docutils0.14-fork/docutils/parsers/rst/directives/__init__.py

# $Id: __init__.py 8024 2017-02-06 00:41:48Z goodger $
# Author: David Goodger <goodger@python.org>
# Copyright: This module has been placed in the public domain.

"""
This package contains directive implementation modules.
"""

__docformat__ = 'reStructuredText'

import re
import codecs
import sys

from docutils import nodes
from docutils.utils import split_escaped_whitespace, escape2null, unescape
from docutils.parsers.rst.languages import en as _fallback_language_module
if sys.version_info < (2,5):
    from docutils._compat import __import__


_directive_registry = {
      'attention': ('admonitions', 'Attention'),
      'caution': ('admonitions', 'Caution'),
      'code': ('body', 'CodeBlock'),
      'danger': ('admonitions', 'Danger'),
      'error': ('admonitions', 'Error'),
      'important': ('admonitions', 'Important'),
      'note': ('admonitions', 'Note'),
      'tip': ('admonitions', 'Tip'),
      'hint': ('admonitions', 'Hint'),
      'warning': ('admonitions', 'Warning'),
      'admonition': ('admonitions', 'Admonition'),
      'sidebar': ('body', 'Sidebar'),
      'topic': ('body', 'Topic'),
      'line-block': ('body', 'LineBlock'),
      'parsed-literal': ('body', 'ParsedLiteral'),
      'math': ('body', 'MathBlock'),
      'rubric': ('body', 'Rubric'),
      'epigraph': ('body', 'Epigraph'),
      'highlights': ('body', 'Highlights'),
      'pull-quote': ('body', 'PullQuote'),
      'compound': ('body', 'Compound'),
      'container': ('body', 'Container'),
      #'questions': ('body', 'question_list'),
      'table': ('tables', 'RSTTable'),
      'csv-table': ('tables', 'CSVTable'),
      'list-table': ('tables', 'ListTable'),
      'image': ('images', 'Image'),
      'figure': ('images', 'Figure'),
      'contents': ('parts', 'Contents'),
      'sectnum': ('parts', 'Sectnum'),
      'header': ('parts', 'Header'),
      'footer': ('parts', 'Footer'),
      #'footnotes': ('parts', 'footnotes'),
      #'citations': ('parts', 'citations'),
      'target-notes': ('references', 'TargetNotes'),
      'meta': ('html', 'Meta'),
      #'imagemap': ('html', 'imagemap'),
      'raw': ('misc', 'Raw'),
      'include': ('misc', 'Include'),
      'replace': ('misc', 'Replace'),
      'unicode': ('misc', 'Unicode'),
      'class': ('misc', 'Class'),
      'role': ('misc', 'Role'),
      'default-role': ('misc', 'DefaultRole'),
      'title': ('misc', 'Title'),
      'date': ('misc', 'Date'),
      'restructuredtext-test-directive': ('misc', 'TestDirective'),}
"""Mapping of directive name to (module name, class name).  The
directive name is canonical & must be lowercase.  Language-dependent
names are defined in the ``language`` subpackage."""

_directives = {}
"""Cache of imported directives."""

def directive(directive_name, language_module, document):
    """
    Locate and return a directive function from its language-dependent name.
    If not found in the current language, check English.  Return None if the
    named directive cannot be found.
    """
    normname = directive_name.lower()
    messages = []
    msg_text = []
    if normname in _directives:
        return _directives[normname], messages
    canonicalname = None
    try:
        canonicalname = language_module.directives[normname]
    except AttributeError, error:
        msg_text.append('Problem retrieving directive entry from language '
                        'module %r: %s.' % (language_module, error))
    except KeyError:
        msg_text.append('No directive entry for "%s" in module "%s".'
                        % (directive_name, language_module.__name__))
    if not canonicalname:
        try:
            canonicalname = _fallback_language_module.directives[normname]
            msg_text.append('Using English fallback for directive "%s".'
                            % directive_name)
        except KeyError:
            msg_text.append('Trying "%s" as canonical directive name.'
                            % directive_name)
            # The canonical name should be an English name, but just in case:
            canonicalname = normname
    if msg_text:
        message = document.reporter.info(
            '\n'.join(msg_text), line=document.current_line)
        messages.append(message)
    try:
        modulename, classname = _directive_registry[canonicalname]
    except KeyError:
        # Error handling done by caller.
        return None, messages
    try:
        module = __import__(modulename, globals(), locals(), level=1)
    except ImportError, detail:
        messages.append(document.reporter.error(
            'Error importing directive module "%s" (directive "%s"):\n%s'
            % (modulename, directive_name, detail),
            line=document.current_line))
        return None, messages
    try:
        directive = getattr(module, classname)
        _directives[normname] = directive
    except AttributeError:
        messages.append(document.reporter.error(
            'No directive class "%s" in module "%s" (directive "%s").'
            % (classname, modulename, directive_name),
            line=document.current_line))
        return None, messages
    return directive, messages

def register_directive(name, directive):
    """
    Register a nonstandard application-defined directive function.
    Language lookups are not needed for such functions.
    """
    _directives[name] = directive

def flag(argument):
    """
    Check for a valid flag option (no argument) and return ``None``.
    (Directive option conversion function.)

    Raise ``ValueError`` if an argument is found.
    """
    if argument and argument.strip():
        raise ValueError('no argument is allowed; "%s" supplied' % argument)
    else:
        return None

def unchanged_required(argument):
    """
    Return the argument text, unchanged.
    (Directive option conversion function.)

    Raise ``ValueError`` if no argument is found.
    """
    if argument is None:
        raise ValueError('argument required but none supplied')
    else:
        return argument  # unchanged!

def unchanged(argument):
    """
    Return the argument text, unchanged.
    (Directive option conversion function.)

    No argument implies empty string ("").
    """
    if argument is None:
        return u''
    else:
        return argument  # unchanged!

def path(argument):
    """
    Return the path argument unwrapped (with newlines removed).
    (Directive option conversion function.)

    Raise ``ValueError`` if no argument is found.
    """
    if argument is None:
        raise ValueError('argument required but none supplied')
    else:
        path = ''.join([s.strip() for s in argument.splitlines()])
        return path

def uri(argument):
    """
    Return the URI argument with unescaped whitespace removed.
    (Directive option conversion function.)

    Raise ``ValueError`` if no argument is found.
    """
    if argument is None:
        raise ValueError('argument required but none supplied')
    else:
        parts = split_escaped_whitespace(escape2null(argument))
        uri = ' '.join(''.join(unescape(part).split()) for part in parts)
        return uri

def nonnegative_int(argument):
    """
    Check for a nonnegative integer argument; raise ``ValueError`` if not.
    (Directive option conversion function.)
    """
    value = int(argument)
    if value < 0:
        raise ValueError('negative value; must be positive or zero')
    return value

def percentage(argument):
    """
    Check for an integer percentage value with optional percent sign.
    """
    try:
        argument = argument.rstrip(' %')
    except AttributeError:
        pass
    return nonnegative_int(argument)

length_units = ['em', 'ex', 'px', 'in', 'cm', 'mm', 'pt', 'pc']

def get_measure(argument, units):
    """
    Check for a positive argument of one of the units and return a
    normalized string of the form "<value><unit>" (without space in
    between).

    To be called from directive option conversion functions.
    """
    match = re.match(r'^([0-9.]+) *(%s)$' % '|'.join(units), argument)
    try:
        float(match.group(1))
    except (AttributeError, ValueError):
        raise ValueError(
            'not a positive measure of one of the following units:\n%s'
            % ' '.join(['"%s"' % i for i in units]))
    return match.group(1) + match.group(2)

def length_or_unitless(argument):
    return get_measure(argument, length_units + [''])

def length_or_percentage_or_unitless(argument, default=''):
    """
    Return normalized string of a length or percentage unit.

    Add <default> if there is no unit. Raise ValueError if the argument is not
    a positive measure of one of the valid CSS units (or without unit).

    >>> length_or_percentage_or_unitless('3 pt')
    '3pt'
    >>> length_or_percentage_or_unitless('3%', 'em')
    '3%'
    >>> length_or_percentage_or_unitless('3')
    '3'
    >>> length_or_percentage_or_unitless('3', 'px')
    '3px'
    """
    try:
        return get_measure(argument, length_units + ['%'])
    except ValueError:
        try:
            return get_measure(argument, ['']) + default
        except ValueError:
            # raise ValueError with list of valid units:
            return get_measure(argument, length_units + ['%'])

def class_option(argument):
    """
    Convert the argument into a list of ID-compatible strings and return it.
    (Directive option conversion function.)

    Raise ``ValueError`` if no argument is found.
    """
    if argument is None:
        raise ValueError('argument required but none supplied')
    names = argument.split()
    class_names = []
    for name in names:
        class_name = nodes.make_id(name)
        if not class_name:
            raise ValueError('cannot make "%s" into a class name' % name)
        class_names.append(class_name)
    return class_names

unicode_pattern = re.compile(
    r'(?:0x|x|\\x|U\+?|\\u)([0-9a-f]+)$|&#x([0-9a-f]+);$', re.IGNORECASE)

def unicode_code(code):
    r"""
    Convert a Unicode character code to a Unicode character.
    (Directive option conversion function.)

    Codes may be decimal numbers, hexadecimal numbers (prefixed by ``0x``,
    ``x``, ``\x``, ``U+``, ``u``, or ``\u``; e.g. ``U+262E``), or XML-style
    numeric character entities (e.g. ``&#x262E;``).  Other text remains as-is.

    Raise ValueError for illegal Unicode code values.
    """
    try:
        if code.isdigit():                  # decimal number
            return unichr(int(code))
        else:
            match = unicode_pattern.match(code)
            if match:                       # hex number
                value = match.group(1) or match.group(2)
                return unichr(int(value, 16))
            else:                           # other text
                return code
    except OverflowError, detail:
        raise ValueError('code too large (%s)' % detail)

def single_char_or_unicode(argument):
    """
    A single character is returned as-is.  Unicode characters codes are
    converted as in `unicode_code`.  (Directive option conversion function.)
    """
    char = unicode_code(argument)
    if len(char) > 1:
        raise ValueError('%r invalid; must be a single character or '
                         'a Unicode code' % char)
    return char

def single_char_or_whitespace_or_unicode(argument):
    """
    As with `single_char_or_unicode`, but "tab" and "space" are also supported.
    (Directive option conversion function.)
    """
    if argument == 'tab':
        char = '\t'
    elif argument == 'space':
        char = ' '
    else:
        char = single_char_or_unicode(argument)
    return char

def positive_int(argument):
    """
    Converts the argument into an integer.  Raises ValueError for negative,
    zero, or non-integer values.  (Directive option conversion function.)
    """
    value = int(argument)
    if value < 1:
        raise ValueError('negative or zero value; must be positive')
    return value

def positive_int_list(argument):
    """
    Converts a space- or comma-separated list of values into a Python list
    of integers.
    (Directive option conversion function.)

    Raises ValueError for non-positive-integer values.
    """
    if ',' in argument:
        entries = argument.split(',')
    else:
        entries = argument.split()
    return [positive_int(entry) for entry in entries]

def encoding(argument):
    """
    Verfies the encoding argument by lookup.
    (Directive option conversion function.)

    Raises ValueError for unknown encodings.
    """
    try:
        codecs.lookup(argument)
    except LookupError:
        raise ValueError('unknown encoding: "%s"' % argument)
    return argument

def choice(argument, values):
    """
    Directive option utility function, supplied to enable options whose
    argument must be a member of a finite set of possible values (must be
    lower case).  A custom conversion function must be written to use it.  For
    example::

        from docutils.parsers.rst import directives

        def yesno(argument):
            return directives.choice(argument, ('yes', 'no'))

    Raise ``ValueError`` if no argument is found or if the argument's value is
    not valid (not an entry in the supplied list).
    """
    try:
        value = argument.lower().strip()
    except AttributeError:
        raise ValueError('must supply an argument; choose from %s'
                         % format_values(values))
    if value in values:
        return value
    else:
        raise ValueError('"%s" unknown; choose from %s'
                         % (argument, format_values(values)))

def format_values(values):
    return '%s, or "%s"' % (', '.join(['"%s"' % s for s in values[:-1]]),
                            values[-1])

def value_or(values, other):
    """
    The argument can be any of `values` or `argument_type`.
    """
    def auto_or_other(argument):
        if argument in values:
            return argument
        else:
            return other(argument)
    return auto_or_other