import os.path import platform import re import sys import textwrap from abc import ABC, abstractmethod from pathlib import Path from typing import (
Any,
Dict,
Iterable,
List,
NamedTuple,
Optional,
Sequence,
Set,
Tuple,
Type,
Union,
)
from pip._vendor.pygments.lexer import Lexer from pip._vendor.pygments.lexers import get_lexer_by_name, guess_lexer_for_filename from pip._vendor.pygments.style import Style as PygmentsStyle from pip._vendor.pygments.styles import get_style_by_name from pip._vendor.pygments.token import (
Comment,
Error,
Generic,
Keyword,
Name,
Number,
Operator,
String,
Token,
Whitespace,
) from pip._vendor.pygments.util import ClassNotFound
from pip._vendor.rich.containers import Lines from pip._vendor.rich.padding import Padding, PaddingDimensions
from ._loop import loop_first from .cells import cell_len from .color import Color, blend_rgb from .console import Console, ConsoleOptions, JustifyMethod, RenderResult from .jupyter import JupyterMixin from .measure import Measurement from .segment import Segment, Segments from .style import Style, StyleType from .text import Text
TokenType = Tuple[str, ...]
WINDOWS = platform.system() == "Windows"
DEFAULT_THEME = "monokai"
class SyntaxTheme(ABC): """Base class for a syntax theme."""
@abstractmethod def get_style_for_token(self, token_type: TokenType) -> Style: """Get a style for a given Pygments token.""" raise NotImplementedError # pragma: no cover
@abstractmethod def get_background_style(self) -> Style: """Get the background color.""" raise NotImplementedError # pragma: no cover
class PygmentsSyntaxTheme(SyntaxTheme): """Syntax theme that delegates to Pygments theme."""
def get_style_for_token(self, token_type: TokenType) -> Style: """Look up style in the style map.""" try: return self._style_cache[token_type] except KeyError: # Styles form a hierarchy # We need to go from most to least specific # e.g. ("foo", "bar", "baz") to ("foo", "bar") to ("foo",)
get_style = self.style_map.get
token = tuple(token_type)
style = self._missing_style while token:
_style = get_style(token) if _style isnotNone:
style = _style break
token = token[:-1]
self._style_cache[token_type] = style return style
class _SyntaxHighlightRange(NamedTuple): """
A range to highlight in a Syntax object.
`start` and `end` are 2-integers tuples, where the first integer is the line number
(starting from 1) and the second integer is the column index (starting from 0). """
class Syntax(JupyterMixin): """Construct a Syntax object to render syntax highlighted code.
Args:
code (str): Code to highlight.
lexer (Lexer | str): Lexer to use (see https://pygments.org/docs/lexers/)
theme (str, optional): Color theme, aka Pygments style (see https://pygments.org/docs/styles/#getting-a-list-of-available-styles). Defaults to "monokai".
dedent (bool, optional): Enable stripping of initial whitespace. Defaults to False.
line_numbers (bool, optional): Enable rendering of line numbers. Defaults to False.
start_line (int, optional): Starting number for line numbers. Defaults to 1.
line_range (Tuple[int | None, int | None], optional): If given should be a tuple of the start andend line to render.
A value of Nonein the tuple indicates the range is open in that direction.
highlight_lines (Set[int]): A set of line numbers to highlight.
code_width: Width of code to render (not including line numbers), or ``None`` to use all available width.
tab_size (int, optional): Size of tabs. Defaults to 4.
word_wrap (bool, optional): Enable word wrapping.
background_color (str, optional): Optional background color, orNone to use theme color. Defaults to None.
indent_guides (bool, optional): Show indent guides. Defaults to False.
padding (PaddingDimensions): Padding to apply around the syntax. Defaults to 0 (no padding). """
Args:
path (str): Path to file to highlight.
encoding (str): Encoding of file.
lexer (str | Lexer, optional): Lexer to use. IfNone, lexer will be auto-detected from path/file content.
theme (str, optional): Color theme, aka Pygments style (see https://pygments.org/docs/styles/#getting-a-list-of-available-styles). Defaults to "emacs".
dedent (bool, optional): Enable stripping of initial whitespace. Defaults to True.
line_numbers (bool, optional): Enable rendering of line numbers. Defaults to False.
start_line (int, optional): Starting number for line numbers. Defaults to 1.
line_range (Tuple[int, int], optional): If given should be a tuple of the start and end line to render.
highlight_lines (Set[int]): A set of line numbers to highlight.
code_width: Width of code to render (not including line numbers), or ``None`` to use all available width.
tab_size (int, optional): Size of tabs. Defaults to 4.
word_wrap (bool, optional): Enable word wrapping of code.
background_color (str, optional): Optional background color, orNone to use theme color. Defaults to None.
indent_guides (bool, optional): Show indent guides. Defaults to False.
padding (PaddingDimensions): Padding to apply around the syntax. Defaults to 0 (no padding).
Returns:
[Syntax]: A Syntax object that may be printed to the console """
code = Path(path).read_text(encoding=encoding)
@classmethod def guess_lexer(cls, path: str, code: Optional[str] = None) -> str: """Guess the alias of the Pygments lexer to use based on a path and an optional string of code. If code is supplied, it will use a combination of the code and the filename to determine the
best lexer to use. For example, if the file is ``index.html`` and the file contains Django
templating syntax, then "html+django" will be returned. If the file is ``index.html``, and no
templating language is used, the "html" lexer will be used. If no string of code is supplied, the lexer will be chosen based on the file extension..
Args:
path (AnyStr): The path to the file containing the code you wish to know the lexer for.
code (str, optional): Optional string of code that will be used as a fallback if no lexer is found for the supplied path.
Returns:
str: The name of the Pygments lexer that best matches the supplied path/code. """
lexer: Optional[Lexer] = None
lexer_name = "default" if code: try:
lexer = guess_lexer_for_filename(path, code) except ClassNotFound: pass
if lexer isNone:
text.append(code) else: if line_range: # More complicated path to only stylize a portion of the code # This speeds up further operations as there are less spans to process
line_start, line_end = line_range
def line_tokenize() -> Iterable[Tuple[Any, str]]: """Split tokens to one per line.""" assert lexer # required to make MyPy happy - we know lexer is not None at this point
for token_type, token in lexer.get_tokens(code): while token:
line_token, new_line, token = token.partition("\n") yield token_type, line_token + new_line
# Skip over tokens until line start while line_no < _line_start: try:
_token_type, token = next(tokens) except StopIteration: break yield (token, None) if token.endswith("\n"):
line_no += 1 # Generate spans until line end for token_type, token in tokens: yield (token, _get_theme_style(token_type)) if token.endswith("\n"):
line_no += 1 if line_end and line_no >= line_end: break
text.append_tokens(tokens_to_spans())
else:
text.append_tokens(
(token, _get_theme_style(token_type)) for token_type, token in lexer.get_tokens(code)
) if self.background_color isnotNone:
text.stylize(f"on {self.background_color}")
if self._stylized_ranges:
self._apply_stylized_ranges(text)
return text
def stylize_range(
self, style: StyleType, start: SyntaxPosition, end: SyntaxPosition
) -> None: """
Adds a custom style on a part of the code, that will be applied to the syntax display when it's rendered.
Line numbers are 1-based, while column indexes are 0-based.
Args:
style (StyleType): The style to apply.
start (Tuple[int, int]): The start of the range, in the form `[line number, column index]`.
end (Tuple[int, int]): The end of the range, in the form `[line number, column index]`. """
self._stylized_ranges.append(_SyntaxHighlightRange(style, start, end))
@property def _numbers_column_width(self) -> int: """Get the number of characters used to render the numbers column."""
column_width = 0 if self.line_numbers:
column_width = (
len(str(self.start_line + self.code.count("\n")))
+ NUMBERS_COLUMN_DEFAULT_PADDING
) return column_width
for line_no, line in enumerate(lines, self.start_line + line_offset): if self.word_wrap:
wrapped_lines = console.render_lines(
line,
render_options.update(height=None, justify="left"),
style=background_style,
pad=not transparent_background,
) else:
segments = list(line.render(console, end="")) if options.no_wrap:
wrapped_lines = [segments] else:
wrapped_lines = [
_Segment.adjust_line_length(
segments,
render_options.max_width,
style=background_style,
pad=not transparent_background,
)
]
if self.line_numbers:
wrapped_line_left_pad = _Segment( " " * numbers_column_width + " ", background_style
) for first, wrapped_line in loop_first(wrapped_lines): if first:
line_column = str(line_no).rjust(numbers_column_width - 2) + " " if highlight_line(line_no): yield _Segment(line_pointer, Style(color="red")) yield _Segment(line_column, highlight_number_style) else: yield _Segment(" ", highlight_number_style) yield _Segment(line_column, number_style) else: yield wrapped_line_left_pad yieldfrom wrapped_line yield new_line else: for wrapped_line in wrapped_lines: yieldfrom wrapped_line yield new_line
def _apply_stylized_ranges(self, text: Text) -> None: """
Apply stylized ranges to a text instance,
using the given code to determine the right portion to apply the style to.
Args:
text (Text): Text instance to apply the style to. """
code = text.plain
newlines_offsets = [ # Let's add outer boundaries at each side of the list:
0, # N.B. using "\n" here is much faster than using metacharacters such as "^" or "\Z":
*[
match.start() + 1 for match in re.finditer("\n", code, flags=re.MULTILINE)
],
len(code) + 1,
]
for stylized_range in self._stylized_ranges:
start = _get_code_index_for_syntax_position(
newlines_offsets, stylized_range.start
)
end = _get_code_index_for_syntax_position(
newlines_offsets, stylized_range.end
) if start isnotNoneand end isnotNone:
text.stylize(stylized_range.style, start, end)
def _process_code(self, code: str) -> Tuple[bool, str]: """
Applies various processing to a raw code string
(normalises it so it always ends with a line return, dedents it if necessary, etc.)
Args:
code (str): The raw code string to process
Returns:
Tuple[bool, str]: the boolean indicates whether the raw code ends with a line return, while the string is the processed code. """
ends_on_nl = code.endswith("\n")
processed_code = code if ends_on_nl else code + "\n"
processed_code = (
textwrap.dedent(processed_code) if self.dedent else processed_code
)
processed_code = processed_code.expandtabs(self.tab_size) return ends_on_nl, processed_code
def _get_code_index_for_syntax_position(
newlines_offsets: Sequence[int], position: SyntaxPosition
) -> Optional[int]: """
Returns the index of the code string for the given positions.
Args:
newlines_offsets (Sequence[int]): The offset of each newline character found in the code snippet.
position (SyntaxPosition): The position to search for.
Returns:
Optional[int]: The index of the code string for this position, or `None` if the given position's line number is out of range (if it's the column that is out of range
we silently clamp its value so that it reaches the end of the line) """
lines_count = len(newlines_offsets)
line_number, column_index = position if line_number > lines_count or len(newlines_offsets) < (line_number + 1): returnNone# `line_number` is out of range
line_index = line_number - 1
line_length = newlines_offsets[line_index + 1] - newlines_offsets[line_index] - 1 # If `column_index` is out of range: let's silently clamp it:
column_index = min(line_length, column_index) return newlines_offsets[line_index] + column_index
if __name__ == "__main__": # pragma: no cover import argparse import sys
parser = argparse.ArgumentParser(
description="Render syntax to the console with Rich"
)
parser.add_argument( "path",
metavar="PATH",
help="path to file, or - for stdin",
)
parser.add_argument( "-c", "--force-color",
dest="force_color",
action="store_true",
default=None,
help="force color for non-terminals",
)
parser.add_argument( "-i", "--indent-guides",
dest="indent_guides",
action="store_true",
default=False,
help="display indent guides",
)
parser.add_argument( "-l", "--line-numbers",
dest="line_numbers",
action="store_true",
help="render line numbers",
)
parser.add_argument( "-w", "--width",
type=int,
dest="width",
default=None,
help="width of output (default will auto-detect)",
)
parser.add_argument( "-r", "--wrap",
dest="word_wrap",
action="store_true",
default=False,
help="word wrap long lines",
)
parser.add_argument( "-s", "--soft-wrap",
action="store_true",
dest="soft_wrap",
default=False,
help="enable soft wrapping mode",
)
parser.add_argument( "-t", "--theme", dest="theme", default="monokai", help="pygments theme"
)
parser.add_argument( "-b", "--background-color",
dest="background_color",
default=None,
help="Override background color",
)
parser.add_argument( "-x", "--lexer",
default=None,
dest="lexer_name",
help="Lexer name",
)
parser.add_argument( "-p", "--padding", type=int, default=0, dest="padding", help="Padding"
)
parser.add_argument( "--highlight-line",
type=int,
default=None,
dest="highlight_line",
help="The line number (not index!) to highlight",
)
args = parser.parse_args()
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.