#!/usr/bin/python
# -*- coding: utf-8 -*-
#
# This script generates a data file containing all Unicode information needed by KCharSelect.
#
#############################################################################
# Copyright (C) 2007 Daniel Laidig <d.laidig@gmx.de>
#
# This script is free software; you can redistribute it and/or
# modify it under the terms of the GNU Library General Public
# License as published by the Free Software Foundation; either
# version 2 of the License, or (at your option) any later version.
#
# This script 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
# Library General Public License for more details.
#
# You should have received a copy of the GNU Library General Public License
# along with this library; see the file COPYING.LIB.  If not, write to
# the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
# Boston, MA 02110-1301, USA.
#############################################################################
#
# The current directory must contain the following files that can be found at
# http://www.unicode.org/Public/UNIDATA/:
# - UnicodeData.txt
# - Unihan_Readings.txt (you need to uncompress it from Unihan.zip)
# - NamesList.txt
# - Blocks.txt
#
# The generated file is named "kcharselect-data" and has to be put in kdelibs/kdeui/widgets/.
# Additionally a translation dummy named "kcharselect-translation.cpp" is generated and has
# to be placed in the same directory.
#
# FILE STRUCTURE
#
# The generated file is a binary file. The first 40 bytes are the header
# and contain the position of each part of the file. Each entry is uint32.
#
# pos   content
# 0     names strings begin
# 4     names offsets begin
# 8     details strings begin
# 12    details offsets begin
# 16    block strings begin
# 20    block offsets begin
# 24    section strings begin
# 28    section offsets begin
# 32    unihan strings begin
# 36    unihan offsets begin
#
# The string parts always contain all strings in a row, followed by a 0x00 byte.
# There is one exception: The data for seeAlso in details is only 2 bytes (as is always is _one_
# unicode character) and _not_ followed by a 0x00 byte.
#
# The offset parts contain entries with a fixed length. Unicode characters are always uint16 and offsets uint32.
# Offsets are positions in the data file.
#
# names_offsets:
# each entry 6 bytes
# 16bit: unicode
# 32bit: offset to name in names_strings
#
# names_strings:
# the first byte is the category (same values as QChar::Category),
# directly followed by the character name (terminated by 0x00)
#
# nameslist_offsets:
# char, alias, alias_count, note, note_count, approxEquiv, approxEquiv_coutn, equiv, equiv_count, seeAlso, seeAlso_count
# 16    32     8            32    8           32           8                  32     8            32       8
# => each entry 27 bytes
#
# blocks_offsets:
# each entry 4 bytes
# 16bit: start unicode
# 16bit: end unicode
# Note that there is no string offset.
#
# section_offsets:
# each entry 4 bytes
# 16bit: section offset
# 16bit: block offset
# Note that these offsets are _not_ positions in the data file but indexes.
# For example 0x0403 means the fourth section includes the third block.
#
# unihan_offsets:
# each entry 30 bytes
# 16bit: unicode
# 32bit: offset to unihan_strings for Definition
# 32bit: offset to unihan_strings for Cantonese
# 32bit: offset to unihan_strings for Mandarin
# 32bit: offset to unihan_strings for Tang
# 32bit: offset to unihan_strings for Korean
# 32bit: offset to unihan_strings for JapaneseKun
# 32bit: offset to unihan_strings for JapaneseOn

from struct import *
import sys
import re
import StringIO

# based on http://www.unicode.org/charts/
sectiondata = '''
SECTION European Alphabets
Basic Latin
Latin-1 Supplement
Latin Extended-A
Latin Extended-B
Latin Extended-C
Latin Extended-D
Latin Extended Additional
Armenian
Coptic
Cyrillic
Cyrillic Supplement
Cyrillic Extended-A
Cyrillic Extended-B
Georgian
Georgian Supplement
Glagolitic
Greek and Coptic
Greek Extended
Ogham
Runic

SECTION African Scripts
Bamum
Ethiopic
Ethiopic Supplement
Ethiopic Extended
Ethiopic Extended-A
NKo
Tifinagh
Vai

SECTION Middle Eastern Scripts
Arabic
Arabic Supplement
Arabic Presentation Forms-A
Arabic Presentation Forms-B
Hebrew
Mandaic
Samaritan
Syriac

SECTION South Asian Scripts
Bengali
Common Indic Number Forms
Devanagari
Devanagari Extended
Gujarati
Gurmukhi
Kannada
Lepcha
Limbu
Malayalam
Meetei Mayek
Ol Chiki
Oriya
Saurashtra
Sinhala
Syloti Nagri
Tamil
Telugu
Thaana
Vedic Extensions

SECTION Philippine Scripts
Buhid
Hanunoo
Tagalog
Tagbanwa


SECTION South East Asian Scripts
Balinese
Batak
Buginese
Cham
Javanese
Kayah Li
Khmer
Khmer Symbols
Lao
Myanmar
Myanmar Extended-A
New Tai Lue
Rejang
Sundanese
Tai Le
Tai Tham
Tai Viet
Thai

SECTION East Asian Scripts
Bopomofo
Bopomofo Extended
CJK Unified Ideographs
CJK Unified Ideographs Extension A
CJK Compatibility
CJK Compatibility Ideographs
CJK Compatibility Forms
CJK Radicals Supplement
CJK Strokes
CJK Symbols and Punctuation
Enclosed CJK Letters and Months
Hangul Jamo
Hangul Jamo Extended-A
Hangul Jamo Extended-B
Hangul Compatibility Jamo
Hangul Syllables
Hiragana
Ideographic Description Characters
Kanbun
Kangxi Radicals
Katakana
Katakana Phonetic Extensions
Lisu
Yi Radicals
Yi Syllables

SECTION Central Asian Scripts
Mongolian
Phags-pa
Tibetan

SECTION Other Scripts
Cherokee
Unified Canadian Aboriginal Syllabics
Unified Canadian Aboriginal Syllabics Extended

SECTION Symbols
Braille Patterns
Control Pictures
Currency Symbols
Dingbats
Enclosed Alphanumerics
General Punctuation
Miscellaneous Symbols
Miscellaneous Technical
Optical Character Recognition
Small Form Variants
Supplemental Punctuation
Vertical Forms
Yijing Hexagram Symbols

SECTION Mathematical Symbols
Arrows
Block Elements
Box Drawing
Supplemental Arrows-A
Supplemental Arrows-B
Geometric Shapes
Letterlike Symbols
Mathematical Operators
Supplemental Mathematical Operators
Miscellaneous Mathematical Symbols-A
Miscellaneous Mathematical Symbols-B
Miscellaneous Symbols and Arrows
Number Forms
Superscripts and Subscripts

SECTION Phonetic Symbols
IPA Extensions
Modifier Tone Letters
Phonetic Extensions
Phonetic Extensions Supplement
Spacing Modifier Letters

SECTION Combining Diacritical Marks
Combining Diacritical Marks
Combining Diacritical Marks Supplement
Combining Diacritical Marks for Symbols
Combining Half Marks

SECTION Other
Alphabetic Presentation Forms
Halfwidth and Fullwidth Forms
High Private Use Surrogates
High Surrogates
Low Surrogates
Private Use Area
Specials
Variation Selectors
'''
# TODO: rename "Other Scripts" to "American Scripts"

categoryMap = { # same values as QChar::Category
    "Mn": 1,
    "Mc": 2,
    "Me": 3,
    "Nd": 4,
    "Nl": 5,
    "No": 6,
    "Zs": 7,
    "Zl": 8,
    "Zp": 9,
    "Cc": 10,
    "Cf": 11,
    "Cs": 12,
    "Co": 13,
    "Cn": 14,
    "Lu":  15,
    "Ll":  16,
    "Lt":  17,
    "Lm":  18,
    "Lo":  19,
    "Pc":  20,
    "Pd":  21,
    "Ps":  22,
    "Pe":  23,
    "Pi":  24,
    "Pf":  25,
    "Po":  26,
    "Sm":  27,
    "Sc":  28,
    "Sk":  29,
    "So":  30
}


class Names:
    def __init__(self):
        self.names = []
        self.controlpos = -1
    def addName(self, uni, name, category):
        self.names.append([uni, name, category])

    def calculateStringSize(self):
        size = 0
        hadcontrol = False
        for entry in self.names:
            if entry[1] == "<control>":
                if not hadcontrol:
                    size += len(entry[1]) + 2
                    hadcontrol = True
            else:
                size += len(entry[1]) + 2
        return size

    def calculateOffsetSize(self):
        return len(self.names)*6

    def writeStrings(self, out, pos):
        hadcontrol = False
        for entry in self.names:
            if entry[1] == "<control>":
                if not hadcontrol:
                    out.write(pack("=b", entry[2]))
                    out.write(entry[1] + "\0")
                    size = len(entry[1]) + 2
                    entry[1] = pos
                    self.controlpos = pos
                    pos += size
                    hadcontrol = True
                else:
                    entry[1] = self.controlpos
            else:
                out.write(pack("=b", entry[2]))
                out.write(entry[1] + "\0")
                size = len(entry[1]) + 2
                entry[1] = pos
                pos += size
        return pos

    def writeOffsets(self, out, pos):
        for entry in self.names:
            out.write(pack("=HI", int(entry[0], 16), entry[1]))
            pos += 6
        return pos

class Details:
    def __init__(self):
        self.details = {}
    def addEntry(self, char, category, text):
        if not self.details.has_key(char):
            self.details[char] = {}
        if not self.details[char].has_key(category):
            self.details[char][category] = []
        self.details[char][category].append(text)

    def calculateStringSize(self):
        size = 0
        for char in self.details.values():
            for cat in char.values():
                for s in cat:
                    if type(s) is str:
                        size += len(s) + 1
                    else:
                        size += 2
        return size

    def calculateOffsetSize(self):
        return len(self.details)*27

    def writeStrings(self, out, pos):
        for char in self.details.values():
            for cat in char.values():
                for i in range(0, len(cat)):
                    s = cat[i]
                    if type(s) is str:
                        out.write(s + "\0")
                        size = len(s) + 1
                    else:
                        out.write(pack("=H", s))
                        size = 2
                    cat[i] = pos
                    pos += size
        return pos

    def writeOffsets(self, out, pos):
        for char in self.details.keys():
            alias = 0
            alias_count = 0
            note = 0
            note_count = 0
            approxEquiv = 0
            approxEquiv_count = 0
            equiv = 0
            equiv_count = 0
            seeAlso = 0
            seeAlso_count = 0
            if self.details[char].has_key("alias"):
                alias = self.details[char]["alias"][0]
                alias_count = len(self.details[char]["alias"])

            if self.details[char].has_key("note"):
                note = self.details[char]["note"][0]
                note_count = len(self.details[char]["note"])

            if self.details[char].has_key("approxEquiv"):
                approxEquiv = self.details[char]["approxEquiv"][0]
                approxEquiv_count = len(self.details[char]["approxEquiv"])

            if self.details[char].has_key("equiv"):
                equiv = self.details[char]["equiv"][0]
                equiv_count = len(self.details[char]["equiv"])

            if self.details[char].has_key("seeAlso"):
                seeAlso = self.details[char]["seeAlso"][0]
                seeAlso_count = len(self.details[char]["seeAlso"])

            out.write(pack("=HIbIbIbIbIb", char, alias, alias_count, note, note_count, approxEquiv, approxEquiv_count, equiv, equiv_count, seeAlso, seeAlso_count))
            pos += 27

        return pos

class SectionsBlocks:
    def __init__(self):
        self.sections = []
        self.blocks = []
        self.blockList = []
        self.sectionList = []

    def addBlock(self, begin, end, name):
        self.blocks.append([begin, end, name])
        self.blockList.append(name)

    def addSection(self, section, block):
        self.sections.append([section, block])
        if not section in self.sectionList:
            self.sectionList.append(section)

    def calculateBlockStringSize(self):
        size = 0
        for block in self.blocks:
            size += len(block[2]) + 1
        return size

    def calculateBlockOffsetSize(self):
        return len(self.blocks) * 4

    def calculateSectionStringSize(self):
        size = 0
        lastsection = ""
        for section in self.sections:
            if section[0] != lastsection:
                size += len(section[0]) + 1
                lastsection = section[0]
        return size

    def calculateSectionOffsetSize(self):
        return len(self.sections) * 4

    def writeBlockStrings(self, out, pos):
        index = 0
        for block in self.blocks:
            out.write(block[2] + "\0")
            size = len(block[2]) + 1
            found = False
            for section in self.sections:
                if section[1] == block[2]:
                    print "found", section
                    section[1] = index
                    found = True
            if not found:
                print "Error: Did not find any category for block \""+block[2]+"\""
                sys.exit(1)
            block[2] = index
            pos += size
            index += 1
        return pos

    def writeBlockOffsets(self, out, pos):
        for block in self.blocks:
            out.write(pack("=HH", int(block[0], 16), int(block[1], 16)))
            pos += 4
        return pos

    def writeSectionStrings(self, out, pos):
        lastsection = ""
        lastpos = 0
        index = -1
        for section in self.sections:
            if section[0] != lastsection:
                index += 1
                lastsection = section[0]
                out.write(section[0] + "\0")
                size = len(section[0]) + 1
                section[0] = index
                lastpos = pos
                pos += size
            else:
                section[0] = index
        return pos

    def writeSectionOffsets(self, out, pos):
        for section in self.sections:
            out.write(pack("=HH", section[0], section[1]))
            pos += 4
        return pos

    def getBlockList(self):
        return self.blockList
    
    def getSectionList(self):
        return self.sectionList

class Unihan:
    def __init__(self):
        self.unihan = {}

    def addUnihan(self, uni, category, value):
        uni = int(uni, 16)
        if category != "kDefinition" and category != "kCantonese" and category != "kMandarin" and category != "kTang" and category != "kKorean" and category != "kJapaneseKun" and category != "kJapaneseOn":
            return
        if not self.unihan.has_key(uni):
            self.unihan[uni] = [None, None, None, None, None, None, None]
        if category == "kDefinition":
            self.unihan[uni][0] = value
        elif category == "kCantonese":
            self.unihan[uni][1] = value
        elif category == "kMandarin":
            self.unihan[uni][2] = value
        elif category == "kTang":
            self.unihan[uni][3] = value
        elif category == "kKorean":
            self.unihan[uni][4] = value
        elif category == "kJapaneseKun":
            self.unihan[uni][5] = value
        elif category == "kJapaneseOn":
            self.unihan[uni][6] = value

    def calculateStringSize(self):
        size = 0
        for char in self.unihan.keys():
            for entry in self.unihan[char]:
                if entry != None:
                    size += len(entry) + 1
        return size

    def calculateOffsetSize(self):
        return len(self.unihan) * 30

    def writeStrings(self, out, pos):
        for char in self.unihan.keys():
            for i in range(0, 7):
                if self.unihan[char][i] != None:
                    out.write(self.unihan[char][i] + "\0")
                    size = len(self.unihan[char][i]) + 1
                    self.unihan[char][i] = pos
                    pos += size
        return pos

    def writeOffsets(self, out, pos):
        for char in self.unihan.keys():
            out.write(pack("=H", char))
            for i in range(0, 7):
                if self.unihan[char][i] != None:
                    out.write(pack("=I", self.unihan[char][i]))
                else:
                    out.write(pack("=I", 0))
            pos += 30
        return pos

class Parser:
    def parseUnicodeData(self, inUnicodeData, names):
        regexp = re.compile(r'^([^;]+);([^;]+);([^;]+)')
        for line in inUnicodeData:
            line = line[:-1]
            m = regexp.match(line)
            if not m:
                continue
            uni = m.group(1)
            name = m.group(2)
            category = m.group(3)
            if len(uni) > 4:
                continue
            names.addName(uni, name, categoryMap[category])

    def parseDetails(self, inNamesList, details):
        invalidRegexp = re.compile(r'^@')
        unicodeRegexp = re.compile(r'^([0-9A-F]+)')

        aliasRegexp = re.compile(r'^\s+=\s+(.+)$') #equal
        seeAlsoRegexp = re.compile(r'^\s+x\s+.*([0-9A-F]{4,6})\)$') #ex
        noteRegexp = re.compile(r'^\s+\*\s+(.+)$') #star
        approxEquivalentRegexp = re.compile(r'^\s+#\s+(.+)$') #pound
        equivalentRegexp = re.compile(r'^\s+:\s+(.+)$') #colon

        drop = 0
        currChar = 0

        for line in inNamesList:
            line = line[:-1]
            m1 = unicodeRegexp.match(line)
            m2 = aliasRegexp.match(line)
            m3 = noteRegexp.match(line)
            m4 = approxEquivalentRegexp.match(line)
            m5 = equivalentRegexp.match(line)
            m6 = seeAlsoRegexp.match(line)
            if invalidRegexp.match(line):
                continue
            elif m1:
                currChar = int(m1.group(1), 16)
                if len(m1.group(1)) > 4: #limit to 16bit
                    drop = 1
                    continue
            elif drop == 1:
                continue
            elif m2:
                value = m2.group(1)
                details.addEntry(currChar, "alias", value)
            elif m3:
                value = m3.group(1)
                details.addEntry(currChar, "note", value)
            elif m4:
                value = m4.group(1)
                details.addEntry(currChar, "approxEquiv", value)
            elif m5:
                value = m5.group(1)
                details.addEntry(currChar, "equiv", value)
            elif m6:
                value = int(m6.group(1), 16)
                details.addEntry(currChar, "seeAlso", value)
    def parseBlocks(self, inBlocks, sectionsBlocks):
        regexp = re.compile(r'^([0-9A-F]+)\.\.([0-9A-F]+); (.+)$')
        for line in inBlocks:
            line = line[:-1]
            m = regexp.match(line)
            if not m:
                continue
            if len(m.group(1)) > 4:
                continue
            sectionsBlocks.addBlock(m.group(1), m.group(2), m.group(3))
    def parseSections(self, inSections, sectionsBlocks):
        currSection = ""
        for line in inSections:
            line = line[:-1]
            if len(line) == 0:
                continue;
            temp = line.split(" ")
            if temp[0] == "SECTION":
                currSection = line[8:]
            elif currSection != "":
                sectionsBlocks.addSection(currSection, line)
            else:
                print "error in data file"
                sys.exit(1)
    def parseUnihan(self, inUnihan, unihan):
        regexp = re.compile(r'^U\+([0-9A-F]+)\s+([^\s]+)\s+(.+)$')
        count = 0
        for line in inUnihan:
            if count % 100000 == 0:
                print "\b.",; sys.stdout.flush()
            count += 1
            line = line[:-1]
            m = regexp.match(line)
            if not m:
                continue
            if len(m.group(1)) <= 4:
                unihan.addUnihan(m.group(1), m.group(2), m.group(3))

def writeTranslationDummy(out, data):
    out.write("""/* This file is part of the KDE libraries

   Copyright (C) 2007 Daniel Laidig <d.laidig@gmx.de>

   This library is free software; you can redistribute it and/or
   modify it under the terms of the GNU Library General Public
   License as published by the Free Software Foundation; either
   version 2 of the License, or (at your option) any later version.

   This library 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
   Library General Public License for more details.

   You should have received a copy of the GNU Library General Public License
   along with this library; see the file COPYING.LIB.  If not, write to
   the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
   Boston, MA 02110-1301, USA.

   This file is autogenerated by kdeutils/kcharselect/kcharselect-generate-datafile.py
*/\n\n""")
    for group in data:
        for entry in group[1]:
            out.write("I18N_NOOP2(\""+group[0]+"\", \""+entry+"\");\n")

out = open("kcharselect-data", "wb")
outTranslationDummy = open("kcharselect-translation.cpp", "wb")

inUnicodeData = open("UnicodeData.txt", "r")
inNamesList = open("NamesList.txt", "r")
inBlocks = open("Blocks.txt", "r")
inSections = StringIO.StringIO(sectiondata)
inUnihan = open("Unihan_Readings.txt", "r")

if calcsize('=H') != 2 or calcsize('=I') != 4:
    print "Error: Sizes of ushort and uint are not 16 and 32 bit as expected"
    sys.exit(1)

names = Names()
details = Details()
sectionsBlocks = SectionsBlocks()
unihan = Unihan()

parser = Parser()

print "========== parsing files ==================="
parser.parseUnicodeData(inUnicodeData, names)
print ".",; sys.stdout.flush()
parser.parseDetails(inNamesList, details)
print "\b.",; sys.stdout.flush()
parser.parseBlocks(inBlocks, sectionsBlocks)
print "\b.",; sys.stdout.flush()
parser.parseSections(inSections, sectionsBlocks)
print "\b.",; sys.stdout.flush()
parser.parseUnihan(inUnihan, unihan)
print "\b.",; sys.stdout.flush()

print "done."

pos = 0

#write header, size: 40 bytes
print "========== writing header =================="
out.write(pack("=I", 40))
print "names strings begin", 40

namesOffsetBegin = names.calculateStringSize() + 40
out.write(pack("=I", namesOffsetBegin))
print "names offsets begin", namesOffsetBegin

detailsStringBegin = namesOffsetBegin + names.calculateOffsetSize()
out.write(pack("=I", detailsStringBegin))
print "details strings begin", detailsStringBegin

detailsOffsetBegin = detailsStringBegin + details.calculateStringSize()
out.write(pack("=I", detailsOffsetBegin))
print "details offsets begin", detailsOffsetBegin

blocksStringBegin = detailsOffsetBegin + details.calculateOffsetSize()
out.write(pack("=I", blocksStringBegin))
print "block strings begin", blocksStringBegin

blocksOffsetBegin = blocksStringBegin + sectionsBlocks.calculateBlockStringSize()
out.write(pack("=I", blocksOffsetBegin))
print "block offsets begin", blocksOffsetBegin

sectionStringBegin = blocksOffsetBegin + sectionsBlocks.calculateBlockOffsetSize()
out.write(pack("=I", sectionStringBegin))
print "section strings begin", sectionStringBegin

sectionOffsetBegin = sectionStringBegin + sectionsBlocks.calculateSectionStringSize()
out.write(pack("=I", sectionOffsetBegin))
print "section offsets begin", sectionOffsetBegin

unihanStringBegin = sectionOffsetBegin + sectionsBlocks.calculateSectionOffsetSize()
out.write(pack("=I", unihanStringBegin))
print "unihan strings begin", unihanStringBegin

unihanOffsetBegin = unihanStringBegin + unihan.calculateStringSize()
out.write(pack("=I", unihanOffsetBegin))
print "unihan offsets begin", unihanOffsetBegin

end = unihanOffsetBegin + unihan.calculateOffsetSize()
print "end should be", end

pos += 40

print "========== writing data ===================="

pos = names.writeStrings(out, pos)
print "names strings written, position", pos
pos = names.writeOffsets(out, pos)
print "names offsets written, position", pos
pos = details.writeStrings(out, pos)
print "details strings written, position", pos
pos = details.writeOffsets(out, pos)
print "details offsets written, position", pos
pos = sectionsBlocks.writeBlockStrings(out, pos)
print "block strings written, position", pos
pos = sectionsBlocks.writeBlockOffsets(out, pos)
print "block offsets written, position", pos
pos = sectionsBlocks.writeSectionStrings(out, pos)
print "section strings written, position", pos
pos = sectionsBlocks.writeSectionOffsets(out, pos)
print "section offsets written, position", pos
pos = unihan.writeStrings(out, pos)
print "unihan strings written, position", pos
pos = unihan.writeOffsets(out, pos)
print "unihan offsets written, position", pos

print "========== writing translation dummy  ======"
translationData = [["KCharSelect section name", sectionsBlocks.getSectionList()], ["KCharselect unicode block name",sectionsBlocks.getBlockList()]]
writeTranslationDummy(outTranslationDummy, translationData)
print "done. make sure to copy both kcharselect-data and kcharselect-translation.cpp."