#!/usr/bin/env python
# vim: tabstop=4: shiftwidth=4: expandtab: softtabstop=4: autoindent:

"""
   Copyright (C) 2007 - 2009 by Mark de Wever <koraq@xs4all.nl>
   Part of the Battle for Wesnoth Project http://www.wesnoth.org/

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License.
   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY.

   See the COPYING file for more details.


The wiki grabber is a tool to convert wiki comment formatting[1] into a text
page which can be used in the wiki.

[1] http://wesnoth.org/wiki/Wiki_grabber
"""
from __future__ import with_statement     # For python < 2.6

import operator
import os
import sys
import re

try:
    import argparse
except ImportError:
    print 'Please install argparse by running "easy_install argparse"'
    sys.exit(1)

if __name__ == "__main__":
    # setup and parse command line arguments
    # The defaults are set to the values of the older hardcoded implementation.
    parser = argparse.ArgumentParser(description='The wiki grabber is a tool'
            + ' to convert wiki comment formatting into a text page which can'
            + ' be used in the wiki. For more details, see'
            + ' http://wesnoth.org/wiki/Wiki_grabber')
    parser.add_argument('-s', '--src-dir', default='../src/gui', dest='src_dir',
            help="the location of wesnoth's source code")
    parser.add_argument('-o', '--output', default='/tmp/', dest='output_dir',
            help='the output directory')
    args = parser.parse_args()

    # contains all output generated:
    # - key filename
    # - value node list
    #
    # every node is a list with 2 items
    # - first the sorting order
    # - second the actual data
    file_map = {}

    # contains all macros:
    # - key macro name
    # - value macro contents
    macro_map = {}

    output_directory = args.output_dir
    src_directory = args.src_dir

    if not os.path.isdir(output_directory):
        raise IOError("'%s' isn't a directory." % output_directory)
    if not os.path.isdir(src_directory):
        raise IOError("'%s' isn't a directory." % src_directory)

    # current file being processed
    current_file = ""

    # current block being processed
    current_block = ""

    re_record_start = '^\s*'
    # There must be whitespace between the field separator (&).
    # However if there is an empty field eg '...foo & & bar...'
    # The first field separator eats the whitespace before the second
    # which would \s+ cause to fail.
    # The solution is to make the whitespace optional and assert the
    # character before the ampersand is whitespace.
    re_field_separator = '\s*(?:(?<=\s))&\s+'
    # Same issue as with the re_field_separator but then with '...foo & $'
    re_record_end = '\s*(?:(?<=\s))\$$'

    re_variable = '([a-zA-Z]\w*)'
    re_string = '(.+?)'

    def is_empty(res, data):
        """
        This checks whether or not a table is empty and writes to stderr if it is.
        It returns True if the table is empty, False otherwise.
        """
        if not res:
            sys.stderr.write("Empty container:\n%s\n" % data)
            return True
        return False

    def reindent(data):
        """Converts the raw input to an easier to use format.

        Lines starting with 8 spaces are concatenated with the previous line.
        The start of line ' *' are removed and if another space exists it's also
        removed."""

        # concatenate
        data = re.sub(r'\n \*(?: ){8,}', " ", data)

        # strip
        data = re.sub(" \*(?: |)", "", data)

        #annotation
        data = re.sub(r'@(?:begin|end|allow|remove)\{(?:parent|tag|link|global|type|key)\}(?:\{.*\})', "", data)
        return data

    def get_value(data, key):
        """
        Extracts data from a key value pair, a key must start at the start of
        a line.
        """

        res = re.compile("^" + key + " *= *(.*)$", re.M).search(data)

        if res != None:
            res = res.group(1)

        return res

    def process_header(data):
        """Processes the header."""

        page = get_value(data, "@page")
        order = get_value(data, "@order")
        if order == None:
            order = 10000

        return [page, order]

    def debug_dump(data, res):
        """Show the data the regex retrieved from a match.

        data is the raw data the regex tried to match.
        res is the result of the regex.findall.
        """
        sys.stderr.write("data : %s" % data)
        for i, val in res:
            for j, sub_val in val:
                sys.stderr.write("Line %s match %s: %s\n" % (i, j, sub_val))

    def format(data):
        """Formats the data for the wiki.

        Replaces the following:
        -  An end of line and its surrounding whitespace to a single space.
        - @$ -> $
        - @& -> &
        - @* -> \n* needed in a list.
        - @- -> \n needed to add text after a list.
        """
        data = re.sub(r'\s*\n\s*', ' ', data)
        data = re.sub(r'@\$', '$', data)
        data = re.sub(r'@&', '&', data)
        data = re.sub(r'@\*', "\n*", data)
        data = re.sub(r'@\-', "\n", data)
        return data

    def create_config_table(data):
        """Creates a table for data in a config table.

        A config table is a table with info about WML configuration key value pairs.
        """

        # matches a line like
        # x1 & f_unsigned & 0 &            The x coordinate of the
        # startpoint. $
        # x1 & f_unsigned & &              The x coordinate of the
        # startpoint. $
        regex = re.compile("([A-Za-z]\w*) +& +([A-Za-z]\w*) +& +([^&]*?) *& +(.*) +\$")
        res = regex.findall(data)

        regex  = re_record_start
        regex += re_variable           # 0 variable id
        regex += re_field_separator
        regex += re_variable           # 1 variable type
        regex += re_field_separator
        regex += '(.*?)'               # 2 optional default value, if omitted mandatory field
        regex += re_field_separator
        regex += re_string             # 3 description
        regex += re_record_end

        regex = re.compile(regex, re.DOTALL | re.MULTILINE)
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty table."

        result = '{| border="1"'
        result += "\n!key\n!type\n!default\n!description\n"
        for i in res:
            result += "|-\n"
            result += "| %s\n" % i[0]
            result += "| [[GUIVariable#%s|%s]]\n" % (i[1], i[1])
            if not i[2]:
                result += "| mandatory\n"
            else:
                result += "| %s\n" % i[2]
            result += "| %s\n" % format(i[3])
        result += "|}"

        return result

    def create_formula_table(data):
        """Creates a table for data in a formula table.

        A formula table is a table with info about which function parameters
        are available for processing in WML formulas.
        """
        regex  = re_record_start
        regex += re_variable           # 0 variable id
        regex += re_field_separator
        regex += re_variable           # 1 variable type
        regex += re_field_separator
        regex += re_string             # 2 description
        regex += re_record_end

        regex = re.compile(regex, re.DOTALL | re.MULTILINE)
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty table."

        result = '{| border="1"'
        result += "\n!Variable\n!type\n!description\n"
        for i in res:
            result += "|-\n"
            result += "| %s\n" % i[0]
            result += "| %s\n" % i[1]
            result += "| %s\n" % format(i[2])
        result += "|}"

        return result

    def create_variable_types_table(data):
        """Creates a table for the variable types."""

        regex  = re_record_start
        regex += re_variable           # 0 variable type
        regex += re_field_separator
        regex += re_string             # 1 description
        regex += re_record_end

        regex = re.compile(regex, re.DOTALL | re.MULTILINE)
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty table."

        result = '{| border="1"'
        result += "\n!Variable\n!description\n"
        for i in res:
            result += "|-\n"
            result += '| <span id="%s">%s</span>\n' % (i[0], i[0])
            result += "| %s\n" % format(i[1])
        result += "|}"

        return result

    def create_widget_overview_table(data):
        """Creates a table for all available widgets."""

        regex  = re_record_start
        regex += re_variable           # widget type
        regex += re_field_separator
        regex += re_string             # description
        regex += re_record_end
        regex = re.compile(regex, re.DOTALL | re.MULTILINE)
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty table."

        result = '{| border="1"'
        result += "\n!Section\n!Description\n"
        for i in res:
            result += "|-\n"
            result += '| <span id="%s">' % i[0].lower()
            result += re.sub(r'_', ' ', i[0])
            result += "</span>"
            result += " ([[GUIWidgetDefinitionWML#%s|definition]]" % i[0]
            result += ", [[GUIWidgetInstanceWML#%s|instantiation]])\n" % i[0]
            result += "| %s\n" % format(i[1])
        result += "|}"

        return result

    def create_window_overview_table(data):
        """Creates a table for all available windows."""

        regex  = re_record_start
        regex += re_variable                # 0 window id
        regex += re_field_separator
        regex += re_string                  # 1 description
        regex += re_record_end

        regex = re.compile(regex, re.DOTALL | re.MULTILINE)
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty table."

        result = '{| border="1"'
        result += "\n!Section\n!Description\n"
        for i in res:
            result += "|-\n"
            result += "| " + re.sub(r'_', ' ', i[0])
            result += " ([[GUIWindowDefinitionWML#%s|definition]])\n" % i[0]
            result += "| %s\n" % format(i[1])
        result += "|}"

        return result

    def create_dialog_widgets_table(data):
        """Creates a table for the widgets in a dialog."""

        regex  = re_record_start
        regex += '(-*)'                     # 0 indention markers
        regex += '((?:[a-zA-Z_]?)\w*)'      # 1 optional id may start with an underscore
        regex += re_field_separator
        regex += '(.*?)'                    # 2 optional retval
        regex += re_field_separator
        regex += re_variable                # 3 type
        regex += re_field_separator
        regex += '([om])'                   # 4 mandatory flag
        regex += re_field_separator
        regex += re_string                  # 5 description
        regex += re_record_end

        regex = re.compile(regex, re.DOTALL | re.MULTILINE)
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty table."

        result = '{| border="1"'
        result += "\n!ID (return value)\n!Type\n!Mandatory\n!Description\n"
        for i in res:
            result += "|-\n| " + "&nbsp;" * len(i[0]) * 8

            if not i[1]:
                result += "''free to choose''"
            else:
                result += i[1]

            if not i[2]:
                result += "\n"
            else:
                result += " (%s)\n" % i[2]

            result += "| [[GUIToolkitWML#%s|%s]]\n" % (i[3], i[3])

            if i[4] == 'm':
                result += "| yes\n"
            else:
                result += "| no\n"

            result += "| %s\n" % format(i[5])

        result += "|}"

        return result

    def validate_table(table):
        """Validates a table.

        At the moments tests for whitespace around separators."""

        regex = '((?<![@\s])&)|([^@]&(?!\s))'
        regex = re.compile(regex)
        invalid_field_separators = regex.findall(table)

        # Not the last test should actually be (\$(?![\s\Z]) but that fails as
        # work-around add a space to the input so when it would match the end
        # of the string it now matches the added space.
        regex = '((?<![@\s])\$)|([^@]\$(?!\s))'
        regex = re.compile(regex)
        invalid_record_terminator = regex.findall(table + ' ')

        if len(invalid_field_separators) or len(invalid_record_terminator):
            result = "Found %i invalid field separators " % (len(invalid_field_separators))
            result += "and %i invalid record separators " % (len(invalid_record_terminator))
            result += "in:\n%s\n\n" % (table)
            return result
        else:
            return None

    def create_table(table):
        """Wrapper for creating tables."""

        type = table.group(1)

        errors = validate_table(table.group(2))
        if errors:
            sys.stderr.write(errors)

        functions = {
        "config": create_config_table,
        "formula": create_formula_table,
        "variable_types": create_variable_types_table,
        "widget_overview": create_widget_overview_table,
        "window_overview": create_window_overview_table,
        "dialog_widgets": create_dialog_widgets_table}

        try:
            return functions[type](table.group(2) + "\n")
        except KeyError:
            sys.stderr.write("Unknown table definition '%s'.\n" % type)
            return "Unknown table definition '%s'." % type

    def create_wml_reference_description(data, nest_level):
        """ Creates a description for wml reference purposes."""
        # Matches a line like:
        # name & string & "" & 1.5 & description
        # [relative] & node & * & & description

        regex = re.compile("(.*) +& +(.*) +& *(.*) *& +(.*) *& *(.*)")
        res = regex.findall(data)

        if is_empty(res, data):
            return "Empty description"

        i = res[0]
        result = "%s '''''%s''''' " % ("*" * nest_level, i[0].strip())
        if i[1] != 'node':
            result += "(%s, ''%s'') " % (i[1], "default " + i[2].rstrip() if i[2] else "'mandatory'")
        else:
            result += "(%s) " % {'*': "zero or more times", '+': "one or more times", '?': "zero or one times", '1': "one time", '': "empty string bug."}[i[2].rstrip()]
        result += "%s" % i[4]
        if i[3]:
            result += " {{DevFeature%s}}" % i[3].replace(' ', '')
        return result + "\n"

    def create_description(description):
        """Wrapper for creating descriptions."""

        # List of descriptions used.
        descriptions = {
            "wml_reference": create_wml_reference_description}

        # We use a list because it's easier to handle nested elements
        type = [description.group(1)]

        # The $ are the separators between rows of a description
        parseme = description.group(2).split("$")
        # We strip everything that we don't need
        parseme = [i.replace("\n", "").strip() for i in parseme]

        nest_level = 1
        data = ""

        for i in parseme:

            # If it's a new block, we need to handle it.
            if i.startswith("@begin"):
                nest_level += 1
                i = i.lstrip("@begin{description}{")
                type += [i[:i.find("}")]]
                i = i.strip(type[nest_level - 1] + "}")

            # We also handle end of blocks.
            elif i.startswith("@end") and i != "@end{description}":
                nest_level -= 1
                i = i.lstrip("@end{description}")
                type = type[:-1]

            # Send content unless special "@end" case.
            if not i == "@end{description}":
                data += descriptions[type[nest_level-1]](i + "\n", nest_level)

        return data

    def process_body(data):
        """Process the body.

        The body needs to be stripped of known markup values.
        """

        table_regex = re.compile("^@begin{table}\{(.*?)\}\n(.*?)\n@end{table}$", re.M | re.S)
        data = table_regex.sub(lambda match: create_table(match), data)

        description_regex = re.compile("^@begin{description}\{(.*?)\}\n(.*?)\n@end{description}$", re.M | re.S)
        data = description_regex.sub(lambda match: create_description(match), data)
        return data

    def process(data):
        """Processes a wiki block."""

        data = replace_macros(data)

        res = re.compile("(.*?)\n\n(.*)", re.S).findall(data)
        if res:
            header = process_header(res[0][0])
            body = process_body(res[0][1])
        else:
            print "File: " + current_file
            print "Block:\n" + current_block
            print "\n\nInvalid wiki block, discarded."
            return

        if not header[0]:
            print "File: " + current_file
            print "Block:\n" + current_block
            print "\n\nNo page defined, dropped."
            return

        if not header[0] in file_map:
            file_map[header[0]] = []

        file_map[header[0]].append([header[1], body])

    def create_output():
        """Generates the output"""

        for file, data_list in file_map.iteritems():
            data_list.sort(key=operator.itemgetter(0))
            with open(os.path.join(output_directory, file), "w") as fd:
                for i in data_list:
                    fd.write(i[1])

    def process_file(name):
        """Processes all wiki blocks (if any) of a file."""

        global current_file
        current_file = name
        with open(name, "r") as file:
            data = file.read()

        regex = re.compile("(/\*WIKI($.*?)^ \*/)", re.M | re.S)
        res = regex.findall(data)
        if res:
            for i in res:
                global current_block
                current_block = i[0]
                section = reindent(i[1])
                wiki_info = process(section)

    def process_directory(dir):
        """Processes a directory.

        Recurses into the sub directory, ignoring hidden directories.
        Processes every .[c|h]pp file.
        """

        items = os.listdir(dir)

        for item in items:

            # Ignore hidden directories.
            if item.startswith("."):
                continue

            if os.path.isdir(os.path.join(dir, item)):
                process_directory(os.path.join(dir, item))
            elif item.endswith(".cpp") or item.endswith(".hpp"):
                process_file(os.path.join(dir, item))

    ##### ##### ##### MACRO PROCESSING ##### ##### #####

    def instanciate_macro(macro):
        """Replaces a macro text by the macro value."""

        global macro_map

        if not macro.group(1) in macro_map:
            print "Macro '%s' is not defined." % macro.group(1)
            return macro.group(0)

        return macro_map[macro.group(1)]

    def replace_macros(data):
        """Replaces all macros found in the data.

        If a macro isn't defined, the text is left as is."""

        macro_regex = re.compile("@macro *= *(\w+)", re.M | re.S)
        data = macro_regex.sub(lambda match: instanciate_macro(match), data)
        return data

    def create_macro_old(macro):
        print "Found old style macro '%s'" % macro.group(1)
        create_macro(macro)

    def create_macro(macro):
        """Wrapper for creating macros."""

        global macro_map

        if macro.group(1) in macro_map:
            print "Macro '%s' is being redefined." % macro.group(1)

        macro_map[macro.group(1)] = macro.group(2)

    def process_file_macros(name):
        """Processes all wiki macro blocks (if any) of a file."""

        global current_file
        current_file = name
        with open(name, "r") as file:
            data = file.read()

        regex = re.compile("(/\*WIKI_MACRO($.*?)^ \*/)", re.M | re.S)
        res = regex.findall(data)
        if res:
            for i in res:
                global current_block
                current_block = i[0]
                section = reindent(i[1])
                macro_regex = re.compile("^@start_macro *= *(.*?)\n(.*?)\n@end_macro.*?$", re.M | re.S)
                macro_regex.sub(lambda match: create_macro_old(match), section)
                macro_regex = re.compile("^@begin{macro}{(.*?)}\n(.*?)\n@end{macro}.*?$", re.M | re.S)
                macro_regex.sub(lambda match: create_macro(match), section)

    def process_directory_macros(dir):
        """Processes a directory looking for macros.

        Recurses into the sub directory, ignoring hidden directories.
        Processes every .[c|h]pp file.
        """

        for item in os.listdir(dir):

            # Ignore hidden directories.
            if item.startswith("."):
                continue

            if os.path.isdir(os.path.join(dir, item)):
                process_directory_macros(os.path.join(dir, item))
            elif item.endswith(".cpp") or item.endswith(".hpp"):
                process_file_macros(os.path.join(dir, item))

    ##### ##### ##### MAIN ##### ##### #####

    process_directory_macros(src_directory)
    process_directory(src_directory)

    create_output()