#!/usr/bin/env python

"""

Parsing of vCard, vCalendar and iCalendar files.

Copyright (C) 2005, 2006, 2007, 2008 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 Lesser 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 Lesser General Public License for more

details.

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

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

--------

References:

RFC 2445: Internet Calendaring and Scheduling Core Object Specification

          (iCalendar)

          http://rfc.net/rfc2445.html

RFC 2425: A MIME Content-Type for Directory Information

          http://rfc.net/rfc2425.html

RFC 2426: vCard MIME Directory Profile

          http://rfc.net/rfc2426.html

"""

try:

    set

except NameError:

    from sets import Set as set

# Encoding-related imports.

import base64, quopri

# Tokenisation help.

import re

# 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 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:

            # NOTE: 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:

                line += self.f.readline()

            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.

        """

        line = self.readline()

        # 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.rstrip("\r\n")]

        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 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.

        """

        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 __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, value):

        "Decode 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

        # 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(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 len(self.components) > 1:

            return self.components.pop()

        elif self.components:

            return self.components[-1]

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

        """

        Record the property with the given 'name', 'parameters' and '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):

        """

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

        """

        return (name, parameters, value)

    # Public methods.

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

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

        ParserBase.parse(self, f, parser_cls)

        return self.components[0]

# Writer classes.

class Writer:

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

    default_line_length = 76

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

        """

        Initialise the object with the file 'f'. If 'line_length' is set, the

        length of written lines will conform to the specified value instead of

        the default value. 

        """

        self.f = f

        self.line_length = line_length or self.default_line_length

        self.char_offset = 0

    def write(self, text):

        "Write the 'text' to the file."

        f = self.f

        line_length = self.line_length

        i = 0

        remaining = len(text)

        while remaining:

            space = line_length - self.char_offset

            if remaining > space:

                f.write(text[i:i + space])

                f.write("\r\n ")

                self.char_offset = 1

                i += space

                remaining -= space

            else:

                f.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.f.write("\r\n")

class StreamWriter:

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

    def __init__(self, f):

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

        self.f = f

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

        """

        Write a content line for the given 'name', 'parameters' and 'value'

        information.

        """

        f = self.f

        f.write(name)

        for parameter_name, parameter_value in parameters.items():

            f.write(";")

            f.write(parameter_name)

            f.write("=")

            f.write(parameter_value)

        f.write(":")

        f.write(self.encode(name, parameters, value))

        f.end_line()

    def encode_content(self, value):

        "Encode the given 'value', quoting characters."

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

    # Internal methods.

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

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

        encoding = parameters.get("ENCODING")

        charset = parameters.get("CHARSET")

        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(value)

# Public functions.

def parse(f, non_standard_newline=0, parser_cls=None):

    """

    Parse the resource data found through the use of the file object 'f', which

    should provide Unicode data. (The codecs module can be used to open files or

    to wrap streams in order to provide Unicode data.)

    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.

    """

    reader = Reader(f, non_standard_newline)

    parser = (parser_cls or Parser)()

    return parser.parse(reader)

def iterparse(f, non_standard_newline=0, parser_cls=None):

    """

    Parse the resource data found through the use of the file object 'f', which

    should provide Unicode data. (The codecs module can be used to open files or

    to wrap streams in order to provide Unicode data.)

    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).

    """

    reader = Reader(f, non_standard_newline)

    parser = (parser_cls or StreamParser)(reader)

    return iter(parser)

def iterwrite(f, line_length=None, writer_cls=None):

    _writer = Writer(f, line_length)

    writer = (writer_cls or StreamWriter)(_writer)

    return writer

# vim: tabstop=4 expandtab shiftwidth=4