tools v3.0
First combined mobi/topaz kindle tool
This commit is contained in:
Apprentice Alf
817
Calibre_Plugins/K4MobiDeDRM_plugin/convert2xml.py
Normal file
817
Calibre_Plugins/K4MobiDeDRM_plugin/convert2xml.py
Normal file
@@ -0,0 +1,817 @@
|
||||
#! /usr/bin/python
|
||||
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
|
||||
# For use with Topaz Scripts Version 2.6
|
||||
|
||||
class Unbuffered:
|
||||
def __init__(self, stream):
|
||||
self.stream = stream
|
||||
def write(self, data):
|
||||
self.stream.write(data)
|
||||
self.stream.flush()
|
||||
def __getattr__(self, attr):
|
||||
return getattr(self.stream, attr)
|
||||
|
||||
import sys
|
||||
sys.stdout=Unbuffered(sys.stdout)
|
||||
|
||||
import csv
|
||||
import os
|
||||
import getopt
|
||||
from struct import pack
|
||||
from struct import unpack
|
||||
|
||||
|
||||
# Get a 7 bit encoded number from string. The most
|
||||
# significant byte comes first and has the high bit (8th) set
|
||||
|
||||
def readEncodedNumber(file):
|
||||
flag = False
|
||||
c = file.read(1)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
data = ord(c)
|
||||
|
||||
if data == 0xFF:
|
||||
flag = True
|
||||
c = file.read(1)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
data = ord(c)
|
||||
|
||||
if data >= 0x80:
|
||||
datax = (data & 0x7F)
|
||||
while data >= 0x80 :
|
||||
c = file.read(1)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
data = ord(c)
|
||||
datax = (datax <<7) + (data & 0x7F)
|
||||
data = datax
|
||||
|
||||
if flag:
|
||||
data = -data
|
||||
return data
|
||||
|
||||
|
||||
# returns a binary string that encodes a number into 7 bits
|
||||
# most significant byte first which has the high bit set
|
||||
|
||||
def encodeNumber(number):
|
||||
result = ""
|
||||
negative = False
|
||||
flag = 0
|
||||
|
||||
if number < 0 :
|
||||
number = -number + 1
|
||||
negative = True
|
||||
|
||||
while True:
|
||||
byte = number & 0x7F
|
||||
number = number >> 7
|
||||
byte += flag
|
||||
result += chr(byte)
|
||||
flag = 0x80
|
||||
if number == 0 :
|
||||
if (byte == 0xFF and negative == False) :
|
||||
result += chr(0x80)
|
||||
break
|
||||
|
||||
if negative:
|
||||
result += chr(0xFF)
|
||||
|
||||
return result[::-1]
|
||||
|
||||
|
||||
|
||||
# create / read a length prefixed string from the file
|
||||
|
||||
def lengthPrefixString(data):
|
||||
return encodeNumber(len(data))+data
|
||||
|
||||
def readString(file):
|
||||
stringLength = readEncodedNumber(file)
|
||||
if (stringLength == None):
|
||||
return ""
|
||||
sv = file.read(stringLength)
|
||||
if (len(sv) != stringLength):
|
||||
return ""
|
||||
return unpack(str(stringLength)+"s",sv)[0]
|
||||
|
||||
|
||||
# convert a binary string generated by encodeNumber (7 bit encoded number)
|
||||
# to the value you would find inside the page*.dat files to be processed
|
||||
|
||||
def convert(i):
|
||||
result = ''
|
||||
val = encodeNumber(i)
|
||||
for j in xrange(len(val)):
|
||||
c = ord(val[j:j+1])
|
||||
result += '%02x' % c
|
||||
return result
|
||||
|
||||
|
||||
|
||||
# the complete string table used to store all book text content
|
||||
# as well as the xml tokens and values that make sense out of it
|
||||
|
||||
class Dictionary(object):
|
||||
def __init__(self, dictFile):
|
||||
self.filename = dictFile
|
||||
self.size = 0
|
||||
self.fo = file(dictFile,'rb')
|
||||
self.stable = []
|
||||
self.size = readEncodedNumber(self.fo)
|
||||
for i in xrange(self.size):
|
||||
self.stable.append(self.escapestr(readString(self.fo)))
|
||||
self.pos = 0
|
||||
|
||||
def escapestr(self, str):
|
||||
str = str.replace('&','&')
|
||||
str = str.replace('<','<')
|
||||
str = str.replace('>','>')
|
||||
str = str.replace('=','=')
|
||||
return str
|
||||
|
||||
def lookup(self,val):
|
||||
if ((val >= 0) and (val < self.size)) :
|
||||
self.pos = val
|
||||
return self.stable[self.pos]
|
||||
else:
|
||||
print "Error - %d outside of string table limits" % val
|
||||
sys.exit(-1)
|
||||
|
||||
def getSize(self):
|
||||
return self.size
|
||||
|
||||
def getPos(self):
|
||||
return self.pos
|
||||
|
||||
def dumpDict(self):
|
||||
for i in xrange(self.size):
|
||||
print "%d %s %s" % (i, convert(i), self.stable[i])
|
||||
return
|
||||
|
||||
# parses the xml snippets that are represented by each page*.dat file.
|
||||
# also parses the other0.dat file - the main stylesheet
|
||||
# and information used to inject the xml snippets into page*.dat files
|
||||
|
||||
class PageParser(object):
|
||||
def __init__(self, filename, dict, debug, flat_xml):
|
||||
self.fo = file(filename,'rb')
|
||||
self.id = os.path.basename(filename).replace('.dat','')
|
||||
self.dict = dict
|
||||
self.debug = debug
|
||||
self.flat_xml = flat_xml
|
||||
self.tagpath = []
|
||||
self.doc = []
|
||||
self.snippetList = []
|
||||
|
||||
|
||||
# hash table used to enable the decoding process
|
||||
# This has all been developed by trial and error so it may still have omissions or
|
||||
# contain errors
|
||||
# Format:
|
||||
# tag : (number of arguments, argument type, subtags present, special case of subtags presents when escaped)
|
||||
|
||||
token_tags = {
|
||||
'x' : (1, 'scalar_number', 0, 0),
|
||||
'y' : (1, 'scalar_number', 0, 0),
|
||||
'h' : (1, 'scalar_number', 0, 0),
|
||||
'w' : (1, 'scalar_number', 0, 0),
|
||||
'firstWord' : (1, 'scalar_number', 0, 0),
|
||||
'lastWord' : (1, 'scalar_number', 0, 0),
|
||||
'rootID' : (1, 'scalar_number', 0, 0),
|
||||
'stemID' : (1, 'scalar_number', 0, 0),
|
||||
'type' : (1, 'scalar_text', 0, 0),
|
||||
|
||||
'info' : (0, 'number', 1, 0),
|
||||
|
||||
'info.word' : (0, 'number', 1, 1),
|
||||
'info.word.ocrText' : (1, 'text', 0, 0),
|
||||
'info.word.firstGlyph' : (1, 'raw', 0, 0),
|
||||
'info.word.lastGlyph' : (1, 'raw', 0, 0),
|
||||
'info.word.bl' : (1, 'raw', 0, 0),
|
||||
'info.word.link_id' : (1, 'number', 0, 0),
|
||||
|
||||
'glyph' : (0, 'number', 1, 1),
|
||||
'glyph.x' : (1, 'number', 0, 0),
|
||||
'glyph.y' : (1, 'number', 0, 0),
|
||||
'glyph.glyphID' : (1, 'number', 0, 0),
|
||||
|
||||
'dehyphen' : (0, 'number', 1, 1),
|
||||
'dehyphen.rootID' : (1, 'number', 0, 0),
|
||||
'dehyphen.stemID' : (1, 'number', 0, 0),
|
||||
'dehyphen.stemPage' : (1, 'number', 0, 0),
|
||||
'dehyphen.sh' : (1, 'number', 0, 0),
|
||||
|
||||
'links' : (0, 'number', 1, 1),
|
||||
'links.page' : (1, 'number', 0, 0),
|
||||
'links.rel' : (1, 'number', 0, 0),
|
||||
'links.row' : (1, 'number', 0, 0),
|
||||
'links.title' : (1, 'text', 0, 0),
|
||||
'links.href' : (1, 'text', 0, 0),
|
||||
'links.type' : (1, 'text', 0, 0),
|
||||
|
||||
'paraCont' : (0, 'number', 1, 1),
|
||||
'paraCont.rootID' : (1, 'number', 0, 0),
|
||||
'paraCont.stemID' : (1, 'number', 0, 0),
|
||||
'paraCont.stemPage' : (1, 'number', 0, 0),
|
||||
|
||||
'paraStems' : (0, 'number', 1, 1),
|
||||
'paraStems.stemID' : (1, 'number', 0, 0),
|
||||
|
||||
'wordStems' : (0, 'number', 1, 1),
|
||||
'wordStems.stemID' : (1, 'number', 0, 0),
|
||||
|
||||
'empty' : (1, 'snippets', 1, 0),
|
||||
|
||||
'page' : (1, 'snippets', 1, 0),
|
||||
'page.pageid' : (1, 'scalar_text', 0, 0),
|
||||
'page.pagelabel' : (1, 'scalar_text', 0, 0),
|
||||
'page.type' : (1, 'scalar_text', 0, 0),
|
||||
'page.h' : (1, 'scalar_number', 0, 0),
|
||||
'page.w' : (1, 'scalar_number', 0, 0),
|
||||
'page.startID' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'group' : (1, 'snippets', 1, 0),
|
||||
'group.type' : (1, 'scalar_text', 0, 0),
|
||||
|
||||
'region' : (1, 'snippets', 1, 0),
|
||||
'region.type' : (1, 'scalar_text', 0, 0),
|
||||
'region.x' : (1, 'scalar_number', 0, 0),
|
||||
'region.y' : (1, 'scalar_number', 0, 0),
|
||||
'region.h' : (1, 'scalar_number', 0, 0),
|
||||
'region.w' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'empty_text_region' : (1, 'snippets', 1, 0),
|
||||
|
||||
'img' : (1, 'snippets', 1, 0),
|
||||
'img.x' : (1, 'scalar_number', 0, 0),
|
||||
'img.y' : (1, 'scalar_number', 0, 0),
|
||||
'img.h' : (1, 'scalar_number', 0, 0),
|
||||
'img.w' : (1, 'scalar_number', 0, 0),
|
||||
'img.src' : (1, 'scalar_number', 0, 0),
|
||||
'img.color_src' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'paragraph' : (1, 'snippets', 1, 0),
|
||||
'paragraph.class' : (1, 'scalar_text', 0, 0),
|
||||
'paragraph.firstWord' : (1, 'scalar_number', 0, 0),
|
||||
'paragraph.lastWord' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'word_semantic' : (1, 'snippets', 1, 1),
|
||||
'word_semantic.type' : (1, 'scalar_text', 0, 0),
|
||||
'word_semantic.firstWord' : (1, 'scalar_number', 0, 0),
|
||||
'word_semantic.lastWord' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'word' : (1, 'snippets', 1, 0),
|
||||
'word.type' : (1, 'scalar_text', 0, 0),
|
||||
'word.class' : (1, 'scalar_text', 0, 0),
|
||||
'word.firstGlyph' : (1, 'scalar_number', 0, 0),
|
||||
'word.lastGlyph' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'_span' : (1, 'snippets', 1, 0),
|
||||
'_span.firstWord' : (1, 'scalar_number', 0, 0),
|
||||
'-span.lastWord' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'span' : (1, 'snippets', 1, 0),
|
||||
'span.firstWord' : (1, 'scalar_number', 0, 0),
|
||||
'span.lastWord' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'extratokens' : (1, 'snippets', 1, 0),
|
||||
'extratokens.type' : (1, 'scalar_text', 0, 0),
|
||||
'extratokens.firstGlyph' : (1, 'scalar_number', 0, 0),
|
||||
'extratokens.lastGlyph' : (1, 'scalar_number', 0, 0),
|
||||
|
||||
'glyph.h' : (1, 'number', 0, 0),
|
||||
'glyph.w' : (1, 'number', 0, 0),
|
||||
'glyph.use' : (1, 'number', 0, 0),
|
||||
'glyph.vtx' : (1, 'number', 0, 1),
|
||||
'glyph.len' : (1, 'number', 0, 1),
|
||||
'glyph.dpi' : (1, 'number', 0, 0),
|
||||
'vtx' : (0, 'number', 1, 1),
|
||||
'vtx.x' : (1, 'number', 0, 0),
|
||||
'vtx.y' : (1, 'number', 0, 0),
|
||||
'len' : (0, 'number', 1, 1),
|
||||
'len.n' : (1, 'number', 0, 0),
|
||||
|
||||
'book' : (1, 'snippets', 1, 0),
|
||||
'version' : (1, 'snippets', 1, 0),
|
||||
'version.FlowEdit_1_id' : (1, 'scalar_text', 0, 0),
|
||||
'version.FlowEdit_1_version' : (1, 'scalar_text', 0, 0),
|
||||
'version.Schema_id' : (1, 'scalar_text', 0, 0),
|
||||
'version.Schema_version' : (1, 'scalar_text', 0, 0),
|
||||
'version.Topaz_version' : (1, 'scalar_text', 0, 0),
|
||||
'version.WordDetailEdit_1_id' : (1, 'scalar_text', 0, 0),
|
||||
'version.WordDetailEdit_1_version' : (1, 'scalar_text', 0, 0),
|
||||
'version.ZoneEdit_1_id' : (1, 'scalar_text', 0, 0),
|
||||
'version.ZoneEdit_1_version' : (1, 'scalar_text', 0, 0),
|
||||
'version.chapterheaders' : (1, 'scalar_text', 0, 0),
|
||||
'version.creation_date' : (1, 'scalar_text', 0, 0),
|
||||
'version.header_footer' : (1, 'scalar_text', 0, 0),
|
||||
'version.init_from_ocr' : (1, 'scalar_text', 0, 0),
|
||||
'version.letter_insertion' : (1, 'scalar_text', 0, 0),
|
||||
'version.xmlinj_convert' : (1, 'scalar_text', 0, 0),
|
||||
'version.xmlinj_reflow' : (1, 'scalar_text', 0, 0),
|
||||
'version.xmlinj_transform' : (1, 'scalar_text', 0, 0),
|
||||
'version.findlists' : (1, 'scalar_text', 0, 0),
|
||||
'version.page_num' : (1, 'scalar_text', 0, 0),
|
||||
'version.page_type' : (1, 'scalar_text', 0, 0),
|
||||
'version.bad_text' : (1, 'scalar_text', 0, 0),
|
||||
'version.glyph_mismatch' : (1, 'scalar_text', 0, 0),
|
||||
'version.margins' : (1, 'scalar_text', 0, 0),
|
||||
'version.staggered_lines' : (1, 'scalar_text', 0, 0),
|
||||
'version.paragraph_continuation' : (1, 'scalar_text', 0, 0),
|
||||
'version.toc' : (1, 'scalar_text', 0, 0),
|
||||
|
||||
'stylesheet' : (1, 'snippets', 1, 0),
|
||||
'style' : (1, 'snippets', 1, 0),
|
||||
'style._tag' : (1, 'scalar_text', 0, 0),
|
||||
'style.type' : (1, 'scalar_text', 0, 0),
|
||||
'style._parent_type' : (1, 'scalar_text', 0, 0),
|
||||
'style.class' : (1, 'scalar_text', 0, 0),
|
||||
'style._after_class' : (1, 'scalar_text', 0, 0),
|
||||
'rule' : (1, 'snippets', 1, 0),
|
||||
'rule.attr' : (1, 'scalar_text', 0, 0),
|
||||
'rule.value' : (1, 'scalar_text', 0, 0),
|
||||
|
||||
'original' : (0, 'number', 1, 1),
|
||||
'original.pnum' : (1, 'number', 0, 0),
|
||||
'original.pid' : (1, 'text', 0, 0),
|
||||
'pages' : (0, 'number', 1, 1),
|
||||
'pages.ref' : (1, 'number', 0, 0),
|
||||
'pages.id' : (1, 'number', 0, 0),
|
||||
'startID' : (0, 'number', 1, 1),
|
||||
'startID.page' : (1, 'number', 0, 0),
|
||||
'startID.id' : (1, 'number', 0, 0),
|
||||
|
||||
}
|
||||
|
||||
|
||||
# full tag path record keeping routines
|
||||
def tag_push(self, token):
|
||||
self.tagpath.append(token)
|
||||
def tag_pop(self):
|
||||
if len(self.tagpath) > 0 :
|
||||
self.tagpath.pop()
|
||||
def tagpath_len(self):
|
||||
return len(self.tagpath)
|
||||
def get_tagpath(self, i):
|
||||
cnt = len(self.tagpath)
|
||||
if i < cnt : result = self.tagpath[i]
|
||||
for j in xrange(i+1, cnt) :
|
||||
result += '.' + self.tagpath[j]
|
||||
return result
|
||||
|
||||
|
||||
# list of absolute command byte values values that indicate
|
||||
# various types of loop meachanisms typically used to generate vectors
|
||||
|
||||
cmd_list = (0x76, 0x76)
|
||||
|
||||
# peek at and return 1 byte that is ahead by i bytes
|
||||
def peek(self, aheadi):
|
||||
c = self.fo.read(aheadi)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
self.fo.seek(-aheadi,1)
|
||||
c = c[-1:]
|
||||
return ord(c)
|
||||
|
||||
|
||||
# get the next value from the file being processed
|
||||
def getNext(self):
|
||||
nbyte = self.peek(1);
|
||||
if (nbyte == None):
|
||||
return None
|
||||
val = readEncodedNumber(self.fo)
|
||||
return val
|
||||
|
||||
|
||||
# format an arg by argtype
|
||||
def formatArg(self, arg, argtype):
|
||||
if (argtype == 'text') or (argtype == 'scalar_text') :
|
||||
result = self.dict.lookup(arg)
|
||||
elif (argtype == 'raw') or (argtype == 'number') or (argtype == 'scalar_number') :
|
||||
result = arg
|
||||
elif (argtype == 'snippets') :
|
||||
result = arg
|
||||
else :
|
||||
print "Error Unknown argtype %s" % argtype
|
||||
sys.exit(-2)
|
||||
return result
|
||||
|
||||
|
||||
# process the next tag token, recursively handling subtags,
|
||||
# arguments, and commands
|
||||
def procToken(self, token):
|
||||
|
||||
known_token = False
|
||||
self.tag_push(token)
|
||||
|
||||
if self.debug : print 'Processing: ', self.get_tagpath(0)
|
||||
cnt = self.tagpath_len()
|
||||
for j in xrange(cnt):
|
||||
tkn = self.get_tagpath(j)
|
||||
if tkn in self.token_tags :
|
||||
num_args = self.token_tags[tkn][0]
|
||||
argtype = self.token_tags[tkn][1]
|
||||
subtags = self.token_tags[tkn][2]
|
||||
splcase = self.token_tags[tkn][3]
|
||||
ntags = -1
|
||||
known_token = True
|
||||
break
|
||||
|
||||
if known_token :
|
||||
|
||||
# handle subtags if present
|
||||
subtagres = []
|
||||
if (splcase == 1):
|
||||
# this type of tag uses of escape marker 0x74 indicate subtag count
|
||||
if self.peek(1) == 0x74:
|
||||
skip = readEncodedNumber(self.fo)
|
||||
subtags = 1
|
||||
num_args = 0
|
||||
|
||||
if (subtags == 1):
|
||||
ntags = readEncodedNumber(self.fo)
|
||||
if self.debug : print 'subtags: ' + token + ' has ' + str(ntags)
|
||||
for j in xrange(ntags):
|
||||
val = readEncodedNumber(self.fo)
|
||||
subtagres.append(self.procToken(self.dict.lookup(val)))
|
||||
|
||||
# arguments can be scalars or vectors of text or numbers
|
||||
argres = []
|
||||
if num_args > 0 :
|
||||
firstarg = self.peek(1)
|
||||
if (firstarg in self.cmd_list) and (argtype != 'scalar_number') and (argtype != 'scalar_text'):
|
||||
# single argument is a variable length vector of data
|
||||
arg = readEncodedNumber(self.fo)
|
||||
argres = self.decodeCMD(arg,argtype)
|
||||
else :
|
||||
# num_arg scalar arguments
|
||||
for i in xrange(num_args):
|
||||
argres.append(self.formatArg(readEncodedNumber(self.fo), argtype))
|
||||
|
||||
# build the return tag
|
||||
result = []
|
||||
tkn = self.get_tagpath(0)
|
||||
result.append(tkn)
|
||||
result.append(subtagres)
|
||||
result.append(argtype)
|
||||
result.append(argres)
|
||||
self.tag_pop()
|
||||
return result
|
||||
|
||||
# all tokens that need to be processed should be in the hash
|
||||
# table if it may indicate a problem, either new token
|
||||
# or an out of sync condition
|
||||
else:
|
||||
result = []
|
||||
if (self.debug):
|
||||
print 'Unknown Token:', token
|
||||
self.tag_pop()
|
||||
return result
|
||||
|
||||
|
||||
# special loop used to process code snippets
|
||||
# it is NEVER used to format arguments.
|
||||
# builds the snippetList
|
||||
def doLoop72(self, argtype):
|
||||
cnt = readEncodedNumber(self.fo)
|
||||
if self.debug :
|
||||
result = 'Set of '+ str(cnt) + ' xml snippets. The overall structure \n'
|
||||
result += 'of the document is indicated by snippet number sets at the\n'
|
||||
result += 'end of each snippet. \n'
|
||||
print result
|
||||
for i in xrange(cnt):
|
||||
if self.debug: print 'Snippet:',str(i)
|
||||
snippet = []
|
||||
snippet.append(i)
|
||||
val = readEncodedNumber(self.fo)
|
||||
snippet.append(self.procToken(self.dict.lookup(val)))
|
||||
self.snippetList.append(snippet)
|
||||
return
|
||||
|
||||
|
||||
|
||||
# general loop code gracisouly submitted by "skindle" - thank you!
|
||||
def doLoop76Mode(self, argtype, cnt, mode):
|
||||
result = []
|
||||
adj = 0
|
||||
if mode & 1:
|
||||
adj = readEncodedNumber(self.fo)
|
||||
mode = mode >> 1
|
||||
x = []
|
||||
for i in xrange(cnt):
|
||||
x.append(readEncodedNumber(self.fo) - adj)
|
||||
for i in xrange(mode):
|
||||
for j in xrange(1, cnt):
|
||||
x[j] = x[j] + x[j - 1]
|
||||
for i in xrange(cnt):
|
||||
result.append(self.formatArg(x[i],argtype))
|
||||
return result
|
||||
|
||||
|
||||
# dispatches loop commands bytes with various modes
|
||||
# The 0x76 style loops are used to build vectors
|
||||
|
||||
# This was all derived by trial and error and
|
||||
# new loop types may exist that are not handled here
|
||||
# since they did not appear in the test cases
|
||||
|
||||
def decodeCMD(self, cmd, argtype):
|
||||
if (cmd == 0x76):
|
||||
|
||||
# loop with cnt, and mode to control loop styles
|
||||
cnt = readEncodedNumber(self.fo)
|
||||
mode = readEncodedNumber(self.fo)
|
||||
|
||||
if self.debug : print 'Loop for', cnt, 'with mode', mode, ': '
|
||||
return self.doLoop76Mode(argtype, cnt, mode)
|
||||
|
||||
if self.dbug: print "Unknown command", cmd
|
||||
result = []
|
||||
return result
|
||||
|
||||
|
||||
|
||||
# add full tag path to injected snippets
|
||||
def updateName(self, tag, prefix):
|
||||
name = tag[0]
|
||||
subtagList = tag[1]
|
||||
argtype = tag[2]
|
||||
argList = tag[3]
|
||||
nname = prefix + '.' + name
|
||||
nsubtaglist = []
|
||||
for j in subtagList:
|
||||
nsubtaglist.append(self.updateName(j,prefix))
|
||||
ntag = []
|
||||
ntag.append(nname)
|
||||
ntag.append(nsubtaglist)
|
||||
ntag.append(argtype)
|
||||
ntag.append(argList)
|
||||
return ntag
|
||||
|
||||
|
||||
|
||||
# perform depth first injection of specified snippets into this one
|
||||
def injectSnippets(self, snippet):
|
||||
snipno, tag = snippet
|
||||
name = tag[0]
|
||||
subtagList = tag[1]
|
||||
argtype = tag[2]
|
||||
argList = tag[3]
|
||||
nsubtagList = []
|
||||
if len(argList) > 0 :
|
||||
for j in argList:
|
||||
asnip = self.snippetList[j]
|
||||
aso, atag = self.injectSnippets(asnip)
|
||||
atag = self.updateName(atag, name)
|
||||
nsubtagList.append(atag)
|
||||
argtype='number'
|
||||
argList=[]
|
||||
if len(nsubtagList) > 0 :
|
||||
subtagList.extend(nsubtagList)
|
||||
tag = []
|
||||
tag.append(name)
|
||||
tag.append(subtagList)
|
||||
tag.append(argtype)
|
||||
tag.append(argList)
|
||||
snippet = []
|
||||
snippet.append(snipno)
|
||||
snippet.append(tag)
|
||||
return snippet
|
||||
|
||||
|
||||
|
||||
# format the tag for output
|
||||
def formatTag(self, node):
|
||||
name = node[0]
|
||||
subtagList = node[1]
|
||||
argtype = node[2]
|
||||
argList = node[3]
|
||||
fullpathname = name.split('.')
|
||||
nodename = fullpathname.pop()
|
||||
ilvl = len(fullpathname)
|
||||
indent = ' ' * (3 * ilvl)
|
||||
result = indent + '<' + nodename + '>'
|
||||
if len(argList) > 0:
|
||||
argres = ''
|
||||
for j in argList:
|
||||
if (argtype == 'text') or (argtype == 'scalar_text') :
|
||||
argres += j + '|'
|
||||
else :
|
||||
argres += str(j) + ','
|
||||
argres = argres[0:-1]
|
||||
if argtype == 'snippets' :
|
||||
result += 'snippets:' + argres
|
||||
else :
|
||||
result += argres
|
||||
if len(subtagList) > 0 :
|
||||
result += '\n'
|
||||
for j in subtagList:
|
||||
if len(j) > 0 :
|
||||
result += self.formatTag(j)
|
||||
result += indent + '</' + nodename + '>\n'
|
||||
else:
|
||||
result += '</' + nodename + '>\n'
|
||||
return result
|
||||
|
||||
|
||||
# flatten tag
|
||||
def flattenTag(self, node):
|
||||
name = node[0]
|
||||
subtagList = node[1]
|
||||
argtype = node[2]
|
||||
argList = node[3]
|
||||
result = name
|
||||
if (len(argList) > 0):
|
||||
argres = ''
|
||||
for j in argList:
|
||||
if (argtype == 'text') or (argtype == 'scalar_text') :
|
||||
argres += j + '|'
|
||||
else :
|
||||
argres += str(j) + '|'
|
||||
argres = argres[0:-1]
|
||||
if argtype == 'snippets' :
|
||||
result += '.snippets=' + argres
|
||||
else :
|
||||
result += '=' + argres
|
||||
result += '\n'
|
||||
for j in subtagList:
|
||||
if len(j) > 0 :
|
||||
result += self.flattenTag(j)
|
||||
return result
|
||||
|
||||
|
||||
# reduce create xml output
|
||||
def formatDoc(self, flat_xml):
|
||||
result = ''
|
||||
for j in self.doc :
|
||||
if len(j) > 0:
|
||||
if flat_xml:
|
||||
result += self.flattenTag(j)
|
||||
else:
|
||||
result += self.formatTag(j)
|
||||
if self.debug : print result
|
||||
return result
|
||||
|
||||
|
||||
|
||||
# main loop - parse the page.dat files
|
||||
# to create structured document and snippets
|
||||
|
||||
# FIXME: value at end of magic appears to be a subtags count
|
||||
# but for what? For now, inject an 'info" tag as it is in
|
||||
# every dictionary and seems close to what is meant
|
||||
# The alternative is to special case the last _ "0x5f" to mean something
|
||||
|
||||
def process(self):
|
||||
|
||||
# peek at the first bytes to see what type of file it is
|
||||
magic = self.fo.read(9)
|
||||
if (magic[0:1] == 'p') and (magic[2:9] == 'marker_'):
|
||||
first_token = 'info'
|
||||
elif (magic[0:1] == 'p') and (magic[2:9] == '__PAGE_'):
|
||||
skip = self.fo.read(2)
|
||||
first_token = 'info'
|
||||
elif (magic[0:1] == 'p') and (magic[2:8] == '_PAGE_'):
|
||||
first_token = 'info'
|
||||
elif (magic[0:1] == 'g') and (magic[2:9] == '__GLYPH'):
|
||||
skip = self.fo.read(3)
|
||||
first_token = 'info'
|
||||
else :
|
||||
# other0.dat file
|
||||
first_token = None
|
||||
self.fo.seek(-9,1)
|
||||
|
||||
|
||||
# main loop to read and build the document tree
|
||||
while True:
|
||||
|
||||
if first_token != None :
|
||||
# use "inserted" first token 'info' for page and glyph files
|
||||
tag = self.procToken(first_token)
|
||||
if len(tag) > 0 :
|
||||
self.doc.append(tag)
|
||||
first_token = None
|
||||
|
||||
v = self.getNext()
|
||||
if (v == None):
|
||||
break
|
||||
|
||||
if (v == 0x72):
|
||||
self.doLoop72('number')
|
||||
elif (v > 0) and (v < self.dict.getSize()) :
|
||||
tag = self.procToken(self.dict.lookup(v))
|
||||
if len(tag) > 0 :
|
||||
self.doc.append(tag)
|
||||
else:
|
||||
if self.debug:
|
||||
print "Main Loop: Unknown value: %x" % v
|
||||
if (v == 0):
|
||||
if (self.peek(1) == 0x5f):
|
||||
skip = self.fo.read(1)
|
||||
first_token = 'info'
|
||||
|
||||
# now do snippet injection
|
||||
if len(self.snippetList) > 0 :
|
||||
if self.debug : print 'Injecting Snippets:'
|
||||
snippet = self.injectSnippets(self.snippetList[0])
|
||||
snipno = snippet[0]
|
||||
tag_add = snippet[1]
|
||||
if self.debug : print self.formatTag(tag_add)
|
||||
if len(tag_add) > 0:
|
||||
self.doc.append(tag_add)
|
||||
|
||||
# handle generation of xml output
|
||||
xmlpage = self.formatDoc(self.flat_xml)
|
||||
|
||||
return xmlpage
|
||||
|
||||
|
||||
def fromData(dict, fname):
|
||||
flat_xml = True
|
||||
debug = False
|
||||
pp = PageParser(fname, dict, debug, flat_xml)
|
||||
xmlpage = pp.process()
|
||||
return xmlpage
|
||||
|
||||
def getXML(dict, fname):
|
||||
flat_xml = False
|
||||
debug = False
|
||||
pp = PageParser(fname, dict, debug, flat_xml)
|
||||
xmlpage = pp.process()
|
||||
return xmlpage
|
||||
|
||||
def usage():
|
||||
print 'Usage: '
|
||||
print ' convert2xml.py dict0000.dat infile.dat '
|
||||
print ' '
|
||||
print ' Options:'
|
||||
print ' -h print this usage help message '
|
||||
print ' -d turn on debug output to check for potential errors '
|
||||
print ' --flat-xml output the flattened xml page description only '
|
||||
print ' '
|
||||
print ' This program will attempt to convert a page*.dat file or '
|
||||
print ' glyphs*.dat file, using the dict0000.dat file, to its xml description. '
|
||||
print ' '
|
||||
print ' Use "cmbtc_dump.py" first to unencrypt, uncompress, and dump '
|
||||
print ' the *.dat files from a Topaz format e-book.'
|
||||
|
||||
#
|
||||
# Main
|
||||
#
|
||||
|
||||
def main(argv):
|
||||
dictFile = ""
|
||||
pageFile = ""
|
||||
debug = False
|
||||
flat_xml = False
|
||||
printOutput = False
|
||||
if len(argv) == 0:
|
||||
printOutput = True
|
||||
argv = sys.argv
|
||||
|
||||
try:
|
||||
opts, args = getopt.getopt(argv[1:], "hd", ["flat-xml"])
|
||||
|
||||
except getopt.GetoptError, err:
|
||||
|
||||
# print help information and exit:
|
||||
print str(err) # will print something like "option -a not recognized"
|
||||
usage()
|
||||
sys.exit(2)
|
||||
|
||||
if len(opts) == 0 and len(args) == 0 :
|
||||
usage()
|
||||
sys.exit(2)
|
||||
|
||||
for o, a in opts:
|
||||
if o =="-d":
|
||||
debug=True
|
||||
if o =="-h":
|
||||
usage()
|
||||
sys.exit(0)
|
||||
if o =="--flat-xml":
|
||||
flat_xml = True
|
||||
|
||||
dictFile, pageFile = args[0], args[1]
|
||||
|
||||
# read in the string table dictionary
|
||||
dict = Dictionary(dictFile)
|
||||
# dict.dumpDict()
|
||||
|
||||
# create a page parser
|
||||
pp = PageParser(pageFile, dict, debug, flat_xml)
|
||||
|
||||
xmlpage = pp.process()
|
||||
|
||||
if printOutput:
|
||||
print xmlpage
|
||||
return 0
|
||||
|
||||
return xmlpage
|
||||
|
||||
if __name__ == '__main__':
|
||||
sys.exit(main(''))
|
||||
706
Calibre_Plugins/K4MobiDeDRM_plugin/flatxml2html.py
Normal file
706
Calibre_Plugins/K4MobiDeDRM_plugin/flatxml2html.py
Normal file
@@ -0,0 +1,706 @@
|
||||
#! /usr/bin/python
|
||||
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
|
||||
# For use with Topaz Scripts Version 2.6
|
||||
|
||||
import sys
|
||||
import csv
|
||||
import os
|
||||
import math
|
||||
import getopt
|
||||
from struct import pack
|
||||
from struct import unpack
|
||||
|
||||
|
||||
class DocParser(object):
|
||||
def __init__(self, flatxml, classlst, fileid, bookDir, gdict, fixedimage):
|
||||
self.id = os.path.basename(fileid).replace('.dat','')
|
||||
self.svgcount = 0
|
||||
self.docList = flatxml.split('\n')
|
||||
self.docSize = len(self.docList)
|
||||
self.classList = {}
|
||||
self.bookDir = bookDir
|
||||
self.gdict = gdict
|
||||
tmpList = classlst.split('\n')
|
||||
for pclass in tmpList:
|
||||
if pclass != '':
|
||||
# remove the leading period from the css name
|
||||
cname = pclass[1:]
|
||||
self.classList[cname] = True
|
||||
self.fixedimage = fixedimage
|
||||
self.ocrtext = []
|
||||
self.link_id = []
|
||||
self.link_title = []
|
||||
self.link_page = []
|
||||
self.link_href = []
|
||||
self.link_type = []
|
||||
self.dehyphen_rootid = []
|
||||
self.paracont_stemid = []
|
||||
self.parastems_stemid = []
|
||||
|
||||
|
||||
def getGlyph(self, gid):
|
||||
result = ''
|
||||
id='id="gl%d"' % gid
|
||||
return self.gdict.lookup(id)
|
||||
|
||||
def glyphs_to_image(self, glyphList):
|
||||
|
||||
def extract(path, key):
|
||||
b = path.find(key) + len(key)
|
||||
e = path.find(' ',b)
|
||||
return int(path[b:e])
|
||||
|
||||
svgDir = os.path.join(self.bookDir,'svg')
|
||||
|
||||
imgDir = os.path.join(self.bookDir,'img')
|
||||
imgname = self.id + '_%04d.svg' % self.svgcount
|
||||
imgfile = os.path.join(imgDir,imgname)
|
||||
|
||||
# get glyph information
|
||||
gxList = self.getData('info.glyph.x',0,-1)
|
||||
gyList = self.getData('info.glyph.y',0,-1)
|
||||
gidList = self.getData('info.glyph.glyphID',0,-1)
|
||||
|
||||
gids = []
|
||||
maxws = []
|
||||
maxhs = []
|
||||
xs = []
|
||||
ys = []
|
||||
gdefs = []
|
||||
|
||||
# get path defintions, positions, dimensions for ecah glyph
|
||||
# that makes up the image, and find min x and min y to reposition origin
|
||||
minx = -1
|
||||
miny = -1
|
||||
for j in glyphList:
|
||||
gid = gidList[j]
|
||||
gids.append(gid)
|
||||
|
||||
xs.append(gxList[j])
|
||||
if minx == -1: minx = gxList[j]
|
||||
else : minx = min(minx, gxList[j])
|
||||
|
||||
ys.append(gyList[j])
|
||||
if miny == -1: miny = gyList[j]
|
||||
else : miny = min(miny, gyList[j])
|
||||
|
||||
path = self.getGlyph(gid)
|
||||
gdefs.append(path)
|
||||
|
||||
maxws.append(extract(path,'width='))
|
||||
maxhs.append(extract(path,'height='))
|
||||
|
||||
|
||||
# change the origin to minx, miny and calc max height and width
|
||||
maxw = maxws[0] + xs[0] - minx
|
||||
maxh = maxhs[0] + ys[0] - miny
|
||||
for j in xrange(0, len(xs)):
|
||||
xs[j] = xs[j] - minx
|
||||
ys[j] = ys[j] - miny
|
||||
maxw = max( maxw, (maxws[j] + xs[j]) )
|
||||
maxh = max( maxh, (maxhs[j] + ys[j]) )
|
||||
|
||||
# open the image file for output
|
||||
ifile = open(imgfile,'w')
|
||||
ifile.write('<?xml version="1.0" standalone="no"?>\n')
|
||||
ifile.write('<!DOCTYPE svg PUBLIC "-//W3C/DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">\n')
|
||||
ifile.write('<svg width="%dpx" height="%dpx" viewBox="0 0 %d %d" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1">\n' % (math.floor(maxw/10), math.floor(maxh/10), maxw, maxh))
|
||||
ifile.write('<defs>\n')
|
||||
for j in xrange(0,len(gdefs)):
|
||||
ifile.write(gdefs[j])
|
||||
ifile.write('</defs>\n')
|
||||
for j in xrange(0,len(gids)):
|
||||
ifile.write('<use xlink:href="#gl%d" x="%d" y="%d" />\n' % (gids[j], xs[j], ys[j]))
|
||||
ifile.write('</svg>')
|
||||
ifile.close()
|
||||
|
||||
return 0
|
||||
|
||||
|
||||
|
||||
# return tag at line pos in document
|
||||
def lineinDoc(self, pos) :
|
||||
if (pos >= 0) and (pos < self.docSize) :
|
||||
item = self.docList[pos]
|
||||
if item.find('=') >= 0:
|
||||
(name, argres) = item.split('=',1)
|
||||
else :
|
||||
name = item
|
||||
argres = ''
|
||||
return name, argres
|
||||
|
||||
|
||||
# find tag in doc if within pos to end inclusive
|
||||
def findinDoc(self, tagpath, pos, end) :
|
||||
result = None
|
||||
if end == -1 :
|
||||
end = self.docSize
|
||||
else:
|
||||
end = min(self.docSize, end)
|
||||
foundat = -1
|
||||
for j in xrange(pos, end):
|
||||
item = self.docList[j]
|
||||
if item.find('=') >= 0:
|
||||
(name, argres) = item.split('=',1)
|
||||
else :
|
||||
name = item
|
||||
argres = ''
|
||||
if name.endswith(tagpath) :
|
||||
result = argres
|
||||
foundat = j
|
||||
break
|
||||
return foundat, result
|
||||
|
||||
|
||||
# return list of start positions for the tagpath
|
||||
def posinDoc(self, tagpath):
|
||||
startpos = []
|
||||
pos = 0
|
||||
res = ""
|
||||
while res != None :
|
||||
(foundpos, res) = self.findinDoc(tagpath, pos, -1)
|
||||
if res != None :
|
||||
startpos.append(foundpos)
|
||||
pos = foundpos + 1
|
||||
return startpos
|
||||
|
||||
|
||||
# returns a vector of integers for the tagpath
|
||||
def getData(self, tagpath, pos, end):
|
||||
argres=[]
|
||||
(foundat, argt) = self.findinDoc(tagpath, pos, end)
|
||||
if (argt != None) and (len(argt) > 0) :
|
||||
argList = argt.split('|')
|
||||
argres = [ int(strval) for strval in argList]
|
||||
return argres
|
||||
|
||||
|
||||
# get the class
|
||||
def getClass(self, pclass):
|
||||
nclass = pclass
|
||||
|
||||
# class names are an issue given topaz may start them with numerals (not allowed),
|
||||
# use a mix of cases (which cause some browsers problems), and actually
|
||||
# attach numbers after "_reclustered*" to the end to deal classeses that inherit
|
||||
# from a base class (but then not actually provide all of these _reclustereed
|
||||
# classes in the stylesheet!
|
||||
|
||||
# so we clean this up by lowercasing, prepend 'cl-', and getting any baseclass
|
||||
# that exists in the stylesheet first, and then adding this specific class
|
||||
# after
|
||||
|
||||
# also some class names have spaces in them so need to convert to dashes
|
||||
if nclass != None :
|
||||
nclass = nclass.replace(' ','-')
|
||||
classres = ''
|
||||
nclass = nclass.lower()
|
||||
nclass = 'cl-' + nclass
|
||||
baseclass = ''
|
||||
# graphic is the base class for captions
|
||||
if nclass.find('cl-cap-') >=0 :
|
||||
classres = 'graphic' + ' '
|
||||
else :
|
||||
# strip to find baseclass
|
||||
p = nclass.find('_')
|
||||
if p > 0 :
|
||||
baseclass = nclass[0:p]
|
||||
if baseclass in self.classList:
|
||||
classres += baseclass + ' '
|
||||
classres += nclass
|
||||
nclass = classres
|
||||
return nclass
|
||||
|
||||
|
||||
# develop a sorted description of the starting positions of
|
||||
# groups and regions on the page, as well as the page type
|
||||
def PageDescription(self):
|
||||
|
||||
def compare(x, y):
|
||||
(xtype, xval) = x
|
||||
(ytype, yval) = y
|
||||
if xval > yval:
|
||||
return 1
|
||||
if xval == yval:
|
||||
return 0
|
||||
return -1
|
||||
|
||||
result = []
|
||||
(pos, pagetype) = self.findinDoc('page.type',0,-1)
|
||||
|
||||
groupList = self.posinDoc('page.group')
|
||||
groupregionList = self.posinDoc('page.group.region')
|
||||
pageregionList = self.posinDoc('page.region')
|
||||
# integrate into one list
|
||||
for j in groupList:
|
||||
result.append(('grpbeg',j))
|
||||
for j in groupregionList:
|
||||
result.append(('gregion',j))
|
||||
for j in pageregionList:
|
||||
result.append(('pregion',j))
|
||||
result.sort(compare)
|
||||
|
||||
# insert group end and page end indicators
|
||||
inGroup = False
|
||||
j = 0
|
||||
while True:
|
||||
if j == len(result): break
|
||||
rtype = result[j][0]
|
||||
rval = result[j][1]
|
||||
if not inGroup and (rtype == 'grpbeg') :
|
||||
inGroup = True
|
||||
j = j + 1
|
||||
elif inGroup and (rtype in ('grpbeg', 'pregion')):
|
||||
result.insert(j,('grpend',rval))
|
||||
inGroup = False
|
||||
else:
|
||||
j = j + 1
|
||||
if inGroup:
|
||||
result.append(('grpend',-1))
|
||||
result.append(('pageend', -1))
|
||||
return pagetype, result
|
||||
|
||||
|
||||
|
||||
# build a description of the paragraph
|
||||
def getParaDescription(self, start, end, regtype):
|
||||
|
||||
result = []
|
||||
|
||||
# paragraph
|
||||
(pos, pclass) = self.findinDoc('paragraph.class',start,end)
|
||||
|
||||
pclass = self.getClass(pclass)
|
||||
|
||||
# build up a description of the paragraph in result and return it
|
||||
# first check for the basic - all words paragraph
|
||||
(pos, sfirst) = self.findinDoc('paragraph.firstWord',start,end)
|
||||
(pos, slast) = self.findinDoc('paragraph.lastWord',start,end)
|
||||
if (sfirst != None) and (slast != None) :
|
||||
first = int(sfirst)
|
||||
last = int(slast)
|
||||
|
||||
makeImage = (regtype == 'vertical') or (regtype == 'table')
|
||||
if self.fixedimage:
|
||||
makeImage = makeImage or (regtype == 'fixed')
|
||||
|
||||
if (pclass != None):
|
||||
makeImage = makeImage or (pclass.find('.inverted') >= 0)
|
||||
if self.fixedimage :
|
||||
makeImage = makeImage or (pclass.find('cl-f-') >= 0)
|
||||
|
||||
if not makeImage :
|
||||
# standard all word paragraph
|
||||
for wordnum in xrange(first, last):
|
||||
result.append(('ocr', wordnum))
|
||||
return pclass, result
|
||||
|
||||
# convert paragraph to svg image
|
||||
# translate first and last word into first and last glyphs
|
||||
# and generate inline image and include it
|
||||
glyphList = []
|
||||
firstglyphList = self.getData('word.firstGlyph',0,-1)
|
||||
gidList = self.getData('info.glyph.glyphID',0,-1)
|
||||
firstGlyph = firstglyphList[first]
|
||||
if last < len(firstglyphList):
|
||||
lastGlyph = firstglyphList[last]
|
||||
else :
|
||||
lastGlyph = len(gidList)
|
||||
for glyphnum in xrange(firstGlyph, lastGlyph):
|
||||
glyphList.append(glyphnum)
|
||||
# include any extratokens if they exist
|
||||
(pos, sfg) = self.findinDoc('extratokens.firstGlyph',start,end)
|
||||
(pos, slg) = self.findinDoc('extratokens.lastGlyph',start,end)
|
||||
if (sfg != None) and (slg != None):
|
||||
for glyphnum in xrange(int(sfg), int(slg)):
|
||||
glyphList.append(glyphnum)
|
||||
num = self.svgcount
|
||||
self.glyphs_to_image(glyphList)
|
||||
self.svgcount += 1
|
||||
result.append(('svg', num))
|
||||
return pclass, result
|
||||
|
||||
# this type of paragraph may be made up of multiple spans, inline
|
||||
# word monograms (images), and words with semantic meaning,
|
||||
# plus glyphs used to form starting letter of first word
|
||||
|
||||
# need to parse this type line by line
|
||||
line = start + 1
|
||||
word_class = ''
|
||||
|
||||
# if end is -1 then we must search to end of document
|
||||
if end == -1 :
|
||||
end = self.docSize
|
||||
|
||||
# seems some xml has last* coming before first* so we have to
|
||||
# handle any order
|
||||
sp_first = -1
|
||||
sp_last = -1
|
||||
|
||||
gl_first = -1
|
||||
gl_last = -1
|
||||
|
||||
ws_first = -1
|
||||
ws_last = -1
|
||||
|
||||
word_class = ''
|
||||
|
||||
while (line < end) :
|
||||
|
||||
(name, argres) = self.lineinDoc(line)
|
||||
|
||||
if name.endswith('span.firstWord') :
|
||||
sp_first = int(argres)
|
||||
|
||||
elif name.endswith('span.lastWord') :
|
||||
sp_last = int(argres)
|
||||
|
||||
elif name.endswith('word.firstGlyph') :
|
||||
gl_first = int(argres)
|
||||
|
||||
elif name.endswith('word.lastGlyph') :
|
||||
gl_last = int(argres)
|
||||
|
||||
elif name.endswith('word_semantic.firstWord'):
|
||||
ws_first = int(argres)
|
||||
|
||||
elif name.endswith('word_semantic.lastWord'):
|
||||
ws_last = int(argres)
|
||||
|
||||
elif name.endswith('word.class'):
|
||||
(cname, space) = argres.split('-',1)
|
||||
if space == '' : space = '0'
|
||||
if (cname == 'spaceafter') and (int(space) > 0) :
|
||||
word_class = 'sa'
|
||||
|
||||
elif name.endswith('word.img.src'):
|
||||
result.append(('img' + word_class, int(argres)))
|
||||
word_class = ''
|
||||
|
||||
elif name.endswith('region.img.src'):
|
||||
result.append(('img' + word_class, int(argres)))
|
||||
|
||||
if (sp_first != -1) and (sp_last != -1):
|
||||
for wordnum in xrange(sp_first, sp_last):
|
||||
result.append(('ocr', wordnum))
|
||||
sp_first = -1
|
||||
sp_last = -1
|
||||
|
||||
if (gl_first != -1) and (gl_last != -1):
|
||||
glyphList = []
|
||||
for glyphnum in xrange(gl_first, gl_last):
|
||||
glyphList.append(glyphnum)
|
||||
num = self.svgcount
|
||||
self.glyphs_to_image(glyphList)
|
||||
self.svgcount += 1
|
||||
result.append(('svg', num))
|
||||
gl_first = -1
|
||||
gl_last = -1
|
||||
|
||||
if (ws_first != -1) and (ws_last != -1):
|
||||
for wordnum in xrange(ws_first, ws_last):
|
||||
result.append(('ocr', wordnum))
|
||||
ws_first = -1
|
||||
ws_last = -1
|
||||
|
||||
line += 1
|
||||
|
||||
return pclass, result
|
||||
|
||||
|
||||
def buildParagraph(self, pclass, pdesc, type, regtype) :
|
||||
parares = ''
|
||||
sep =''
|
||||
|
||||
classres = ''
|
||||
if pclass :
|
||||
classres = ' class="' + pclass + '"'
|
||||
|
||||
br_lb = (regtype == 'fixed') or (regtype == 'chapterheading') or (regtype == 'vertical')
|
||||
|
||||
handle_links = len(self.link_id) > 0
|
||||
|
||||
if (type == 'full') or (type == 'begin') :
|
||||
parares += '<p' + classres + '>'
|
||||
|
||||
if (type == 'end'):
|
||||
parares += ' '
|
||||
|
||||
lstart = len(parares)
|
||||
|
||||
cnt = len(pdesc)
|
||||
|
||||
for j in xrange( 0, cnt) :
|
||||
|
||||
(wtype, num) = pdesc[j]
|
||||
|
||||
if wtype == 'ocr' :
|
||||
word = self.ocrtext[num]
|
||||
sep = ' '
|
||||
|
||||
if handle_links:
|
||||
link = self.link_id[num]
|
||||
if (link > 0):
|
||||
linktype = self.link_type[link-1]
|
||||
title = self.link_title[link-1]
|
||||
if (title == "") or (parares.rfind(title) < 0):
|
||||
title=parares[lstart:]
|
||||
if linktype == 'external' :
|
||||
linkhref = self.link_href[link-1]
|
||||
linkhtml = '<a href="%s">' % linkhref
|
||||
else :
|
||||
if len(self.link_page) >= link :
|
||||
ptarget = self.link_page[link-1] - 1
|
||||
linkhtml = '<a href="#page%04d">' % ptarget
|
||||
else :
|
||||
# just link to the current page
|
||||
linkhtml = '<a href="#' + self.id + '">'
|
||||
linkhtml += title + '</a>'
|
||||
pos = parares.rfind(title)
|
||||
if pos >= 0:
|
||||
parares = parares[0:pos] + linkhtml + parares[pos+len(title):]
|
||||
else :
|
||||
parares += linkhtml
|
||||
lstart = len(parares)
|
||||
if word == '_link_' : word = ''
|
||||
elif (link < 0) :
|
||||
if word == '_link_' : word = ''
|
||||
|
||||
if word == '_lb_':
|
||||
if ((num-1) in self.dehyphen_rootid ) or handle_links:
|
||||
word = ''
|
||||
sep = ''
|
||||
elif br_lb :
|
||||
word = '<br />\n'
|
||||
sep = ''
|
||||
else :
|
||||
word = '\n'
|
||||
sep = ''
|
||||
|
||||
if num in self.dehyphen_rootid :
|
||||
word = word[0:-1]
|
||||
sep = ''
|
||||
|
||||
parares += word + sep
|
||||
|
||||
elif wtype == 'img' :
|
||||
sep = ''
|
||||
parares += '<img src="img/img%04d.jpg" alt="" />' % num
|
||||
parares += sep
|
||||
|
||||
elif wtype == 'imgsa' :
|
||||
sep = ' '
|
||||
parares += '<img src="img/img%04d.jpg" alt="" />' % num
|
||||
parares += sep
|
||||
|
||||
elif wtype == 'svg' :
|
||||
sep = ''
|
||||
parares += '<img src="img/' + self.id + '_%04d.svg" alt="" />' % num
|
||||
parares += sep
|
||||
|
||||
if len(sep) > 0 : parares = parares[0:-1]
|
||||
if (type == 'full') or (type == 'end') :
|
||||
parares += '</p>'
|
||||
return parares
|
||||
|
||||
|
||||
|
||||
# walk the document tree collecting the information needed
|
||||
# to build an html page using the ocrText
|
||||
|
||||
def process(self):
|
||||
|
||||
htmlpage = ''
|
||||
|
||||
# get the ocr text
|
||||
(pos, argres) = self.findinDoc('info.word.ocrText',0,-1)
|
||||
if argres : self.ocrtext = argres.split('|')
|
||||
|
||||
# get information to dehyphenate the text
|
||||
self.dehyphen_rootid = self.getData('info.dehyphen.rootID',0,-1)
|
||||
|
||||
# determine if first paragraph is continued from previous page
|
||||
(pos, self.parastems_stemid) = self.findinDoc('info.paraStems.stemID',0,-1)
|
||||
first_para_continued = (self.parastems_stemid != None)
|
||||
|
||||
# determine if last paragraph is continued onto the next page
|
||||
(pos, self.paracont_stemid) = self.findinDoc('info.paraCont.stemID',0,-1)
|
||||
last_para_continued = (self.paracont_stemid != None)
|
||||
|
||||
# collect link ids
|
||||
self.link_id = self.getData('info.word.link_id',0,-1)
|
||||
|
||||
# collect link destination page numbers
|
||||
self.link_page = self.getData('info.links.page',0,-1)
|
||||
|
||||
# collect link types (container versus external)
|
||||
(pos, argres) = self.findinDoc('info.links.type',0,-1)
|
||||
if argres : self.link_type = argres.split('|')
|
||||
|
||||
# collect link destinations
|
||||
(pos, argres) = self.findinDoc('info.links.href',0,-1)
|
||||
if argres : self.link_href = argres.split('|')
|
||||
|
||||
# collect link titles
|
||||
(pos, argres) = self.findinDoc('info.links.title',0,-1)
|
||||
if argres :
|
||||
self.link_title = argres.split('|')
|
||||
else:
|
||||
self.link_title.append('')
|
||||
|
||||
# get a descriptions of the starting points of the regions
|
||||
# and groups on the page
|
||||
(pagetype, pageDesc) = self.PageDescription()
|
||||
regcnt = len(pageDesc) - 1
|
||||
|
||||
anchorSet = False
|
||||
breakSet = False
|
||||
inGroup = False
|
||||
|
||||
# process each region on the page and convert what you can to html
|
||||
|
||||
for j in xrange(regcnt):
|
||||
|
||||
(etype, start) = pageDesc[j]
|
||||
(ntype, end) = pageDesc[j+1]
|
||||
|
||||
|
||||
# set anchor for link target on this page
|
||||
if not anchorSet and not first_para_continued:
|
||||
htmlpage += '<div style="visibility: hidden; height: 0; width: 0;" id="'
|
||||
htmlpage += self.id + '" title="pagetype_' + pagetype + '"></div>\n'
|
||||
anchorSet = True
|
||||
|
||||
# handle groups of graphics with text captions
|
||||
if (etype == 'grpbeg'):
|
||||
(pos, grptype) = self.findinDoc('group.type', start, end)
|
||||
if grptype != None:
|
||||
if grptype == 'graphic':
|
||||
gcstr = ' class="' + grptype + '"'
|
||||
htmlpage += '<div' + gcstr + '>'
|
||||
inGroup = True
|
||||
|
||||
elif (etype == 'grpend'):
|
||||
if inGroup:
|
||||
htmlpage += '</div>\n'
|
||||
inGroup = False
|
||||
|
||||
else:
|
||||
(pos, regtype) = self.findinDoc('region.type',start,end)
|
||||
|
||||
if regtype == 'graphic' :
|
||||
(pos, simgsrc) = self.findinDoc('img.src',start,end)
|
||||
if simgsrc:
|
||||
if inGroup:
|
||||
htmlpage += '<img src="img/img%04d.jpg" alt="" />' % int(simgsrc)
|
||||
else:
|
||||
htmlpage += '<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc)
|
||||
|
||||
elif regtype == 'chapterheading' :
|
||||
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
|
||||
if not breakSet:
|
||||
htmlpage += '<div style="page-break-after: always;"> </div>\n'
|
||||
breakSet = True
|
||||
tag = 'h1'
|
||||
if pclass and (len(pclass) >= 7):
|
||||
if pclass[3:7] == 'ch1-' : tag = 'h1'
|
||||
if pclass[3:7] == 'ch2-' : tag = 'h2'
|
||||
if pclass[3:7] == 'ch3-' : tag = 'h3'
|
||||
htmlpage += '<' + tag + ' class="' + pclass + '">'
|
||||
else:
|
||||
htmlpage += '<' + tag + '>'
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, 'middle', regtype)
|
||||
htmlpage += '</' + tag + '>'
|
||||
|
||||
elif (regtype == 'text') or (regtype == 'fixed') or (regtype == 'insert') or (regtype == 'listitem'):
|
||||
ptype = 'full'
|
||||
# check to see if this is a continution from the previous page
|
||||
if first_para_continued :
|
||||
ptype = 'end'
|
||||
first_para_continued = False
|
||||
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
|
||||
if pclass and (len(pclass) >= 6) and (ptype == 'full'):
|
||||
tag = 'p'
|
||||
if pclass[3:6] == 'h1-' : tag = 'h4'
|
||||
if pclass[3:6] == 'h2-' : tag = 'h5'
|
||||
if pclass[3:6] == 'h3-' : tag = 'h6'
|
||||
htmlpage += '<' + tag + ' class="' + pclass + '">'
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, 'middle', regtype)
|
||||
htmlpage += '</' + tag + '>'
|
||||
else :
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, ptype, regtype)
|
||||
|
||||
elif (regtype == 'tocentry') :
|
||||
ptype = 'full'
|
||||
if first_para_continued :
|
||||
ptype = 'end'
|
||||
first_para_continued = False
|
||||
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, ptype, regtype)
|
||||
|
||||
|
||||
elif (regtype == 'vertical') or (regtype == 'table') :
|
||||
ptype = 'full'
|
||||
if inGroup:
|
||||
ptype = 'middle'
|
||||
if first_para_continued :
|
||||
ptype = 'end'
|
||||
first_para_continued = False
|
||||
(pclass, pdesc) = self.getParaDescription(start, end, regtype)
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, ptype, regtype)
|
||||
|
||||
|
||||
elif (regtype == 'synth_fcvr.center'):
|
||||
(pos, simgsrc) = self.findinDoc('img.src',start,end)
|
||||
if simgsrc:
|
||||
htmlpage += '<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc)
|
||||
|
||||
else :
|
||||
print ' Making region type', regtype,
|
||||
(pos, temp) = self.findinDoc('paragraph',start,end)
|
||||
(pos2, temp) = self.findinDoc('span',start,end)
|
||||
if pos != -1 or pos2 != -1:
|
||||
print ' a "text" region'
|
||||
orig_regtype = regtype
|
||||
regtype = 'fixed'
|
||||
ptype = 'full'
|
||||
# check to see if this is a continution from the previous page
|
||||
if first_para_continued :
|
||||
ptype = 'end'
|
||||
first_para_continued = False
|
||||
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
|
||||
if not pclass:
|
||||
if orig_regtype.endswith('.right') : pclass = 'cl-right'
|
||||
elif orig_regtype.endswith('.center') : pclass = 'cl-center'
|
||||
elif orig_regtype.endswith('.left') : pclass = 'cl-left'
|
||||
elif orig_regtype.endswith('.justify') : pclass = 'cl-justify'
|
||||
if pclass and (ptype == 'full') and (len(pclass) >= 6):
|
||||
tag = 'p'
|
||||
if pclass[3:6] == 'h1-' : tag = 'h4'
|
||||
if pclass[3:6] == 'h2-' : tag = 'h5'
|
||||
if pclass[3:6] == 'h3-' : tag = 'h6'
|
||||
htmlpage += '<' + tag + ' class="' + pclass + '">'
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, 'middle', regtype)
|
||||
htmlpage += '</' + tag + '>'
|
||||
else :
|
||||
htmlpage += self.buildParagraph(pclass, pdesc, ptype, regtype)
|
||||
else :
|
||||
print ' a "graphic" region'
|
||||
(pos, simgsrc) = self.findinDoc('img.src',start,end)
|
||||
if simgsrc:
|
||||
htmlpage += '<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc)
|
||||
|
||||
|
||||
if last_para_continued :
|
||||
if htmlpage[-4:] == '</p>':
|
||||
htmlpage = htmlpage[0:-4]
|
||||
last_para_continued = False
|
||||
|
||||
return htmlpage
|
||||
|
||||
|
||||
|
||||
def convert2HTML(flatxml, classlst, fileid, bookDir, gdict, fixedimage):
|
||||
# create a document parser
|
||||
dp = DocParser(flatxml, classlst, fileid, bookDir, gdict, fixedimage)
|
||||
htmlpage = dp.process()
|
||||
return htmlpage
|
||||
151
Calibre_Plugins/K4MobiDeDRM_plugin/flatxml2svg.py
Normal file
151
Calibre_Plugins/K4MobiDeDRM_plugin/flatxml2svg.py
Normal file
@@ -0,0 +1,151 @@
|
||||
#! /usr/bin/python
|
||||
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
|
||||
|
||||
import sys
|
||||
import csv
|
||||
import os
|
||||
import getopt
|
||||
from struct import pack
|
||||
from struct import unpack
|
||||
|
||||
|
||||
class PParser(object):
|
||||
def __init__(self, gd, flatxml):
|
||||
self.gd = gd
|
||||
self.flatdoc = flatxml.split('\n')
|
||||
self.temp = []
|
||||
foo = self.getData('page.h') or self.getData('book.h')
|
||||
self.ph = foo[0]
|
||||
foo = self.getData('page.w') or self.getData('book.w')
|
||||
self.pw = foo[0]
|
||||
self.gx = self.getData('info.glyph.x')
|
||||
self.gy = self.getData('info.glyph.y')
|
||||
self.gid = self.getData('info.glyph.glyphID')
|
||||
def getData(self, path):
|
||||
result = None
|
||||
cnt = len(self.flatdoc)
|
||||
for j in xrange(cnt):
|
||||
item = self.flatdoc[j]
|
||||
if item.find('=') >= 0:
|
||||
(name, argt) = item.split('=')
|
||||
argres = argt.split('|')
|
||||
else:
|
||||
name = item
|
||||
argres = []
|
||||
if (name.endswith(path)):
|
||||
result = argres
|
||||
break
|
||||
if (len(argres) > 0) :
|
||||
for j in xrange(0,len(argres)):
|
||||
argres[j] = int(argres[j])
|
||||
return result
|
||||
def getDataTemp(self, path):
|
||||
result = None
|
||||
cnt = len(self.temp)
|
||||
for j in xrange(cnt):
|
||||
item = self.temp[j]
|
||||
if item.find('=') >= 0:
|
||||
(name, argt) = item.split('=')
|
||||
argres = argt.split('|')
|
||||
else:
|
||||
name = item
|
||||
argres = []
|
||||
if (name.endswith(path)):
|
||||
result = argres
|
||||
self.temp.pop(j)
|
||||
break
|
||||
if (len(argres) > 0) :
|
||||
for j in xrange(0,len(argres)):
|
||||
argres[j] = int(argres[j])
|
||||
return result
|
||||
def getImages(self):
|
||||
result = []
|
||||
self.temp = self.flatdoc
|
||||
while (self.getDataTemp('img') != None):
|
||||
h = self.getDataTemp('img.h')[0]
|
||||
w = self.getDataTemp('img.w')[0]
|
||||
x = self.getDataTemp('img.x')[0]
|
||||
y = self.getDataTemp('img.y')[0]
|
||||
src = self.getDataTemp('img.src')[0]
|
||||
result.append('<image xlink:href="../img/img%04d.jpg" x="%d" y="%d" width="%d" height="%d" />\n' % (src, x, y, w, h))
|
||||
return result
|
||||
def getGlyphs(self):
|
||||
result = []
|
||||
if (self.gid != None) and (len(self.gid) > 0):
|
||||
glyphs = []
|
||||
for j in set(self.gid):
|
||||
glyphs.append(j)
|
||||
glyphs.sort()
|
||||
for gid in glyphs:
|
||||
id='id="gl%d"' % gid
|
||||
path = self.gd.lookup(id)
|
||||
if path:
|
||||
result.append(id + ' ' + path)
|
||||
return result
|
||||
|
||||
|
||||
def convert2SVG(gdict, flat_xml, counter, numfiles, svgDir, raw, meta_array, scaledpi):
|
||||
ml = ''
|
||||
pp = PParser(gdict, flat_xml)
|
||||
ml += '<?xml version="1.0" standalone="no"?>\n'
|
||||
if (raw):
|
||||
ml += '<!DOCTYPE svg PUBLIC "-//W3C/DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">\n'
|
||||
ml += '<svg width="%fin" height="%fin" viewBox="0 0 %d %d" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1">\n' % (pp.pw / scaledpi, pp.ph / scaledpi, pp.pw -1, pp.ph -1)
|
||||
ml += '<title>Page %d - %s by %s</title>\n' % (counter, meta_array['Title'],meta_array['Authors'])
|
||||
else:
|
||||
ml += '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">\n'
|
||||
ml += '<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" ><head>\n'
|
||||
ml += '<title>Page %d - %s by %s</title>\n' % (counter, meta_array['Title'],meta_array['Authors'])
|
||||
ml += '<script><![CDATA[\n'
|
||||
ml += 'function gd(){var p=window.location.href.replace(/^.*\?dpi=(\d+).*$/i,"$1");return p;}\n'
|
||||
ml += 'var dpi=%d;\n' % scaledpi
|
||||
if (counter) :
|
||||
ml += 'var prevpage="page%04d.xhtml";\n' % (counter - 1)
|
||||
if (counter < numfiles-1) :
|
||||
ml += 'var nextpage="page%04d.xhtml";\n' % (counter + 1)
|
||||
ml += 'var pw=%d;var ph=%d;' % (pp.pw, pp.ph)
|
||||
ml += 'function zoomin(){dpi=dpi*(0.8);setsize();}\n'
|
||||
ml += 'function zoomout(){dpi=dpi*1.25;setsize();}\n'
|
||||
ml += 'function setsize(){var svg=document.getElementById("svgimg");var prev=document.getElementById("prevsvg");var next=document.getElementById("nextsvg");var width=(pw/dpi)+"in";var height=(ph/dpi)+"in";svg.setAttribute("width",width);svg.setAttribute("height",height);prev.setAttribute("height",height);prev.setAttribute("width","50px");next.setAttribute("height",height);next.setAttribute("width","50px");}\n'
|
||||
ml += 'function ppage(){window.location.href=prevpage+"?dpi="+Math.round(dpi);}\n'
|
||||
ml += 'function npage(){window.location.href=nextpage+"?dpi="+Math.round(dpi);}\n'
|
||||
ml += 'var gt=gd();if(gt>0){dpi=gt;}\n'
|
||||
ml += 'window.onload=setsize;\n'
|
||||
ml += ']]></script>\n'
|
||||
ml += '</head>\n'
|
||||
ml += '<body onLoad="setsize();" style="background-color:#777;text-align:center;">\n'
|
||||
ml += '<div style="white-space:nowrap;">\n'
|
||||
if (counter == 0) :
|
||||
ml += '<a href="javascript:ppage();"><svg id="prevsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"></svg></a>\n'
|
||||
else:
|
||||
ml += '<a href="javascript:ppage();"><svg id="prevsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"><polygon points="5,150,95,5,95,295" fill="#AAAAAA" /></svg></a>\n'
|
||||
ml += '<a href="javascript:npage();"><svg id="svgimg" viewBox="0 0 %d %d" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" style="background-color:#FFF;border:1px solid black;">' % (pp.pw, pp.ph)
|
||||
if (pp.gid != None):
|
||||
ml += '<defs>\n'
|
||||
gdefs = pp.getGlyphs()
|
||||
for j in xrange(0,len(gdefs)):
|
||||
ml += gdefs[j]
|
||||
ml += '</defs>\n'
|
||||
img = pp.getImages()
|
||||
if (img != None):
|
||||
for j in xrange(0,len(img)):
|
||||
ml += img[j]
|
||||
if (pp.gid != None):
|
||||
for j in xrange(0,len(pp.gid)):
|
||||
ml += '<use xlink:href="#gl%d" x="%d" y="%d" />\n' % (pp.gid[j], pp.gx[j], pp.gy[j])
|
||||
if (img == None or len(img) == 0) and (pp.gid == None or len(pp.gid) == 0):
|
||||
ml += '<text x="10" y="10" font-family="Helvetica" font-size="100" stroke="black">This page intentionally left blank.</text>\n<text x="10" y="110" font-family="Helvetica" font-size="50" stroke="black">Until this notice unintentionally gave it content. (gensvg.py)</text>\n'
|
||||
if (raw) :
|
||||
ml += '</svg>'
|
||||
else :
|
||||
ml += '</svg></a>\n'
|
||||
if (counter == numfiles - 1) :
|
||||
ml += '<a href="javascript:npage();"><svg id="nextsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"></svg></a>\n'
|
||||
else :
|
||||
ml += '<a href="javascript:npage();"><svg id="nextsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"><polygon points="5,5,5,295,95,150" fill="#AAAAAA" /></svg></a>\n'
|
||||
ml += '</div>\n'
|
||||
ml += '<div><a href="javascript:zoomin();">zoom in</a> - <a href="javascript:zoomout();">zoom out</a></div>\n'
|
||||
ml += '</body>\n'
|
||||
ml += '</html>\n'
|
||||
return ml
|
||||
|
||||
561
Calibre_Plugins/K4MobiDeDRM_plugin/genbook.py
Normal file
561
Calibre_Plugins/K4MobiDeDRM_plugin/genbook.py
Normal file
@@ -0,0 +1,561 @@
|
||||
#! /usr/bin/python
|
||||
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
|
||||
|
||||
class Unbuffered:
|
||||
def __init__(self, stream):
|
||||
self.stream = stream
|
||||
def write(self, data):
|
||||
self.stream.write(data)
|
||||
self.stream.flush()
|
||||
def __getattr__(self, attr):
|
||||
return getattr(self.stream, attr)
|
||||
|
||||
import sys
|
||||
sys.stdout=Unbuffered(sys.stdout)
|
||||
|
||||
import csv
|
||||
import os
|
||||
import getopt
|
||||
from struct import pack
|
||||
from struct import unpack
|
||||
|
||||
|
||||
# local support routines
|
||||
import convert2xml
|
||||
import flatxml2html
|
||||
import flatxml2svg
|
||||
import stylexml2css
|
||||
|
||||
|
||||
# Get a 7 bit encoded number from a file
|
||||
def readEncodedNumber(file):
|
||||
flag = False
|
||||
c = file.read(1)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
data = ord(c)
|
||||
if data == 0xFF:
|
||||
flag = True
|
||||
c = file.read(1)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
data = ord(c)
|
||||
if data >= 0x80:
|
||||
datax = (data & 0x7F)
|
||||
while data >= 0x80 :
|
||||
c = file.read(1)
|
||||
if (len(c) == 0):
|
||||
return None
|
||||
data = ord(c)
|
||||
datax = (datax <<7) + (data & 0x7F)
|
||||
data = datax
|
||||
if flag:
|
||||
data = -data
|
||||
return data
|
||||
|
||||
# Get a length prefixed string from the file
|
||||
def lengthPrefixString(data):
|
||||
return encodeNumber(len(data))+data
|
||||
|
||||
def readString(file):
|
||||
stringLength = readEncodedNumber(file)
|
||||
if (stringLength == None):
|
||||
return None
|
||||
sv = file.read(stringLength)
|
||||
if (len(sv) != stringLength):
|
||||
return ""
|
||||
return unpack(str(stringLength)+"s",sv)[0]
|
||||
|
||||
def getMetaArray(metaFile):
|
||||
# parse the meta file
|
||||
result = {}
|
||||
fo = file(metaFile,'rb')
|
||||
size = readEncodedNumber(fo)
|
||||
for i in xrange(size):
|
||||
tag = readString(fo)
|
||||
value = readString(fo)
|
||||
result[tag] = value
|
||||
# print tag, value
|
||||
fo.close()
|
||||
return result
|
||||
|
||||
|
||||
# dictionary of all text strings by index value
|
||||
class Dictionary(object):
|
||||
def __init__(self, dictFile):
|
||||
self.filename = dictFile
|
||||
self.size = 0
|
||||
self.fo = file(dictFile,'rb')
|
||||
self.stable = []
|
||||
self.size = readEncodedNumber(self.fo)
|
||||
for i in xrange(self.size):
|
||||
self.stable.append(self.escapestr(readString(self.fo)))
|
||||
self.pos = 0
|
||||
def escapestr(self, str):
|
||||
str = str.replace('&','&')
|
||||
str = str.replace('<','<')
|
||||
str = str.replace('>','>')
|
||||
str = str.replace('=','=')
|
||||
return str
|
||||
def lookup(self,val):
|
||||
if ((val >= 0) and (val < self.size)) :
|
||||
self.pos = val
|
||||
return self.stable[self.pos]
|
||||
else:
|
||||
print "Error - %d outside of string table limits" % val
|
||||
sys.exit(-1)
|
||||
def getSize(self):
|
||||
return self.size
|
||||
def getPos(self):
|
||||
return self.pos
|
||||
|
||||
|
||||
class PageDimParser(object):
|
||||
def __init__(self, flatxml):
|
||||
self.flatdoc = flatxml.split('\n')
|
||||
# find tag if within pos to end inclusive
|
||||
def findinDoc(self, tagpath, pos, end) :
|
||||
result = None
|
||||
docList = self.flatdoc
|
||||
cnt = len(docList)
|
||||
if end == -1 :
|
||||
end = cnt
|
||||
else:
|
||||
end = min(cnt,end)
|
||||
foundat = -1
|
||||
for j in xrange(pos, end):
|
||||
item = docList[j]
|
||||
if item.find('=') >= 0:
|
||||
(name, argres) = item.split('=')
|
||||
else :
|
||||
name = item
|
||||
argres = ''
|
||||
if name.endswith(tagpath) :
|
||||
result = argres
|
||||
foundat = j
|
||||
break
|
||||
return foundat, result
|
||||
def process(self):
|
||||
(pos, sph) = self.findinDoc('page.h',0,-1)
|
||||
(pos, spw) = self.findinDoc('page.w',0,-1)
|
||||
if (sph == None): sph = '-1'
|
||||
if (spw == None): spw = '-1'
|
||||
return sph, spw
|
||||
|
||||
def getPageDim(flatxml):
|
||||
# create a document parser
|
||||
dp = PageDimParser(flatxml)
|
||||
(ph, pw) = dp.process()
|
||||
return ph, pw
|
||||
|
||||
class GParser(object):
|
||||
def __init__(self, flatxml):
|
||||
self.flatdoc = flatxml.split('\n')
|
||||
self.dpi = 1440
|
||||
self.gh = self.getData('info.glyph.h')
|
||||
self.gw = self.getData('info.glyph.w')
|
||||
self.guse = self.getData('info.glyph.use')
|
||||
if self.guse :
|
||||
self.count = len(self.guse)
|
||||
else :
|
||||
self.count = 0
|
||||
self.gvtx = self.getData('info.glyph.vtx')
|
||||
self.glen = self.getData('info.glyph.len')
|
||||
self.gdpi = self.getData('info.glyph.dpi')
|
||||
self.vx = self.getData('info.vtx.x')
|
||||
self.vy = self.getData('info.vtx.y')
|
||||
self.vlen = self.getData('info.len.n')
|
||||
if self.vlen :
|
||||
self.glen.append(len(self.vlen))
|
||||
elif self.glen:
|
||||
self.glen.append(0)
|
||||
if self.vx :
|
||||
self.gvtx.append(len(self.vx))
|
||||
elif self.gvtx :
|
||||
self.gvtx.append(0)
|
||||
def getData(self, path):
|
||||
result = None
|
||||
cnt = len(self.flatdoc)
|
||||
for j in xrange(cnt):
|
||||
item = self.flatdoc[j]
|
||||
if item.find('=') >= 0:
|
||||
(name, argt) = item.split('=')
|
||||
argres = argt.split('|')
|
||||
else:
|
||||
name = item
|
||||
argres = []
|
||||
if (name == path):
|
||||
result = argres
|
||||
break
|
||||
if (len(argres) > 0) :
|
||||
for j in xrange(0,len(argres)):
|
||||
argres[j] = int(argres[j])
|
||||
return result
|
||||
def getGlyphDim(self, gly):
|
||||
maxh = (self.gh[gly] * self.dpi) / self.gdpi[gly]
|
||||
maxw = (self.gw[gly] * self.dpi) / self.gdpi[gly]
|
||||
return maxh, maxw
|
||||
def getPath(self, gly):
|
||||
path = ''
|
||||
if (gly < 0) or (gly >= self.count):
|
||||
return path
|
||||
tx = self.vx[self.gvtx[gly]:self.gvtx[gly+1]]
|
||||
ty = self.vy[self.gvtx[gly]:self.gvtx[gly+1]]
|
||||
p = 0
|
||||
for k in xrange(self.glen[gly], self.glen[gly+1]):
|
||||
if (p == 0):
|
||||
zx = tx[0:self.vlen[k]+1]
|
||||
zy = ty[0:self.vlen[k]+1]
|
||||
else:
|
||||
zx = tx[self.vlen[k-1]+1:self.vlen[k]+1]
|
||||
zy = ty[self.vlen[k-1]+1:self.vlen[k]+1]
|
||||
p += 1
|
||||
j = 0
|
||||
while ( j < len(zx) ):
|
||||
if (j == 0):
|
||||
# Start Position.
|
||||
path += 'M %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly])
|
||||
elif (j <= len(zx)-3):
|
||||
# Cubic Bezier Curve
|
||||
path += 'C %d %d %d %d %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly], zx[j+1] * self.dpi / self.gdpi[gly], zy[j+1] * self.dpi / self.gdpi[gly], zx[j+2] * self.dpi / self.gdpi[gly], zy[j+2] * self.dpi / self.gdpi[gly])
|
||||
j += 2
|
||||
elif (j == len(zx)-2):
|
||||
# Cubic Bezier Curve to Start Position
|
||||
path += 'C %d %d %d %d %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly], zx[j+1] * self.dpi / self.gdpi[gly], zy[j+1] * self.dpi / self.gdpi[gly], zx[0] * self.dpi / self.gdpi[gly], zy[0] * self.dpi / self.gdpi[gly])
|
||||
j += 1
|
||||
elif (j == len(zx)-1):
|
||||
# Quadratic Bezier Curve to Start Position
|
||||
path += 'Q %d %d %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly], zx[0] * self.dpi / self.gdpi[gly], zy[0] * self.dpi / self.gdpi[gly])
|
||||
|
||||
j += 1
|
||||
path += 'z'
|
||||
return path
|
||||
|
||||
|
||||
|
||||
# dictionary of all text strings by index value
|
||||
class GlyphDict(object):
|
||||
def __init__(self):
|
||||
self.gdict = {}
|
||||
def lookup(self, id):
|
||||
# id='id="gl%d"' % val
|
||||
if id in self.gdict:
|
||||
return self.gdict[id]
|
||||
return None
|
||||
def addGlyph(self, val, path):
|
||||
id='id="gl%d"' % val
|
||||
self.gdict[id] = path
|
||||
|
||||
|
||||
def generateBook(bookDir, raw, fixedimage):
|
||||
# sanity check Topaz file extraction
|
||||
if not os.path.exists(bookDir) :
|
||||
print "Can not find directory with unencrypted book"
|
||||
return 1
|
||||
|
||||
dictFile = os.path.join(bookDir,'dict0000.dat')
|
||||
if not os.path.exists(dictFile) :
|
||||
print "Can not find dict0000.dat file"
|
||||
return 1
|
||||
|
||||
pageDir = os.path.join(bookDir,'page')
|
||||
if not os.path.exists(pageDir) :
|
||||
print "Can not find page directory in unencrypted book"
|
||||
return 1
|
||||
|
||||
imgDir = os.path.join(bookDir,'img')
|
||||
if not os.path.exists(imgDir) :
|
||||
print "Can not find image directory in unencrypted book"
|
||||
return 1
|
||||
|
||||
glyphsDir = os.path.join(bookDir,'glyphs')
|
||||
if not os.path.exists(glyphsDir) :
|
||||
print "Can not find glyphs directory in unencrypted book"
|
||||
return 1
|
||||
|
||||
metaFile = os.path.join(bookDir,'metadata0000.dat')
|
||||
if not os.path.exists(metaFile) :
|
||||
print "Can not find metadata0000.dat in unencrypted book"
|
||||
return 1
|
||||
|
||||
svgDir = os.path.join(bookDir,'svg')
|
||||
if not os.path.exists(svgDir) :
|
||||
os.makedirs(svgDir)
|
||||
|
||||
xmlDir = os.path.join(bookDir,'xml')
|
||||
if not os.path.exists(xmlDir) :
|
||||
os.makedirs(xmlDir)
|
||||
|
||||
otherFile = os.path.join(bookDir,'other0000.dat')
|
||||
if not os.path.exists(otherFile) :
|
||||
print "Can not find other0000.dat in unencrypted book"
|
||||
return 1
|
||||
|
||||
print "Updating to color images if available"
|
||||
spath = os.path.join(bookDir,'color_img')
|
||||
dpath = os.path.join(bookDir,'img')
|
||||
filenames = os.listdir(spath)
|
||||
filenames = sorted(filenames)
|
||||
for filename in filenames:
|
||||
imgname = filename.replace('color','img')
|
||||
sfile = os.path.join(spath,filename)
|
||||
dfile = os.path.join(dpath,imgname)
|
||||
imgdata = file(sfile,'rb').read()
|
||||
file(dfile,'wb').write(imgdata)
|
||||
|
||||
print "Creating cover.jpg"
|
||||
isCover = False
|
||||
cpath = os.path.join(bookDir,'img')
|
||||
cpath = os.path.join(cpath,'img0000.jpg')
|
||||
if os.path.isfile(cpath):
|
||||
cover = file(cpath, 'rb').read()
|
||||
cpath = os.path.join(bookDir,'cover.jpg')
|
||||
file(cpath, 'wb').write(cover)
|
||||
isCover = True
|
||||
|
||||
|
||||
print 'Processing Dictionary'
|
||||
dict = Dictionary(dictFile)
|
||||
|
||||
print 'Processing Meta Data and creating OPF'
|
||||
meta_array = getMetaArray(metaFile)
|
||||
|
||||
xname = os.path.join(xmlDir, 'metadata.xml')
|
||||
metastr = ''
|
||||
for key in meta_array:
|
||||
metastr += '<meta name="' + key + '" content="' + meta_array[key] + '" />\n'
|
||||
file(xname, 'wb').write(metastr)
|
||||
|
||||
print 'Processing StyleSheet'
|
||||
# get some scaling info from metadata to use while processing styles
|
||||
fontsize = '135'
|
||||
if 'fontSize' in meta_array:
|
||||
fontsize = meta_array['fontSize']
|
||||
|
||||
# also get the size of a normal text page
|
||||
spage = '1'
|
||||
if 'firstTextPage' in meta_array:
|
||||
spage = meta_array['firstTextPage']
|
||||
pnum = int(spage)
|
||||
|
||||
# get page height and width from first text page for use in stylesheet scaling
|
||||
pname = 'page%04d.dat' % (pnum + 1)
|
||||
fname = os.path.join(pageDir,pname)
|
||||
flat_xml = convert2xml.fromData(dict, fname)
|
||||
|
||||
(ph, pw) = getPageDim(flat_xml)
|
||||
if (ph == '-1') or (ph == '0') : ph = '11000'
|
||||
if (pw == '-1') or (pw == '0') : pw = '8500'
|
||||
|
||||
# print ' ', 'other0000.dat'
|
||||
xname = os.path.join(bookDir, 'style.css')
|
||||
flat_xml = convert2xml.fromData(dict, otherFile)
|
||||
cssstr , classlst = stylexml2css.convert2CSS(flat_xml, fontsize, ph, pw)
|
||||
file(xname, 'wb').write(cssstr)
|
||||
xname = os.path.join(xmlDir, 'other0000.xml')
|
||||
file(xname, 'wb').write(convert2xml.getXML(dict, otherFile))
|
||||
|
||||
print 'Processing Glyphs'
|
||||
gd = GlyphDict()
|
||||
filenames = os.listdir(glyphsDir)
|
||||
filenames = sorted(filenames)
|
||||
glyfname = os.path.join(svgDir,'glyphs.svg')
|
||||
glyfile = open(glyfname, 'w')
|
||||
glyfile.write('<?xml version="1.0" standalone="no"?>\n')
|
||||
glyfile.write('<!DOCTYPE svg PUBLIC "-//W3C/DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">\n')
|
||||
glyfile.write('<svg width="512" height="512" viewBox="0 0 511 511" xmlns="http://www.w3.org/2000/svg" version="1.1">\n')
|
||||
glyfile.write('<title>Glyphs for %s</title>\n' % meta_array['Title'])
|
||||
glyfile.write('<defs>\n')
|
||||
counter = 0
|
||||
for filename in filenames:
|
||||
# print ' ', filename
|
||||
print '.',
|
||||
fname = os.path.join(glyphsDir,filename)
|
||||
flat_xml = convert2xml.fromData(dict, fname)
|
||||
|
||||
xname = os.path.join(xmlDir, filename.replace('.dat','.xml'))
|
||||
file(xname, 'wb').write(convert2xml.getXML(dict, fname))
|
||||
|
||||
gp = GParser(flat_xml)
|
||||
for i in xrange(0, gp.count):
|
||||
path = gp.getPath(i)
|
||||
maxh, maxw = gp.getGlyphDim(i)
|
||||
fullpath = '<path id="gl%d" d="%s" fill="black" /><!-- width=%d height=%d -->\n' % (counter * 256 + i, path, maxw, maxh)
|
||||
glyfile.write(fullpath)
|
||||
gd.addGlyph(counter * 256 + i, fullpath)
|
||||
counter += 1
|
||||
glyfile.write('</defs>\n')
|
||||
glyfile.write('</svg>\n')
|
||||
glyfile.close()
|
||||
print " "
|
||||
|
||||
# start up the html
|
||||
htmlFileName = "book.html"
|
||||
htmlstr = '<?xml version="1.0" encoding="utf-8"?>\n'
|
||||
htmlstr += '<!DOCTYPE HTML PUBLIC "-//W3C//DTD XHTML 1.1 Strict//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11-strict.dtd">\n'
|
||||
htmlstr += '<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">\n'
|
||||
htmlstr += '<head>\n'
|
||||
htmlstr += '<meta http-equiv="content-type" content="text/html; charset=utf-8"/>\n'
|
||||
htmlstr += '<title>' + meta_array['Title'] + ' by ' + meta_array['Authors'] + '</title>\n'
|
||||
htmlstr += '<meta name="Author" content="' + meta_array['Authors'] + '" />\n'
|
||||
htmlstr += '<meta name="Title" content="' + meta_array['Title'] + '" />\n'
|
||||
htmlstr += '<meta name="ASIN" content="' + meta_array['ASIN'] + '" />\n'
|
||||
htmlstr += '<meta name="GUID" content="' + meta_array['GUID'] + '" />\n'
|
||||
htmlstr += '<link href="style.css" rel="stylesheet" type="text/css" />\n'
|
||||
htmlstr += '</head>\n<body>\n'
|
||||
|
||||
print 'Processing Pages'
|
||||
# Books are at 1440 DPI. This is rendering at twice that size for
|
||||
# readability when rendering to the screen.
|
||||
scaledpi = 1440.0
|
||||
|
||||
svgindex = '<?xml version="1.0" encoding="utf-8"?>\n'
|
||||
svgindex += '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">\n'
|
||||
svgindex += '<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" >'
|
||||
svgindex += '<head>\n'
|
||||
svgindex += '<title>' + meta_array['Title'] + '</title>\n'
|
||||
svgindex += '<meta name="Author" content="' + meta_array['Authors'] + '" />\n'
|
||||
svgindex += '<meta name="Title" content="' + meta_array['Title'] + '" />\n'
|
||||
svgindex += '<meta name="ASIN" content="' + meta_array['ASIN'] + '" />\n'
|
||||
svgindex += '<meta name="GUID" content="' + meta_array['GUID'] + '" />\n'
|
||||
svgindex += '</head>\n'
|
||||
svgindex += '<body>\n'
|
||||
|
||||
filenames = os.listdir(pageDir)
|
||||
filenames = sorted(filenames)
|
||||
numfiles = len(filenames)
|
||||
counter = 0
|
||||
|
||||
for filename in filenames:
|
||||
# print ' ', filename
|
||||
print ".",
|
||||
|
||||
fname = os.path.join(pageDir,filename)
|
||||
flat_xml = convert2xml.fromData(dict, fname)
|
||||
|
||||
xname = os.path.join(xmlDir, filename.replace('.dat','.xml'))
|
||||
file(xname, 'wb').write(convert2xml.getXML(dict, fname))
|
||||
|
||||
# first get the html
|
||||
htmlstr += flatxml2html.convert2HTML(flat_xml, classlst, fname, bookDir, gd, fixedimage)
|
||||
|
||||
# now get the svg image of the page
|
||||
svgxml = flatxml2svg.convert2SVG(gd, flat_xml, counter, numfiles, svgDir, raw, meta_array, scaledpi)
|
||||
|
||||
if (raw) :
|
||||
pfile = open(os.path.join(svgDir,filename.replace('.dat','.svg')), 'w')
|
||||
svgindex += '<a href="svg/page%04d.svg">Page %d</a>\n' % (counter, counter)
|
||||
else :
|
||||
pfile = open(os.path.join(svgDir,'page%04d.xhtml' % counter), 'w')
|
||||
svgindex += '<a href="svg/page%04d.xhtml">Page %d</a>\n' % (counter, counter)
|
||||
|
||||
|
||||
pfile.write(svgxml)
|
||||
pfile.close()
|
||||
|
||||
counter += 1
|
||||
|
||||
print " "
|
||||
|
||||
# finish up the html string and output it
|
||||
htmlstr += '</body>\n</html>\n'
|
||||
file(os.path.join(bookDir, htmlFileName), 'wb').write(htmlstr)
|
||||
|
||||
# finish up the svg index string and output it
|
||||
svgindex += '</body>\n</html>\n'
|
||||
file(os.path.join(bookDir, 'index_svg.xhtml'), 'wb').write(svgindex)
|
||||
|
||||
# build the opf file
|
||||
opfname = os.path.join(bookDir, 'book.opf')
|
||||
opfstr = '<?xml version="1.0" encoding="utf-8"?>\n'
|
||||
opfstr += '<package xmlns="http://www.idpf.org/2007/opf" unique-identifier="guid_id">\n'
|
||||
# adding metadata
|
||||
opfstr += ' <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:opf="http://www.idpf.org/2007/opf">\n'
|
||||
opfstr += ' <dc:identifier opf:scheme="GUID" id="guid_id">' + meta_array['GUID'] + '</dc:identifier>\n'
|
||||
opfstr += ' <dc:identifier opf:scheme="ASIN">' + meta_array['ASIN'] + '</dc:identifier>\n'
|
||||
opfstr += ' <dc:identifier opf:scheme="oASIN">' + meta_array['oASIN'] + '</dc:identifier>\n'
|
||||
opfstr += ' <dc:title>' + meta_array['Title'] + '</dc:title>\n'
|
||||
opfstr += ' <dc:creator opf:role="aut">' + meta_array['Authors'] + '</dc:creator>\n'
|
||||
opfstr += ' <dc:language>en</dc:language>\n'
|
||||
opfstr += ' <dc:date>' + meta_array['UpdateTime'] + '</dc:date>\n'
|
||||
if isCover:
|
||||
opfstr += ' <meta name="cover" content="bookcover"/>\n'
|
||||
opfstr += ' </metadata>\n'
|
||||
opfstr += '<manifest>\n'
|
||||
opfstr += ' <item id="book" href="book.html" media-type="application/xhtml+xml"/>\n'
|
||||
opfstr += ' <item id="stylesheet" href="style.css" media-type="text.css"/>\n'
|
||||
# adding image files to manifest
|
||||
filenames = os.listdir(imgDir)
|
||||
filenames = sorted(filenames)
|
||||
for filename in filenames:
|
||||
imgname, imgext = os.path.splitext(filename)
|
||||
if imgext == '.jpg':
|
||||
imgext = 'jpeg'
|
||||
if imgext == '.svg':
|
||||
imgext = 'svg+xml'
|
||||
opfstr += ' <item id="' + imgname + '" href="img/' + filename + '" media-type="image/' + imgext + '"/>\n'
|
||||
if isCover:
|
||||
opfstr += ' <item id="bookcover" href="cover.jpg" media-type="image/jpeg" />\n'
|
||||
opfstr += '</manifest>\n'
|
||||
# adding spine
|
||||
opfstr += '<spine>\n <itemref idref="book" />\n</spine>\n'
|
||||
if isCover:
|
||||
opfstr += ' <guide>\n'
|
||||
opfstr += ' <reference href="cover.jpg" type="cover" title="Cover"/>\n'
|
||||
opfstr += ' </guide>\n'
|
||||
opfstr += '</package>\n'
|
||||
file(opfname, 'wb').write(opfstr)
|
||||
|
||||
print 'Processing Complete'
|
||||
|
||||
return 0
|
||||
|
||||
def usage():
|
||||
print "genbook.py generates a book from the extract Topaz Files"
|
||||
print "Usage:"
|
||||
print " genbook.py [-r] [-h [--fixed-image] <bookDir> "
|
||||
print " "
|
||||
print "Options:"
|
||||
print " -h : help - print this usage message"
|
||||
print " -r : generate raw svg files (not wrapped in xhtml)"
|
||||
print " --fixed-image : genearate any Fixed Area as an svg image in the html"
|
||||
print " "
|
||||
|
||||
|
||||
def main(argv):
|
||||
bookDir = ''
|
||||
|
||||
if len(argv) == 0:
|
||||
argv = sys.argv
|
||||
|
||||
try:
|
||||
opts, args = getopt.getopt(argv[1:], "rh:",["fixed-image"])
|
||||
|
||||
except getopt.GetoptError, err:
|
||||
print str(err)
|
||||
usage()
|
||||
return 1
|
||||
|
||||
if len(opts) == 0 and len(args) == 0 :
|
||||
usage()
|
||||
return 1
|
||||
|
||||
raw = 0
|
||||
fixedimage = False
|
||||
for o, a in opts:
|
||||
if o =="-h":
|
||||
usage()
|
||||
return 0
|
||||
if o =="-r":
|
||||
raw = 1
|
||||
if o =="--fixed-image":
|
||||
fixedimage = True
|
||||
|
||||
bookDir = args[0]
|
||||
|
||||
rv = generateBook(bookDir, raw, fixedimage)
|
||||
return rv
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
sys.exit(main(''))
|
||||
312
Calibre_Plugins/K4MobiDeDRM_plugin/kgenpids.py
Normal file
312
Calibre_Plugins/K4MobiDeDRM_plugin/kgenpids.py
Normal file
@@ -0,0 +1,312 @@
|
||||
#!/usr/bin/env python
|
||||
|
||||
from __future__ import with_statement
|
||||
import sys
|
||||
import os, csv
|
||||
import binascii
|
||||
import zlib
|
||||
import re
|
||||
from struct import pack, unpack, unpack_from
|
||||
|
||||
class DrmException(Exception):
|
||||
pass
|
||||
|
||||
global kindleDatabase
|
||||
global charMap1
|
||||
global charMap2
|
||||
global charMap3
|
||||
global charMap4
|
||||
|
||||
if sys.platform.startswith('win'):
|
||||
from k4pcutils import openKindleInfo, CryptUnprotectData, GetUserName, GetVolumeSerialNumber, charMap2
|
||||
if sys.platform.startswith('darwin'):
|
||||
from k4mutils import openKindleInfo, CryptUnprotectData, GetUserName, GetVolumeSerialNumber, charMap2
|
||||
|
||||
charMap1 = "n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M"
|
||||
charMap3 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
|
||||
charMap4 = "ABCDEFGHIJKLMNPQRSTUVWXYZ123456789"
|
||||
|
||||
# crypto digestroutines
|
||||
import hashlib
|
||||
|
||||
def MD5(message):
|
||||
ctx = hashlib.md5()
|
||||
ctx.update(message)
|
||||
return ctx.digest()
|
||||
|
||||
def SHA1(message):
|
||||
ctx = hashlib.sha1()
|
||||
ctx.update(message)
|
||||
return ctx.digest()
|
||||
|
||||
|
||||
# Encode the bytes in data with the characters in map
|
||||
def encode(data, map):
|
||||
result = ""
|
||||
for char in data:
|
||||
value = ord(char)
|
||||
Q = (value ^ 0x80) // len(map)
|
||||
R = value % len(map)
|
||||
result += map[Q]
|
||||
result += map[R]
|
||||
return result
|
||||
|
||||
# Hash the bytes in data and then encode the digest with the characters in map
|
||||
def encodeHash(data,map):
|
||||
return encode(MD5(data),map)
|
||||
|
||||
# Decode the string in data with the characters in map. Returns the decoded bytes
|
||||
def decode(data,map):
|
||||
result = ""
|
||||
for i in range (0,len(data)-1,2):
|
||||
high = map.find(data[i])
|
||||
low = map.find(data[i+1])
|
||||
if (high == -1) or (low == -1) :
|
||||
break
|
||||
value = (((high * len(map)) ^ 0x80) & 0xFF) + low
|
||||
result += pack("B",value)
|
||||
return result
|
||||
|
||||
|
||||
# Parse the Kindle.info file and return the records as a list of key-values
|
||||
def parseKindleInfo(kInfoFile):
|
||||
DB = {}
|
||||
infoReader = openKindleInfo(kInfoFile)
|
||||
infoReader.read(1)
|
||||
data = infoReader.read()
|
||||
if sys.platform.startswith('win'):
|
||||
items = data.split('{')
|
||||
else :
|
||||
items = data.split('[')
|
||||
for item in items:
|
||||
splito = item.split(':')
|
||||
DB[splito[0]] =splito[1]
|
||||
return DB
|
||||
|
||||
# Get a record from the Kindle.info file for the key "hashedKey" (already hashed and encoded). Return the decoded and decrypted record
|
||||
def getKindleInfoValueForHash(hashedKey):
|
||||
global kindleDatabase
|
||||
global charMap1
|
||||
global charMap2
|
||||
encryptedValue = decode(kindleDatabase[hashedKey],charMap2)
|
||||
if sys.platform.startswith('win'):
|
||||
return CryptUnprotectData(encryptedValue,"")
|
||||
else:
|
||||
cleartext = CryptUnprotectData(encryptedValue)
|
||||
return decode(cleartext, charMap1)
|
||||
|
||||
# Get a record from the Kindle.info file for the string in "key" (plaintext). Return the decoded and decrypted record
|
||||
def getKindleInfoValueForKey(key):
|
||||
global charMap2
|
||||
return getKindleInfoValueForHash(encodeHash(key,charMap2))
|
||||
|
||||
# Find if the original string for a hashed/encoded string is known. If so return the original string othwise return an empty string.
|
||||
def findNameForHash(hash):
|
||||
global charMap2
|
||||
names = ["kindle.account.tokens","kindle.cookie.item","eulaVersionAccepted","login_date","kindle.token.item","login","kindle.key.item","kindle.name.info","kindle.device.info", "MazamaRandomNumber"]
|
||||
result = ""
|
||||
for name in names:
|
||||
if hash == encodeHash(name, charMap2):
|
||||
result = name
|
||||
break
|
||||
return result
|
||||
|
||||
# Print all the records from the kindle.info file (option -i)
|
||||
def printKindleInfo():
|
||||
for record in kindleDatabase:
|
||||
name = findNameForHash(record)
|
||||
if name != "" :
|
||||
print (name)
|
||||
print ("--------------------------")
|
||||
else :
|
||||
print ("Unknown Record")
|
||||
print getKindleInfoValueForHash(record)
|
||||
print "\n"
|
||||
|
||||
#
|
||||
# PID generation routines
|
||||
#
|
||||
|
||||
# Returns two bit at offset from a bit field
|
||||
def getTwoBitsFromBitField(bitField,offset):
|
||||
byteNumber = offset // 4
|
||||
bitPosition = 6 - 2*(offset % 4)
|
||||
return ord(bitField[byteNumber]) >> bitPosition & 3
|
||||
|
||||
# Returns the six bits at offset from a bit field
|
||||
def getSixBitsFromBitField(bitField,offset):
|
||||
offset *= 3
|
||||
value = (getTwoBitsFromBitField(bitField,offset) <<4) + (getTwoBitsFromBitField(bitField,offset+1) << 2) +getTwoBitsFromBitField(bitField,offset+2)
|
||||
return value
|
||||
|
||||
# 8 bits to six bits encoding from hash to generate PID string
|
||||
def encodePID(hash):
|
||||
global charMap3
|
||||
PID = ""
|
||||
for position in range (0,8):
|
||||
PID += charMap3[getSixBitsFromBitField(hash,position)]
|
||||
return PID
|
||||
|
||||
# Encryption table used to generate the device PID
|
||||
def generatePidEncryptionTable() :
|
||||
table = []
|
||||
for counter1 in range (0,0x100):
|
||||
value = counter1
|
||||
for counter2 in range (0,8):
|
||||
if (value & 1 == 0) :
|
||||
value = value >> 1
|
||||
else :
|
||||
value = value >> 1
|
||||
value = value ^ 0xEDB88320
|
||||
table.append(value)
|
||||
return table
|
||||
|
||||
# Seed value used to generate the device PID
|
||||
def generatePidSeed(table,dsn) :
|
||||
value = 0
|
||||
for counter in range (0,4) :
|
||||
index = (ord(dsn[counter]) ^ value) &0xFF
|
||||
value = (value >> 8) ^ table[index]
|
||||
return value
|
||||
|
||||
# Generate the device PID
|
||||
def generateDevicePID(table,dsn,nbRoll):
|
||||
global charMap4
|
||||
seed = generatePidSeed(table,dsn)
|
||||
pidAscii = ""
|
||||
pid = [(seed >>24) &0xFF,(seed >> 16) &0xff,(seed >> 8) &0xFF,(seed) & 0xFF,(seed>>24) & 0xFF,(seed >> 16) &0xff,(seed >> 8) &0xFF,(seed) & 0xFF]
|
||||
index = 0
|
||||
for counter in range (0,nbRoll):
|
||||
pid[index] = pid[index] ^ ord(dsn[counter])
|
||||
index = (index+1) %8
|
||||
for counter in range (0,8):
|
||||
index = ((((pid[counter] >>5) & 3) ^ pid[counter]) & 0x1f) + (pid[counter] >> 7)
|
||||
pidAscii += charMap4[index]
|
||||
return pidAscii
|
||||
|
||||
def crc32(s):
|
||||
return (~binascii.crc32(s,-1))&0xFFFFFFFF
|
||||
|
||||
# convert from 8 digit PID to 10 digit PID with checksum
|
||||
def checksumPid(s):
|
||||
global charMap4
|
||||
crc = crc32(s)
|
||||
crc = crc ^ (crc >> 16)
|
||||
res = s
|
||||
l = len(charMap4)
|
||||
for i in (0,1):
|
||||
b = crc & 0xff
|
||||
pos = (b // l) ^ (b % l)
|
||||
res += charMap4[pos%l]
|
||||
crc >>= 8
|
||||
return res
|
||||
|
||||
|
||||
# old kindle serial number to fixed pid
|
||||
def pidFromSerial(s, l):
|
||||
global charMap4
|
||||
crc = crc32(s)
|
||||
arr1 = [0]*l
|
||||
for i in xrange(len(s)):
|
||||
arr1[i%l] ^= ord(s[i])
|
||||
crc_bytes = [crc >> 24 & 0xff, crc >> 16 & 0xff, crc >> 8 & 0xff, crc & 0xff]
|
||||
for i in xrange(l):
|
||||
arr1[i] ^= crc_bytes[i&3]
|
||||
pid = ""
|
||||
for i in xrange(l):
|
||||
b = arr1[i] & 0xff
|
||||
pid+=charMap4[(b >> 7) + ((b >> 5 & 3) ^ (b & 0x1f))]
|
||||
return pid
|
||||
|
||||
|
||||
# Parse the EXTH header records and use the Kindle serial number to calculate the book pid.
|
||||
def getKindlePid(pidlst, rec209, token, serialnum):
|
||||
|
||||
if rec209 != None:
|
||||
# Compute book PID
|
||||
pidHash = SHA1(serialnum+rec209+token)
|
||||
bookPID = encodePID(pidHash)
|
||||
bookPID = checksumPid(bookPID)
|
||||
pidlst.append(bookPID)
|
||||
|
||||
# compute fixed pid for old pre 2.5 firmware update pid as well
|
||||
bookPID = pidFromSerial(serialnum, 7) + "*"
|
||||
bookPID = checksumPid(bookPID)
|
||||
pidlst.append(bookPID)
|
||||
|
||||
return pidlst
|
||||
|
||||
|
||||
# Parse the EXTH header records and parse the Kindleinfo
|
||||
# file to calculate the book pid.
|
||||
|
||||
def getK4Pids(pidlst, rec209, token, kInfoFile=None):
|
||||
global kindleDatabase
|
||||
global charMap1
|
||||
kindleDatabase = None
|
||||
try:
|
||||
kindleDatabase = parseKindleInfo(kInfoFile)
|
||||
except Exception, message:
|
||||
print(message)
|
||||
pass
|
||||
|
||||
if kindleDatabase == None :
|
||||
return pidlst
|
||||
|
||||
# Get the Mazama Random number
|
||||
MazamaRandomNumber = getKindleInfoValueForKey("MazamaRandomNumber")
|
||||
|
||||
# Get the HDD serial
|
||||
encodedSystemVolumeSerialNumber = encodeHash(GetVolumeSerialNumber(),charMap1)
|
||||
|
||||
# Get the current user name
|
||||
encodedUsername = encodeHash(GetUserName(),charMap1)
|
||||
|
||||
# concat, hash and encode to calculate the DSN
|
||||
DSN = encode(SHA1(MazamaRandomNumber+encodedSystemVolumeSerialNumber+encodedUsername),charMap1)
|
||||
|
||||
# Compute the device PID (for which I can tell, is used for nothing).
|
||||
table = generatePidEncryptionTable()
|
||||
devicePID = generateDevicePID(table,DSN,4)
|
||||
devicePID = checksumPid(devicePID)
|
||||
pidlst.append(devicePID)
|
||||
|
||||
# Compute book PID
|
||||
if rec209 == None:
|
||||
print "\nNo EXTH record type 209 - Perhaps not a K4 file?"
|
||||
return pidlst
|
||||
|
||||
# Get the kindle account token
|
||||
kindleAccountToken = getKindleInfoValueForKey("kindle.account.tokens")
|
||||
|
||||
# book pid
|
||||
pidHash = SHA1(DSN+kindleAccountToken+rec209+token)
|
||||
bookPID = encodePID(pidHash)
|
||||
bookPID = checksumPid(bookPID)
|
||||
pidlst.append(bookPID)
|
||||
|
||||
# variant 1
|
||||
pidHash = SHA1(kindleAccountToken+rec209+token)
|
||||
bookPID = encodePID(pidHash)
|
||||
bookPID = checksumPid(bookPID)
|
||||
pidlst.append(bookPID)
|
||||
|
||||
# variant 2
|
||||
pidHash = SHA1(DSN+rec209+token)
|
||||
bookPID = encodePID(pidHash)
|
||||
bookPID = checksumPid(bookPID)
|
||||
pidlst.append(bookPID)
|
||||
|
||||
return pidlst
|
||||
|
||||
def getPidList(md1, md2, k4, pids, serials, kInfoFiles):
|
||||
pidlst = []
|
||||
if k4:
|
||||
pidlst = getK4Pids(pidlst, md1, md2)
|
||||
for infoFile in kInfoFiles:
|
||||
pidlst = getK4Pids(pidlst, md1, md2, infoFile)
|
||||
for serialnum in serials:
|
||||
pidlst = getKindlePid(pidlst, md1, md2, serialnum)
|
||||
for pid in pids:
|
||||
pidlst.append(pid)
|
||||
return pidlst
|
||||
243
Calibre_Plugins/K4MobiDeDRM_plugin/stylexml2css.py
Normal file
243
Calibre_Plugins/K4MobiDeDRM_plugin/stylexml2css.py
Normal file
@@ -0,0 +1,243 @@
|
||||
#! /usr/bin/python
|
||||
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
|
||||
# For use with Topaz Scripts Version 2.6
|
||||
|
||||
import csv
|
||||
import sys
|
||||
import os
|
||||
import getopt
|
||||
from struct import pack
|
||||
from struct import unpack
|
||||
|
||||
|
||||
class DocParser(object):
|
||||
def __init__(self, flatxml, fontsize, ph, pw):
|
||||
self.flatdoc = flatxml.split('\n')
|
||||
self.fontsize = int(fontsize)
|
||||
self.ph = int(ph) * 1.0
|
||||
self.pw = int(pw) * 1.0
|
||||
|
||||
stags = {
|
||||
'paragraph' : 'p',
|
||||
'graphic' : '.graphic'
|
||||
}
|
||||
|
||||
attr_val_map = {
|
||||
'hang' : 'text-indent: ',
|
||||
'indent' : 'text-indent: ',
|
||||
'line-space' : 'line-height: ',
|
||||
'margin-bottom' : 'margin-bottom: ',
|
||||
'margin-left' : 'margin-left: ',
|
||||
'margin-right' : 'margin-right: ',
|
||||
'margin-top' : 'margin-top: ',
|
||||
'space-after' : 'padding-bottom: ',
|
||||
}
|
||||
|
||||
attr_str_map = {
|
||||
'align-center' : 'text-align: center; margin-left: auto; margin-right: auto;',
|
||||
'align-left' : 'text-align: left;',
|
||||
'align-right' : 'text-align: right;',
|
||||
'align-justify' : 'text-align: justify;',
|
||||
'display-inline' : 'display: inline;',
|
||||
'pos-left' : 'text-align: left;',
|
||||
'pos-right' : 'text-align: right;',
|
||||
'pos-center' : 'text-align: center; margin-left: auto; margin-right: auto;',
|
||||
}
|
||||
|
||||
|
||||
# find tag if within pos to end inclusive
|
||||
def findinDoc(self, tagpath, pos, end) :
|
||||
result = None
|
||||
docList = self.flatdoc
|
||||
cnt = len(docList)
|
||||
if end == -1 :
|
||||
end = cnt
|
||||
else:
|
||||
end = min(cnt,end)
|
||||
foundat = -1
|
||||
for j in xrange(pos, end):
|
||||
item = docList[j]
|
||||
if item.find('=') >= 0:
|
||||
(name, argres) = item.split('=',1)
|
||||
else :
|
||||
name = item
|
||||
argres = ''
|
||||
if name.endswith(tagpath) :
|
||||
result = argres
|
||||
foundat = j
|
||||
break
|
||||
return foundat, result
|
||||
|
||||
|
||||
# return list of start positions for the tagpath
|
||||
def posinDoc(self, tagpath):
|
||||
startpos = []
|
||||
pos = 0
|
||||
res = ""
|
||||
while res != None :
|
||||
(foundpos, res) = self.findinDoc(tagpath, pos, -1)
|
||||
if res != None :
|
||||
startpos.append(foundpos)
|
||||
pos = foundpos + 1
|
||||
return startpos
|
||||
|
||||
|
||||
def process(self):
|
||||
|
||||
classlst = ''
|
||||
csspage = '.cl-center { text-align: center; margin-left: auto; margin-right: auto; }\n'
|
||||
csspage += '.cl-right { text-align: right; }\n'
|
||||
csspage += '.cl-left { text-align: left; }\n'
|
||||
csspage += '.cl-justify { text-align: justify; }\n'
|
||||
|
||||
# generate a list of each <style> starting point in the stylesheet
|
||||
styleList= self.posinDoc('book.stylesheet.style')
|
||||
stylecnt = len(styleList)
|
||||
styleList.append(-1)
|
||||
|
||||
# process each style converting what you can
|
||||
|
||||
for j in xrange(stylecnt):
|
||||
start = styleList[j]
|
||||
end = styleList[j+1]
|
||||
|
||||
(pos, tag) = self.findinDoc('style._tag',start,end)
|
||||
if tag == None :
|
||||
(pos, tag) = self.findinDoc('style.type',start,end)
|
||||
|
||||
# Is this something we know how to convert to css
|
||||
if tag in self.stags :
|
||||
|
||||
# get the style class
|
||||
(pos, sclass) = self.findinDoc('style.class',start,end)
|
||||
if sclass != None:
|
||||
sclass = sclass.replace(' ','-')
|
||||
sclass = '.cl-' + sclass.lower()
|
||||
else :
|
||||
sclass = ''
|
||||
|
||||
# check for any "after class" specifiers
|
||||
(pos, aftclass) = self.findinDoc('style._after_class',start,end)
|
||||
if aftclass != None:
|
||||
aftclass = aftclass.replace(' ','-')
|
||||
aftclass = '.cl-' + aftclass.lower()
|
||||
else :
|
||||
aftclass = ''
|
||||
|
||||
cssargs = {}
|
||||
|
||||
while True :
|
||||
|
||||
(pos1, attr) = self.findinDoc('style.rule.attr', start, end)
|
||||
(pos2, val) = self.findinDoc('style.rule.value', start, end)
|
||||
|
||||
if attr == None : break
|
||||
|
||||
if (attr == 'display') or (attr == 'pos') or (attr == 'align'):
|
||||
# handle text based attributess
|
||||
attr = attr + '-' + val
|
||||
if attr in self.attr_str_map :
|
||||
cssargs[attr] = (self.attr_str_map[attr], '')
|
||||
else :
|
||||
# handle value based attributes
|
||||
if attr in self.attr_val_map :
|
||||
name = self.attr_val_map[attr]
|
||||
if attr in ('margin-bottom', 'margin-top', 'space-after') :
|
||||
scale = self.ph
|
||||
elif attr in ('margin-right', 'indent', 'margin-left', 'hang') :
|
||||
scale = self.pw
|
||||
elif attr == 'line-space':
|
||||
scale = self.fontsize * 2.0
|
||||
|
||||
if not ((attr == 'hang') and (int(val) == 0)) :
|
||||
pv = float(val)/scale
|
||||
cssargs[attr] = (self.attr_val_map[attr], pv)
|
||||
keep = True
|
||||
|
||||
start = max(pos1, pos2) + 1
|
||||
|
||||
# disable all of the after class tags until I figure out how to handle them
|
||||
if aftclass != "" : keep = False
|
||||
|
||||
if keep :
|
||||
# make sure line-space does not go below 100% or above 300% since
|
||||
# it can be wacky in some styles
|
||||
if 'line-space' in cssargs:
|
||||
seg = cssargs['line-space'][0]
|
||||
val = cssargs['line-space'][1]
|
||||
if val < 1.0: val = 1.0
|
||||
if val > 3.0: val = 3.0
|
||||
del cssargs['line-space']
|
||||
cssargs['line-space'] = (self.attr_val_map['line-space'], val)
|
||||
|
||||
|
||||
# handle modifications for css style hanging indents
|
||||
if 'hang' in cssargs:
|
||||
hseg = cssargs['hang'][0]
|
||||
hval = cssargs['hang'][1]
|
||||
del cssargs['hang']
|
||||
cssargs['hang'] = (self.attr_val_map['hang'], -hval)
|
||||
mval = 0
|
||||
mseg = 'margin-left: '
|
||||
mval = hval
|
||||
if 'margin-left' in cssargs:
|
||||
mseg = cssargs['margin-left'][0]
|
||||
mval = cssargs['margin-left'][1]
|
||||
if mval < 0: mval = 0
|
||||
mval = hval + mval
|
||||
cssargs['margin-left'] = (mseg, mval)
|
||||
if 'indent' in cssargs:
|
||||
del cssargs['indent']
|
||||
|
||||
cssline = sclass + ' { '
|
||||
for key in iter(cssargs):
|
||||
mseg = cssargs[key][0]
|
||||
mval = cssargs[key][1]
|
||||
if mval == '':
|
||||
cssline += mseg + ' '
|
||||
else :
|
||||
aseg = mseg + '%.1f%%;' % (mval * 100.0)
|
||||
cssline += aseg + ' '
|
||||
|
||||
cssline += '}'
|
||||
|
||||
if sclass != '' :
|
||||
classlst += sclass + '\n'
|
||||
|
||||
# handle special case of paragraph class used inside chapter heading
|
||||
# and non-chapter headings
|
||||
if sclass != '' :
|
||||
ctype = sclass[4:7]
|
||||
if ctype == 'ch1' :
|
||||
csspage += 'h1' + cssline + '\n'
|
||||
if ctype == 'ch2' :
|
||||
csspage += 'h2' + cssline + '\n'
|
||||
if ctype == 'ch3' :
|
||||
csspage += 'h3' + cssline + '\n'
|
||||
if ctype == 'h1-' :
|
||||
csspage += 'h4' + cssline + '\n'
|
||||
if ctype == 'h2-' :
|
||||
csspage += 'h5' + cssline + '\n'
|
||||
if ctype == 'h3_' :
|
||||
csspage += 'h6' + cssline + '\n'
|
||||
|
||||
if cssline != ' { }':
|
||||
csspage += self.stags[tag] + cssline + '\n'
|
||||
|
||||
|
||||
return csspage, classlst
|
||||
|
||||
|
||||
|
||||
def convert2CSS(flatxml, fontsize, ph, pw):
|
||||
|
||||
print ' ', 'Using font size:',fontsize
|
||||
print ' ', 'Using page height:', ph
|
||||
print ' ', 'Using page width:', pw
|
||||
|
||||
# create a document parser
|
||||
dp = DocParser(flatxml, fontsize, ph, pw)
|
||||
|
||||
csspage = dp.process()
|
||||
|
||||
return csspage
|
||||
434
Calibre_Plugins/K4MobiDeDRM_plugin/topazextract.py
Normal file
434
Calibre_Plugins/K4MobiDeDRM_plugin/topazextract.py
Normal file
@@ -0,0 +1,434 @@
|
||||
#!/usr/bin/env python
|
||||
|
||||
class Unbuffered:
|
||||
def __init__(self, stream):
|
||||
self.stream = stream
|
||||
def write(self, data):
|
||||
self.stream.write(data)
|
||||
self.stream.flush()
|
||||
def __getattr__(self, attr):
|
||||
return getattr(self.stream, attr)
|
||||
|
||||
import sys
|
||||
sys.stdout=Unbuffered(sys.stdout)
|
||||
import os, csv, getopt
|
||||
import zlib, zipfile, tempfile, shutil
|
||||
from struct import pack
|
||||
from struct import unpack
|
||||
|
||||
class TpzDRMError(Exception):
|
||||
pass
|
||||
|
||||
# local support routines
|
||||
import kgenpids
|
||||
import genbook
|
||||
#
|
||||
# Utility routines
|
||||
#
|
||||
|
||||
# Get a 7 bit encoded number from file
|
||||
def bookReadEncodedNumber(fo):
|
||||
flag = False
|
||||
data = ord(fo.read(1))
|
||||
if data == 0xFF:
|
||||
flag = True
|
||||
data = ord(fo.read(1))
|
||||
if data >= 0x80:
|
||||
datax = (data & 0x7F)
|
||||
while data >= 0x80 :
|
||||
data = ord(fo.read(1))
|
||||
datax = (datax <<7) + (data & 0x7F)
|
||||
data = datax
|
||||
if flag:
|
||||
data = -data
|
||||
return data
|
||||
|
||||
# Get a length prefixed string from file
|
||||
def bookReadString(fo):
|
||||
stringLength = bookReadEncodedNumber(fo)
|
||||
return unpack(str(stringLength)+"s",fo.read(stringLength))[0]
|
||||
|
||||
#
|
||||
# crypto routines
|
||||
#
|
||||
|
||||
# Context initialisation for the Topaz Crypto
|
||||
def topazCryptoInit(key):
|
||||
ctx1 = 0x0CAFFE19E
|
||||
for keyChar in key:
|
||||
keyByte = ord(keyChar)
|
||||
ctx2 = ctx1
|
||||
ctx1 = ((((ctx1 >>2) * (ctx1 >>7))&0xFFFFFFFF) ^ (keyByte * keyByte * 0x0F902007)& 0xFFFFFFFF )
|
||||
return [ctx1,ctx2]
|
||||
|
||||
# decrypt data with the context prepared by topazCryptoInit()
|
||||
def topazCryptoDecrypt(data, ctx):
|
||||
ctx1 = ctx[0]
|
||||
ctx2 = ctx[1]
|
||||
plainText = ""
|
||||
for dataChar in data:
|
||||
dataByte = ord(dataChar)
|
||||
m = (dataByte ^ ((ctx1 >> 3) &0xFF) ^ ((ctx2<<3) & 0xFF)) &0xFF
|
||||
ctx2 = ctx1
|
||||
ctx1 = (((ctx1 >> 2) * (ctx1 >> 7)) &0xFFFFFFFF) ^((m * m * 0x0F902007) &0xFFFFFFFF)
|
||||
plainText += chr(m)
|
||||
return plainText
|
||||
|
||||
# Decrypt data with the PID
|
||||
def decryptRecord(data,PID):
|
||||
ctx = topazCryptoInit(PID)
|
||||
return topazCryptoDecrypt(data, ctx)
|
||||
|
||||
# Try to decrypt a dkey record (contains the bookPID)
|
||||
def decryptDkeyRecord(data,PID):
|
||||
record = decryptRecord(data,PID)
|
||||
fields = unpack("3sB8sB8s3s",record)
|
||||
if fields[0] != "PID" or fields[5] != "pid" :
|
||||
raise TpzDRMError("Didn't find PID magic numbers in record")
|
||||
elif fields[1] != 8 or fields[3] != 8 :
|
||||
raise TpzDRMError("Record didn't contain correct length fields")
|
||||
elif fields[2] != PID :
|
||||
raise TpzDRMError("Record didn't contain PID")
|
||||
return fields[4]
|
||||
|
||||
# Decrypt all dkey records (contain the book PID)
|
||||
def decryptDkeyRecords(data,PID):
|
||||
nbKeyRecords = ord(data[0])
|
||||
records = []
|
||||
data = data[1:]
|
||||
for i in range (0,nbKeyRecords):
|
||||
length = ord(data[0])
|
||||
try:
|
||||
key = decryptDkeyRecord(data[1:length+1],PID)
|
||||
records.append(key)
|
||||
except TpzDRMError:
|
||||
pass
|
||||
data = data[1+length:]
|
||||
if len(records) == 0:
|
||||
raise TpzDRMError("BookKey Not Found")
|
||||
return records
|
||||
|
||||
|
||||
class TopazBook:
|
||||
def __init__(self, filename, outdir):
|
||||
self.fo = file(filename, 'rb')
|
||||
self.outdir = outdir
|
||||
self.bookPayloadOffset = 0
|
||||
self.bookHeaderRecords = {}
|
||||
self.bookMetadata = {}
|
||||
self.bookKey = None
|
||||
magic = unpack("4s",self.fo.read(4))[0]
|
||||
if magic != 'TPZ0':
|
||||
raise TpzDRMError("Parse Error : Invalid Header, not a Topaz file")
|
||||
self.parseTopazHeaders()
|
||||
self.parseMetadata()
|
||||
|
||||
def parseTopazHeaders(self):
|
||||
def bookReadHeaderRecordData():
|
||||
# Read and return the data of one header record at the current book file position
|
||||
# [[offset,decompressedLength,compressedLength],...]
|
||||
nbValues = bookReadEncodedNumber(self.fo)
|
||||
values = []
|
||||
for i in range (0,nbValues):
|
||||
values.append([bookReadEncodedNumber(self.fo),bookReadEncodedNumber(self.fo),bookReadEncodedNumber(self.fo)])
|
||||
return values
|
||||
def parseTopazHeaderRecord():
|
||||
# Read and parse one header record at the current book file position and return the associated data
|
||||
# [[offset,decompressedLength,compressedLength],...]
|
||||
if ord(self.fo.read(1)) != 0x63:
|
||||
raise TpzDRMError("Parse Error : Invalid Header")
|
||||
tag = bookReadString(self.fo)
|
||||
record = bookReadHeaderRecordData()
|
||||
return [tag,record]
|
||||
nbRecords = bookReadEncodedNumber(self.fo)
|
||||
for i in range (0,nbRecords):
|
||||
result = parseTopazHeaderRecord()
|
||||
# print result[0], result[1]
|
||||
self.bookHeaderRecords[result[0]] = result[1]
|
||||
if ord(self.fo.read(1)) != 0x64 :
|
||||
raise TpzDRMError("Parse Error : Invalid Header")
|
||||
self.bookPayloadOffset = self.fo.tell()
|
||||
|
||||
def parseMetadata(self):
|
||||
# Parse the metadata record from the book payload and return a list of [key,values]
|
||||
self.fo.seek(self.bookPayloadOffset + self.bookHeaderRecords["metadata"][0][0])
|
||||
tag = bookReadString(self.fo)
|
||||
if tag != "metadata" :
|
||||
raise TpzDRMError("Parse Error : Record Names Don't Match")
|
||||
flags = ord(self.fo.read(1))
|
||||
nbRecords = ord(self.fo.read(1))
|
||||
for i in range (0,nbRecords) :
|
||||
record = [bookReadString(self.fo), bookReadString(self.fo)]
|
||||
self.bookMetadata[record[0]] = record[1]
|
||||
return self.bookMetadata
|
||||
|
||||
def getPIDMetaInfo(self):
|
||||
keysRecord = None
|
||||
KeysRecordRecord = None
|
||||
if 'keys' in self.bookMetadata:
|
||||
keysRecord = self.bookMetadata['keys']
|
||||
keysRecordRecord = self.bookMetadata[keysRecord]
|
||||
return keysRecord, keysRecordRecord
|
||||
|
||||
def getBookTitle(self):
|
||||
title = ''
|
||||
if 'Title' in self.bookMetadata:
|
||||
title = self.bookMetadata['Title']
|
||||
return title
|
||||
|
||||
def setBookKey(self, key):
|
||||
self.bookKey = key
|
||||
|
||||
def getBookPayloadRecord(self, name, index):
|
||||
# Get a record in the book payload, given its name and index.
|
||||
# decrypted and decompressed if necessary
|
||||
encrypted = False
|
||||
compressed = False
|
||||
try:
|
||||
recordOffset = self.bookHeaderRecords[name][index][0]
|
||||
except:
|
||||
raise TpzDRMError("Parse Error : Invalid Record, record not found")
|
||||
|
||||
self.fo.seek(self.bookPayloadOffset + recordOffset)
|
||||
|
||||
tag = bookReadString(self.fo)
|
||||
if tag != name :
|
||||
raise TpzDRMError("Parse Error : Invalid Record, record name doesn't match")
|
||||
|
||||
recordIndex = bookReadEncodedNumber(self.fo)
|
||||
if recordIndex < 0 :
|
||||
encrypted = True
|
||||
recordIndex = -recordIndex -1
|
||||
|
||||
if recordIndex != index :
|
||||
raise TpzDRMError("Parse Error : Invalid Record, index doesn't match")
|
||||
|
||||
if (self.bookHeaderRecords[name][index][2] > 0):
|
||||
compressed = True
|
||||
record = self.fo.read(self.bookHeaderRecords[name][index][2])
|
||||
else:
|
||||
record = self.fo.read(self.bookHeaderRecords[name][index][1])
|
||||
|
||||
if encrypted:
|
||||
if self.bookKey:
|
||||
ctx = topazCryptoInit(self.bookKey)
|
||||
record = topazCryptoDecrypt(record,ctx)
|
||||
else :
|
||||
raise TpzDRMError("Error: Attempt to decrypt without bookKey")
|
||||
|
||||
if compressed:
|
||||
record = zlib.decompress(record)
|
||||
|
||||
return record
|
||||
|
||||
def processBook(self, pidlst):
|
||||
raw = 0
|
||||
fixedimage=True
|
||||
try:
|
||||
keydata = self.getBookPayloadRecord('dkey', 0)
|
||||
except TpzDRMError, e:
|
||||
print "no dkey record found, book may not be encrypted"
|
||||
print "attempting to extrct files without a book key"
|
||||
self.createBookDirectory()
|
||||
self.extractFiles()
|
||||
print "Successfully Extracted Topaz contents"
|
||||
rv = genbook.generateBook(self.outdir, raw, fixedimage)
|
||||
if rv == 0:
|
||||
print "\nBook Successfully generated"
|
||||
return rv
|
||||
|
||||
# try each pid to decode the file
|
||||
bookKey = None
|
||||
for pid in pidlst:
|
||||
# use 8 digit pids here
|
||||
pid = pid[0:8]
|
||||
print "\nTrying: ", pid
|
||||
bookKeys = []
|
||||
data = keydata
|
||||
try:
|
||||
bookKeys+=decryptDkeyRecords(data,pid)
|
||||
except TpzDRMError, e:
|
||||
pass
|
||||
else:
|
||||
bookKey = bookKeys[0]
|
||||
print "Book Key Found!"
|
||||
break
|
||||
|
||||
if not bookKey:
|
||||
raise TpzDRMError('Decryption Unsucessful; No valid pid found')
|
||||
|
||||
self.setBookKey(bookKey)
|
||||
self.createBookDirectory()
|
||||
self.extractFiles()
|
||||
print "Successfully Extracted Topaz contents"
|
||||
rv = genbook.generateBook(self.outdir, raw, fixedimage)
|
||||
if rv == 0:
|
||||
print "\nBook Successfully generated"
|
||||
return rv
|
||||
|
||||
def createBookDirectory(self):
|
||||
outdir = self.outdir
|
||||
# create output directory structure
|
||||
if not os.path.exists(outdir):
|
||||
os.makedirs(outdir)
|
||||
destdir = os.path.join(outdir,'img')
|
||||
if not os.path.exists(destdir):
|
||||
os.makedirs(destdir)
|
||||
destdir = os.path.join(outdir,'color_img')
|
||||
if not os.path.exists(destdir):
|
||||
os.makedirs(destdir)
|
||||
destdir = os.path.join(outdir,'page')
|
||||
if not os.path.exists(destdir):
|
||||
os.makedirs(destdir)
|
||||
destdir = os.path.join(outdir,'glyphs')
|
||||
if not os.path.exists(destdir):
|
||||
os.makedirs(destdir)
|
||||
|
||||
def extractFiles(self):
|
||||
outdir = self.outdir
|
||||
for headerRecord in self.bookHeaderRecords:
|
||||
name = headerRecord
|
||||
if name != "dkey" :
|
||||
ext = '.dat'
|
||||
if name == 'img' : ext = '.jpg'
|
||||
if name == 'color' : ext = '.jpg'
|
||||
print "\nProcessing Section: %s " % name
|
||||
for index in range (0,len(self.bookHeaderRecords[name])) :
|
||||
fnum = "%04d" % index
|
||||
fname = name + fnum + ext
|
||||
destdir = outdir
|
||||
if name == 'img':
|
||||
destdir = os.path.join(outdir,'img')
|
||||
if name == 'color':
|
||||
destdir = os.path.join(outdir,'color_img')
|
||||
if name == 'page':
|
||||
destdir = os.path.join(outdir,'page')
|
||||
if name == 'glyphs':
|
||||
destdir = os.path.join(outdir,'glyphs')
|
||||
outputFile = os.path.join(destdir,fname)
|
||||
print ".",
|
||||
record = self.getBookPayloadRecord(name,index)
|
||||
if record != '':
|
||||
file(outputFile, 'wb').write(record)
|
||||
print " "
|
||||
|
||||
|
||||
def zipUpDir(myzip, tempdir,localname):
|
||||
currentdir = tempdir
|
||||
if localname != "":
|
||||
currentdir = os.path.join(currentdir,localname)
|
||||
list = os.listdir(currentdir)
|
||||
for file in list:
|
||||
afilename = file
|
||||
localfilePath = os.path.join(localname, afilename)
|
||||
realfilePath = os.path.join(currentdir,file)
|
||||
if os.path.isfile(realfilePath):
|
||||
myzip.write(realfilePath, localfilePath)
|
||||
elif os.path.isdir(realfilePath):
|
||||
zipUpDir(myzip, tempdir, localfilePath)
|
||||
|
||||
|
||||
def usage(progname):
|
||||
print "Removes DRM protection from Topaz ebooks and extract the contents"
|
||||
print "Usage:"
|
||||
print " %s [-k <kindle.info>] [-p <pidnums>] [-s <kindleSerialNumbers>] <infile> <outdir> " % progname
|
||||
|
||||
|
||||
# Main
|
||||
def main(argv=sys.argv):
|
||||
progname = os.path.basename(argv[0])
|
||||
k4 = False
|
||||
pids = []
|
||||
serials = []
|
||||
kInfoFiles = []
|
||||
|
||||
try:
|
||||
opts, args = getopt.getopt(sys.argv[1:], "k:p:s:")
|
||||
except getopt.GetoptError, err:
|
||||
print str(err)
|
||||
usage(progname)
|
||||
return 1
|
||||
if len(args)<2:
|
||||
usage(progname)
|
||||
return 1
|
||||
|
||||
for o, a in opts:
|
||||
if o == "-k":
|
||||
if a == None :
|
||||
print "Invalid parameter for -k"
|
||||
return 1
|
||||
kInfoFiles.append(a)
|
||||
if o == "-p":
|
||||
if a == None :
|
||||
print "Invalid parameter for -p"
|
||||
return 1
|
||||
pids = a.split(',')
|
||||
if o == "-s":
|
||||
if a == None :
|
||||
print "Invalid parameter for -s"
|
||||
return 1
|
||||
serials = a.split(',')
|
||||
k4 = True
|
||||
|
||||
infile = args[0]
|
||||
outdir = args[1]
|
||||
|
||||
if not os.path.isfile(infile):
|
||||
print "Input File Does Not Exist"
|
||||
return 1
|
||||
|
||||
bookname = os.path.splitext(os.path.basename(infile))[0]
|
||||
tempdir = tempfile.mkdtemp()
|
||||
|
||||
tb = TopazBook(infile, tempdir)
|
||||
title = tb.getBookTitle()
|
||||
print "Processing Book: ", title
|
||||
keysRecord, keysRecordRecord = tb.getPIDMetaInfo()
|
||||
pidlst = kgenpids.getPidList(keysRecord, keysRecordRecord, k4, pids, serials, kInfoFiles)
|
||||
|
||||
try:
|
||||
tb.processBook(pidlst)
|
||||
except TpzDRMError, e:
|
||||
print str(e)
|
||||
print " Creating DeBug Full Zip Archive of Book"
|
||||
zipname = os.path.join(outdir, bookname + '_debug' + '.zip')
|
||||
myzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
|
||||
zipUpDir(myzip, tempdir, '')
|
||||
myzip.close()
|
||||
return 1
|
||||
|
||||
print " Creating HTML ZIP Archive"
|
||||
zipname = os.path.join(outdir, bookname + '_nodrm' + '.zip')
|
||||
myzip1 = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
|
||||
myzip1.write(os.path.join(tempdir,'book.html'),'book.html')
|
||||
myzip1.write(os.path.join(tempdir,'book.opf'),'book.opf')
|
||||
if os.path.isfile(os.path.join(tempdir,'cover.jpg')):
|
||||
myzip1.write(os.path.join(tempdir,'cover.jpg'),'cover.jpg')
|
||||
myzip1.write(os.path.join(tempdir,'style.css'),'style.css')
|
||||
zipUpDir(myzip1, tempdir, 'img')
|
||||
myzip1.close()
|
||||
|
||||
print " Creating SVG ZIP Archive"
|
||||
zipname = os.path.join(outdir, bookname + '_SVG' + '.zip')
|
||||
myzip2 = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
|
||||
myzip2.write(os.path.join(tempdir,'index_svg.xhtml'),'index_svg.xhtml')
|
||||
zipUpDir(myzip2, tempdir, 'svg')
|
||||
zipUpDir(myzip2, tempdir, 'img')
|
||||
myzip2.close()
|
||||
|
||||
print " Creating XML ZIP Archive"
|
||||
zipname = os.path.join(outdir, bookname + '_XML' + '.zip')
|
||||
myzip3 = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
|
||||
targetdir = os.path.join(tempdir,'xml')
|
||||
zipUpDir(myzip3, targetdir, '')
|
||||
zipUpDir(myzip3, tempdir, 'img')
|
||||
myzip3.close()
|
||||
|
||||
shutil.rmtree(tempdir)
|
||||
|
||||
return 0
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
sys.exit(main())
|
||||
|
||||
Reference in New Issue
Block a user