        return s

    def repr_instance(self, x, level):
        try:
            s = builtins.repr(x)
            # Bugs in x.__repr__() can cause arbitrary
            # exceptions -- then make up something
        except Exception:
            return "<%s instance at %#x>" % (x.__class__.__name__, id(x))
        if len(s) > self.limit:
            i = max(0, (self.limit - 3) // 2)
            j = max(0, self.limit - 3 - i)
            fillvalue = self.gen_trim_msg(len(s))
            s = s[:i] + fillvalue + s[len(s) - j :]
        return s

    def repr_str(self, x, level):
        return "'%s'" % (x[: self.limit] + self.gen_trim_msg(len(x)))

    def print(self, value):
        return self.repr(value)

    def gen_trim_msg(self, length):
        if length <= self.limit:
            return ""
        return "...<trimmed %d bytes string> " % (length - self.limit)


def _possibly_sorted(x):
    # Since not all sequences of items can be sorted and comparison
    # functions may raise arbitrary exceptions, return an unsorted
    # sequence in that case.
    try:
        return sorted(x)
    except Exception:
        return list(x)