#!/usr/bin/env python

"""

Parsing of vCard, vCalendar and iCalendar files.

Copyright (C) 2005, 2006, 2007, 2008, 2009, 2011, 2013,

              2014, 2015, 2017 Paul Boddie <paul@boddie.org.uk>

This program is free software; you can redistribute it and/or modify it under

the terms of the GNU General Public License as published by the Free Software

Foundation; either version 3 of the License, or (at your option) any later

version.

This program is distributed in the hope that it will be useful, but WITHOUT

ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS

FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more

details.

You should have received a copy of the GNU General Public License along with

this program.  If not, see <http://www.gnu.org/licenses/>.

--------

References:

RFC 5545: Internet Calendaring and Scheduling Core Object Specification

          (iCalendar)

          http://tools.ietf.org/html/rfc5545

RFC 2445: Internet Calendaring and Scheduling Core Object Specification

          (iCalendar)

          http://tools.ietf.org/html/rfc2445

RFC 2425: A MIME Content-Type for Directory Information

          http://tools.ietf.org/html/rfc2425

RFC 2426: vCard MIME Directory Profile

          http://tools.ietf.org/html/rfc2426

"""

try:

    set

except NameError:

    from sets import Set as set

# Encoding-related imports.

import base64, quopri

import codecs

# Tokenisation help.

import re

# Configuration.

default_encoding = "utf-8"

class ParseError(Exception):

    "General parsing errors."

    pass

class WriteError(Exception):

    "General writing errors."

    pass

# Reader and parser classes.

class Reader:

    "A simple class wrapping a file, providing simple pushback capabilities."

    def __init__(self, f, non_standard_newline=0):

        """

        Initialise the object with the file 'f'. If 'non_standard_newline' is

        set to a true value (unlike the default), lines ending with CR will be

        treated as complete lines.

        """

        self.f = f

        self.non_standard_newline = non_standard_newline

        self.lines = []

        self.line_number = 1 # about to read line 1

    def close(self):

        "Close the reader."

        self.f.close()

    def pushback(self, line):

        """

        Push the given 'line' back so that the next line read is actually the

        given 'line' and not the next line from the underlying file.

        """

        self.lines.append(line)

        self.line_number -= 1

    def readline(self):

        """

        If no pushed-back lines exist, read a line directly from the file.

        Otherwise, read from the list of pushed-back lines.

        """

        self.line_number += 1

        if self.lines:

            return self.lines.pop()

        else:

            # Sanity check for broken lines (\r instead of \r\n or \n).

            line = self.f.readline()

            while line.endswith("\r") and not self.non_standard_newline:

                s = self.f.readline()

                if not s:

                    break

                line += s

            if line.endswith("\r") and self.non_standard_newline:

                return line + "\n"

            else:

                return line

    def read_content_line(self):

        """

        Read an entire content line, itself potentially consisting of many

        physical lines of text, returning a string.

        """

        # Skip blank lines.

        line = self.readline()

        while line:

            line_stripped = line.rstrip("\r\n")

            if not line_stripped:

                line = self.readline()

            else:

                break

        else:

            return ""

        # Strip all appropriate whitespace from the right end of each line.

        # For subsequent lines, remove the first whitespace character.

        # See section 4.1 of the iCalendar specification.

        lines = [line_stripped]

        line = self.readline()

        while line.startswith(" ") or line.startswith("\t"):

            lines.append(line[1:].rstrip("\r\n"))

            line = self.readline()

        # Since one line too many will have been read, push the line back into

        # the file.

        if line:

            self.pushback(line)

        return "".join(lines)

    def get_content_line(self):

        "Return a content line object for the current line."

        return ContentLine(self.read_content_line())

class ContentLine:

    "A content line which can be searched."

    SEPARATORS = re.compile('[;:"]')

    SEPARATORS_PLUS_EQUALS = re.compile('[=;:"]')

    def __init__(self, text):

        self.text = text

        self.start = 0

    def __repr__(self):

        return "ContentLine(%r)" % self.text

    def get_remaining(self):

        "Get the remaining text from the content line."

        return self.text[self.start:]

    def search(self, targets):

        """

        Find one of the 'targets' in the text, returning the string from the

        current position up to the target found, along with the target string,

        using a tuple of the form (string, target). If no target was found,

        return the entire string together with a target of None.

        The 'targets' parameter must be a regular expression object or an object

        compatible with the API of such objects.

        """

        text = self.text

        start = pos = self.start

        length = len(text)

        # Remember the first target.

        first = None

        first_pos = None

        in_quoted_region = 0

        # Process the text, looking for the targets.

        while pos < length:

            match = targets.search(text, pos)

            # Where nothing matches, end the search.

            if match is None:

                pos = length

            # Where a double quote matches, toggle the region state.

            elif match.group() == '"':

                in_quoted_region = not in_quoted_region

                pos = match.end()

            # Where something else matches outside a region, stop searching.

            elif not in_quoted_region:

                first = match.group()

                first_pos = match.start()

                break

            # Otherwise, keep looking for the end of the region.

            else:

                pos = match.end()

        # Where no more input can provide the targets, return a special result.

        else:

            self.start = length

            return text[start:], None

        self.start = match.end()

        return text[start:first_pos], first

class StreamParser:

    "A stream parser for content in vCard/vCalendar/iCalendar-like formats."

    def __init__(self, f):

        "Initialise the parser for the given file 'f'."

        self.f = f

    def close(self):

        "Close the reader."

        self.f.close()

    def __iter__(self):

        "Return self as the iterator."

        return self

    def next(self):

        """

        Return the next content item in the file as a tuple of the form

        (name, parameters, values).

        """

        return self.parse_content_line()

    def decode_content(self, name, value):

        """

        Decode for property 'name' the given 'value', replacing quoted

        characters.

        """

        return value.replace("\r", "").replace("\\N", "\n").replace("\\n", "\n")

    # Internal methods.

    def parse_content_line(self):

        """

        Return the name, parameters and value information for the current

        content line in the file being parsed.

        """

        f = self.f

        line_number = f.line_number

        line = f.get_content_line()

        # Read the property name.

        name, sep = line.search(line.SEPARATORS)

        name = name.strip()

        if not name and sep is None:

            raise StopIteration

        # Read the parameters.

        parameters = {}

        while sep == ";":

            # Find the actual modifier.

            parameter_name, sep = line.search(line.SEPARATORS_PLUS_EQUALS)

            parameter_name = parameter_name.strip()

            if sep == "=":

                parameter_value, sep = line.search(line.SEPARATORS)

                parameter_value = parameter_value.strip()

            else:

                parameter_value = None

            # Append a key, value tuple to the parameters list.

            parameters[parameter_name] = parameter_value

        # Get the value content.

        if sep != ":":

            raise ValueError, (line_number, line)

        # Obtain and decode the value.

        value = self.decode(name, parameters, line.get_remaining())

        return name, parameters, value

    def decode(self, name, parameters, value):

        "Decode using 'name' and 'parameters' the given 'value'."

        encoding = parameters.get("ENCODING")

        charset = parameters.get("CHARSET")

        value = self.decode_content(name, value)

        if encoding == "QUOTED-PRINTABLE":

            return unicode(quopri.decodestring(value), charset or "iso-8859-1")

        elif encoding == "BASE64":

            return base64.decodestring(value)

        else:

            return value

class ParserBase:

    "An abstract parser for content in vCard/vCalendar/iCalendar-like formats."

    def __init__(self):

        "Initialise the parser."

        self.names = []

    def parse(self, f, parser_cls=None):

        "Parse the contents of the file 'f'."

        parser = (parser_cls or StreamParser)(f)

        for name, parameters, value in parser:

            if name == "BEGIN":

                self.names.append(value)

                self.startComponent(value, parameters)

            elif name == "END":

                start_name = self.names.pop()

                if start_name != value:

                    raise ParseError, "Mismatch in BEGIN and END declarations (%r and %r) at line %d." % (

                        start_name, value, f.line_number)

                self.endComponent(value)

            else:

                self.handleProperty(name, parameters, value)

class Parser(ParserBase):

    "A SAX-like parser for vCard/vCalendar/iCalendar-like formats."

    def __init__(self):

        ParserBase.__init__(self)

        self.components = []

    def startComponent(self, name, parameters):

        """

        Add the component with the given 'name' and 'parameters', recording an

        empty list of children as part of the component's content.

        """

        component = self.handleProperty(name, parameters)

        self.components.append(component)

        return component

    def endComponent(self, name):

        """

        End the component with the given 'name' by removing it from the active

        component stack. If only one component exists on the stack, retain it

        for later inspection.

        """

        if len(self.components) > 1:

            return self.components.pop()

        # Or return the only element.

        elif self.components:

            return self.components[0]

    def handleProperty(self, name, parameters, value=None):

        """

        Record the property with the given 'name', 'parameters' and optional

        'value' as part of the current component's children.

        """

        component = self.makeComponent(name, parameters, value)

        self.attachComponent(component)

        return component

    # Component object construction/manipulation methods.

    def attachComponent(self, component):

        "Attach the given 'component' to its parent."

        if self.components:

            component_name, component_parameters, component_children = self.components[-1]

            component_children.append(component)

    def makeComponent(self, name, parameters, value=None):

        """

        Make a component object from the given 'name', 'parameters' and optional

        'value'.

        """

        return (name, parameters, value or [])

    # Public methods.

    def parse(self, f, parser_cls=None):

        "Parse the contents of the file 'f'."

        ParserBase.parse(self, f, parser_cls)

        try:

            return self.components[0]

        except IndexError:

            raise ParseError, "No vContent component found in file."

# Writer classes.

class Writer:

    "A simple class wrapping a file, providing simple output capabilities."

    default_line_length = 76

    def __init__(self, write, line_length=None):

        """

        Initialise the object with the given 'write' operation. If 'line_length'

        is set, the length of written lines will conform to the specified value

        instead of the default value. 

        """

        self._write = write

        self.line_length = line_length or self.default_line_length

        self.char_offset = 0

    def write(self, text):

        "Write the 'text' to the file."

        write = self._write

        line_length = self.line_length

        i = 0

        remaining = len(text)

        while remaining:

            space = line_length - self.char_offset

            if remaining > space:

                write(text[i:i + space])

                write("\r\n ")

                self.char_offset = 1

                i += space

                remaining -= space

            else:

                write(text[i:])

                self.char_offset += remaining

                i += remaining

                remaining = 0

    def end_line(self):

        "End the current content line."

        if self.char_offset > 0:

            self.char_offset = 0

            self._write("\r\n")

class StreamWriter:

    "A stream writer for content in vCard/vCalendar/iCalendar-like formats."

    def __init__(self, f):

        "Initialise the stream writer with the given 'f' stream object."

        self.f = f

    def append(self, record):

        self.write(*record)

    def write(self, name, parameters, value):

        """

        Write a content line, serialising the given 'name', 'parameters' and

        'value' information.

        """

        self.write_content_line(name, self.encode_parameters(parameters), self.encode_value(name, parameters, value))

    # Internal methods.

    def write_content_line(self, name, encoded_parameters, encoded_value):

        """

        Write a content line for the given 'name', 'encoded_parameters' and

        'encoded_value' information.

        """

        f = self.f

        f.write(name)

        for param_name, param_value in encoded_parameters.items():

            f.write(";")

            f.write(param_name)

            f.write("=")

            f.write(param_value)

        f.write(":")

        f.write(encoded_value)

        f.end_line()

    def encode_quoted_parameter_value(self, value):

        "Encode the given 'value'."

        return '"%s"' % value

    def encode_value(self, name, parameters, value):

        """

        Encode using 'name' and 'parameters' the given 'value' so that the

        resulting encoded form employs any specified character encodings.

        """

        encoding = parameters.get("ENCODING")

        charset = parameters.get("CHARSET")

        try:

            if encoding == "QUOTED-PRINTABLE":

                value = quopri.encodestring(value.encode(charset or "iso-8859-1"))

            elif encoding == "BASE64":

                value = base64.encodestring(value)

            return self.encode_content(name, value)

        except TypeError:

            raise WriteError, "Property %r value with parameters %r cannot be encoded: %r" % (name, parameters, value)

    # Overrideable methods.

    def encode_parameters(self, parameters):

        """

        Encode the given 'parameters' according to the vCalendar specification.

        """

        encoded_parameters = {}

        for param_name, param_value in parameters.items():

            # Basic format support merely involves quoting values which seem to

            # need it. Other more specific formats may define exactly which

            # parameters should be quoted.

            if ContentLine.SEPARATORS.search(param_value):

                param_value = self.encode_quoted_parameter_value(param_value)

            encoded_parameters[param_name] = param_value

        return encoded_parameters

    def encode_content(self, name, value):

        "Encode for property 'name' the given 'value', quoting characters."

        return (value or "").replace("\n", "\\n")

# Utility functions.

def is_input_stream(stream_or_string):

    return hasattr(stream_or_string, "read")

def get_input_stream(stream_or_string, encoding=None):

    if is_input_stream(stream_or_string):

        if isinstance(stream_or_string, codecs.StreamReader):

            return stream_or_string

        else:

            return codecs.getreader(encoding or default_encoding)(stream_or_string)

    else:

        return codecs.open(stream_or_string, encoding=(encoding or default_encoding))

def get_output_stream(stream_or_string, encoding=None):

    if hasattr(stream_or_string, "write"):

        if isinstance(stream_or_string, codecs.StreamWriter):

            return stream_or_string

        else:

            return codecs.getwriter(encoding or default_encoding)(stream_or_string)

    else:

        return codecs.open(stream_or_string, "w", encoding=(encoding or default_encoding))

def items_to_dict(items, sections=None):

    """

    Return the given 'items' as a dictionary mapping names to tuples of the form

    (value, attributes). Where 'sections' is provided, only items whose names

    occur in the given 'sections' collection will be treated as groups or

    sections of definitions.

    """

    d = {}

    for name, attr, value in items:

        if not d.has_key(name):

            d[name] = []

        if isinstance(value, list) and (not sections or name in sections):

            d[name].append((items_to_dict(value, sections), attr))

        else:

            d[name].append((value, attr))

    return d

def dict_to_items(d):

    """

    Return 'd' converted to a list of items suitable for serialisation using

    iterwrite.

    """

    items = []

    for name, value in d.items():

        if isinstance(value, list):

            for v, a in value:

                if isinstance(v, dict):

                    items.append((name, a, dict_to_items(v)))

                else:

                    items.append((name, a, v))

        else:

            v, a = value

            items.append((name, a, dict_to_items(v)))

    return items

# Public functions.

def parse(stream_or_string, encoding=None, non_standard_newline=0, parser_cls=None):

    """

    Parse the resource data found through the use of the 'stream_or_string',

    which is either a stream providing Unicode data (the codecs module can be

    used to open files or to wrap streams in order to provide Unicode data) or a

    filename identifying a file to be parsed.

    The optional 'encoding' can be used to specify the character encoding used

    by the file to be parsed.

    The optional 'non_standard_newline' can be set to a true value (unlike the

    default) in order to attempt to process files with CR as the end of line

    character.

    As a result of parsing the resource, the root node of the imported resource

    is returned.

    """

    stream = get_input_stream(stream_or_string, encoding)

    reader = Reader(stream, non_standard_newline)

    # Parse using the reader.

    try:

        parser = (parser_cls or Parser)()

        return parser.parse(reader)

    # Close any opened streams.

    finally:

        if not is_input_stream(stream_or_string):

            reader.close()

def iterparse(stream_or_string, encoding=None, non_standard_newline=0, parser_cls=None):

    """

    Parse the resource data found through the use of the 'stream_or_string',

    which is either a stream providing Unicode data (the codecs module can be

    used to open files or to wrap streams in order to provide Unicode data) or a

    filename identifying a file to be parsed.

    The optional 'encoding' can be used to specify the character encoding used

    by the file to be parsed.

    The optional 'non_standard_newline' can be set to a true value (unlike the

    default) in order to attempt to process files with CR as the end of line

    character.

    An iterator is returned which provides event tuples describing parsing

    events of the form (name, parameters, value).

    """

    stream = get_input_stream(stream_or_string, encoding)

    reader = Reader(stream, non_standard_newline)

    parser = (parser_cls or StreamParser)(reader)

    return parser

def iterwrite(stream_or_string=None, write=None, encoding=None, line_length=None, writer_cls=None):

    """

    Return a writer which will either send data to the resource found through

    the use of 'stream_or_string' or using the given 'write' operation.

    The 'stream_or_string' parameter may be either a stream accepting Unicode

    data (the codecs module can be used to open files or to wrap streams in

    order to accept Unicode data) or a filename identifying a file to be

    written.

    The optional 'encoding' can be used to specify the character encoding used

    by the file to be written.

    The optional 'line_length' can be used to specify how long lines should be

    in the resulting data.

    """

    if stream_or_string:

        stream = get_output_stream(stream_or_string, encoding)

        _writer = Writer(stream.write, line_length)

    elif write:

        _writer = Writer(write, line_length)

    else:

        raise IOError, "No stream, filename or write operation specified."

    return (writer_cls or StreamWriter)(_writer)

def to_dict(node, sections=None):

    "Return the 'node' converted to a dictionary representation."

    name, attr, items = node

    return {name : (isinstance(items, list) and items_to_dict(items, sections) or items, attr)}

def to_node(d):

    "Return 'd' converted to a items-based representation."

    return dict_to_items(d)[0]

# vim: tabstop=4 expandtab shiftwidth=4