mirrors/DeDRM_tools
mirrors/DeDRM_tools
1
0
Fork 0
mirror of https://github.com/noDRM/DeDRM_tools.git synced 2024-11-16 19:06:09 +06:00
Code Issues Projects Releases Wiki Activity

topazscripts 1.3 by some_updates

Browse Source
This commit is contained in:
some_updates 2010-01-19 12:11:59 +00:00 committed by Apprentice Alf
parent 0a437510f6
commit c1e5943471
6 changed files with 651 additions and 554 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

379
Topaz_Tools/lib/convert2xml.py
View File

@ -160,101 +160,159 @@ class PageParser(object):
# tag : (number of arguments, argument type, subtags present, special case of subtags presents when escaped)
token_tags = {
'book' : (1, 'snippets', 1, 0),
'version' : (1, 'snippets', 1, 0),
'stylesheet' : (1, 'snippets', 1, 0),
'links' : (0, 'number', 0, 1),
'pages' : (0, 'number', 0, 1),
'page' : (1, 'snippets', 1, 0),
'group' : (1, 'snippets', 1, 0),
'region' : (1, 'snippets', 1, 0),
'reflow' : (1, 'number', 1, 0),
'img' : (1, 'snippets', 1, 0),
'paragraph' : (1, 'snippets', 1, 0),
'extratokens' : (1, 'snippets', 1, 0),
'style' : (1, 'snippets', 1, 0),
'rule' : (1, 'snippets', 1, 0),
'_span' : (1, 'snippets', 1, 0),
'word_semantic': (1, 'snippets', 1, 1),
'value' : (1, 'scalar_text', 0, 0),
'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),
'x' : (1, 'number', 0, 0),
'y' : (1, 'number', 0, 0),
'links.page' : (1, 'number', 0, 0),
'link_id' : (1, '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),
'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),
'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),
'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),
'_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),
'sh' : (1, 'number', 0, 0),
'word' : (0, 'number', 1, 1),
'src' : (1, 'scalar_number', 0, 0),
'rel' : (1, 'number', 0, 0),
'row' : (1, 'number', 0, 0),
'startID' : (1, 'number', 0, 1),
'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),
'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),
'glyphID' : (1, 'number', 0, 0),
'rootID' : (1, 'number', 0, 0),
'stemID' : (1, 'number', 0, 0),
'margin-top' : (1, 'number', 0, 0),
'stemPage' : (1, 'number', 0, 0),
'dehyphen' : (1, 'number', 1, 1),
'rootID' : (1, 'number', 0, 0),
'paraCont' : (1, 'number', 1, 1),
'paraStems' : (1, 'number', 1, 1),
'wordStems' : (1, 'number', 1, 1),
'original' : (0, 'number', 0, 1),
'use' : (1, 'number', 0, 0),
'vtx' : (1, 'number', 0, 1),
'len' : (1, 'number', 0, 1),
'dpi' : (1, 'number', 0, 0),
'n' : (1, 'number', 0, 0),
'id' : (1, 'number', 0, 0),
'ref' : (1, 'number', 0, 0),
'pnum' : (1, 'number', 0, 0),
'pid' : (1, 'text', 0, 0),
'info' : (0, 'number', 1, 0),
'bl' : (1, 'raw', 0, 0),
'firstGlyph' : (1, 'raw', 0, 0),
'lastGlyph' : (1, 'raw', 0, 0),
'ocrText' : (1, 'text', 0, 0),
'title' : (1, 'text', 0, 0),
'href' : (1, 'text', 0, 0),
'_parent_type' : (1, 'text', 0, 0),
'attr' : (1, 'scalar_text', 0, 0),
'justify' : (1, 'scalar_text', 0, 0),
'align' : (1, 'scalar_text', 0, 0),
'layout' : (1, 'scalar_text', 0, 0),
'pageid' : (1, 'scalar_text', 0, 0),
'pagelabel' : (1, 'scalar_text', 0, 0),
'type' : (1, 'text', 0, 0),
'class' : (1, 'scalar_text', 0, 0),
'container' : (1, 'scalar_text', 0, 0),
'_after_class' : (1, 'scalar_text', 0, 0),
'_tag' : (1, 'scalar_text', 0, 0),
'pos' : (1, 'scalar_text', 0, 0),
'page_num' : (1, 'scalar_text', 0, 0),
'page_type' : (1, 'scalar_text', 0, 0),
'findlists' : (1, 'scalar_text', 0, 0),
'FlowEdit_1_id' : (1, 'scalar_text', 0, 0),
'FlowEdit_1_version' : (1, 'scalar_text', 0, 0),
'Schema_id' : (1, 'scalar_text', 0, 0),
'Schema_version' : (1, 'scalar_text', 0, 0),
'Topaz_version' : (1, 'scalar_text', 0, 0),
'WordDetailEdit_1_id' : (1, 'scalar_text', 0, 0),
'WordDetailEdit_1_version' : (1, 'scalar_text', 0, 0),
'ZoneEdit_1_id' : (1, 'scalar_text', 0, 0),
'ZoneEdit_1_version' : (1, 'scalar_text', 0, 0),
'chapterheaders' : (1, 'scalar_text', 0, 0),
'creation_date' : (1, 'scalar_text', 0, 0),
'header_footer' : (1, 'scalar_text', 0, 0),
'init_from_ocr' : (1, 'scalar_text', 0, 0),
'letter_insertion' : (1, 'scalar_text', 0, 0),
'xmlinj_convert' : (1, 'scalar_text', 0, 0),
'xmlinj_reflow' : (1, 'scalar_text', 0, 0),
'xmlinj_transform' : (1, 'scalar_text', 0, 0),
'startID.id' : (1, 'number', 0, 0),
}
@ -404,101 +462,25 @@ class PageParser(object):
return
# loop: pass though values unchanged
# DO NOT CHANGE - this has proven to be correct
def doLoop76Mode0(self, argtype, cnt):
result = []
for i in xrange(cnt):
result.append(self.formatArg(readEncodedNumber(self.fo), argtype))
return result
# loop generating values relative to the *negative*
# of the offset - don't ask why - it just is
# DO NOT CHANGE - this has proven to be correct
def doLoop76Mode1(self, argtype, cnt):
result = []
offset = -readEncodedNumber(self.fo)
for i in xrange(cnt):
val = readEncodedNumber(self.fo) + offset
result.append(self.formatArg(val, argtype))
return result
# loop generating values with starting value and accumulation
# DO NOT CHANGE - this has proven to be the correct
def doLoop76Mode2(self, argtype, cnt):
result = []
ptr = readEncodedNumber(self.fo)
result.append(self.formatArg(ptr, argtype))
for i in xrange(cnt-1):
ptr = ptr + readEncodedNumber(self.fo)
result.append(self.formatArg(ptr, argtype))
return result
# loop generating values with starting value and accumulation
# **after** subtracting adjustment value from each
# DO NOT CHANGE - this has been proven to be correct
def doLoop76Mode3(self, argtype, cnt):
# 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)
ptr = readEncodedNumber(self.fo)
ptr = ptr - adj
result.append(self.formatArg(ptr, argtype))
for i in xrange(cnt-1):
ptr = ptr + readEncodedNumber(self.fo) - adj
result.append(self.formatArg(ptr,argtype))
return result
# loop using runing sum of data values and starting value
# with accumulation to get new value
# Again, don't ask it took me forever to figure this out
# DO NOT CHANGE - this has been proven to be correct
def doLoop76Mode4(self, argtype, cnt):
result = []
val = readEncodedNumber(self.fo)
runsum = val
ptr = val
result.append(self.formatArg(ptr, argtype))
for i in xrange(cnt-1):
runsum += readEncodedNumber(self.fo)
ptr = ptr + runsum
result.append(self.formatArg(ptr,argtype))
return result
# loop using and extra value as an adjustment
# and a running sum of the values after subtracting
# the adjustment, added to a ptr to get a new pointer
def doLoop76Mode5(self, argtype, cnt):
result = []
adj = readEncodedNumber(self.fo)
ptr = 0
runsum = 0
mode = mode >> 1
x = []
for i in xrange(cnt):
val = readEncodedNumber(self.fo)
runsum += (val - adj)
ptr = ptr +runsum
result.append(self.formatArg(ptr,argtype))
return result
# FIXME: I have only 4 points to work this out with inside my book
# So may be wrong but it is correct for my 4 points
def doLoop76Mode6(self, argtype, cnt):
result = []
oldval = 0
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):
val = readEncodedNumber(self.fo)
ptr= (3 * oldval) + val + 1
result.append(self.formatArg(ptr,argtype))
oldval = val
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
@ -507,58 +489,21 @@ class PageParser(object):
# since they did not appear in the test cases
def decodeCMD(self, cmd, argtype):
# if (cmd == 0x72):
# self.doLoop72(argtype)
# result =[]
# return result
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, ': '
if (mode == 0x00):
return self.doLoop76Mode0(argtype, cnt)
elif (mode == 0x01):
return self.doLoop76Mode1(argtype, cnt)
elif (mode == 0x02):
return self.doLoop76Mode2(argtype, cnt)
elif (mode == 0x03):
return self.doLoop76Mode3(argtype, cnt)
elif (mode == 0x04):
return self.doLoop76Mode4(argtype, cnt)
elif (mode == 0x05):
return self.doLoop76Mode5(argtype, cnt)
elif (mode == 0x06):
return self.doLoop76Mode6(argtype, cnt)
else:
if self.debug :
# try to mark any unknown loop comands
# if they exist, unless they are used to process
# text or some other known list, we won't be able to prove them correct
print '*** Unknown Loop 0x%x %d %d :' % (cmd, cnt, mode)
for i in xrange(cnt):
val = readEncodedNumber(self.fo)
print ' 0x%x' % val,
print ' '
result = []
return result
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]
@ -727,7 +672,7 @@ class PageParser(object):
self.doc.append(tag)
else:
if self.debug:
print "Mina Loop: Unknown value: %x" % v
print "Main Loop: Unknown value: %x" % v
# now do snippet injection

192
Topaz_Tools/lib/flatxml2html.py
View File

@ -11,9 +11,16 @@ from struct import unpack
class DocParser(object):
def __init__(self, flatxml, fileid):
def __init__(self, flatxml, classlst, fileid):
self.id = os.path.basename(fileid).replace('.dat','')
self.flatdoc = flatxml.split('\n')
self.classList = {}
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.ocrtext = []
self.link_id = []
self.link_title = []
@ -22,6 +29,18 @@ class DocParser(object):
self.paracont_stemid = []
self.parastems_stemid = []
# find tag if within pos to end inclusive
def lineinDoc(self, pos) :
docList = self.flatdoc
cnt = len(docList)
if (pos >= 0) and (pos < cnt) :
item = docList[pos]
if item.find('=') >= 0:
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
return name, argres
# find tag if within pos to end inclusive
@ -61,57 +80,117 @@ class DocParser(object):
return startpos
# get a description of the paragraph
# build a description of the paragraph
def getParaDescription(self, start, end):
result = []
# normal paragraph
(pos, pclass) = self.findinDoc('paragraph.class',start,end)
# class names are an issue given topaz starts 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 with reflow issues
# so we clean this up by lowercasing, prepend 'cl_', and remove all end pieces after reclustered
# 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 with reflow issues
# but then not actually provide all of these _reclustereed classes in the stylesheet!
# so we clean this up by lowercasing, prepend 'cl_', and if not in the class
# list from the stylesheet, trying once more with "_reclustered*" removed
# if still not in stylesheet, let it pass as is
pclass = pclass.lower()
pclass = 'cl_' + pclass
p = pclass.find('reclustered')
if p > 0 : pclass = pclass[0:p+11]
if pclass not in self.classList:
p = pclass.find('_reclustered')
if p > 0 :
baseclass = pclass[0:p]
if baseclass in self.classList:
pclass = baseclass
# 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) :
return pclass, int(sfirst), int(slast)
# some paragraphs are instead split into multiple spans and some even have word_semantic tags as well
# so walk through this region keeping track of the first firstword, and the last lastWord
# on any items that have it
(pos, sfirst) = self.findinDoc('firstWord',start, end)
first = int(sfirst)
last = -1
for i in xrange(pos+1,end):
(pos, slast) = self.findinDoc('lastWord',i,i+1)
if slast != None:
last = int(slast)
return pclass, first, last
for wordnum in xrange(first, last):
result.append(('ocr', wordnum))
return pclass, result
# this type of paragrph may be made up of multiple _spans, inline
# word monograms (images) and words with semantic meaning
# need to parse this type line by line
line = start + 1
word_class = ''
while (line < end) :
(name, argres) = self.lineinDoc(line)
if name.endswith('_span.firstWord') :
first = int(argres)
(name, argres) = self.lineinDoc(line+1)
if not name.endswith('_span.lastWord'):
print 'Error: - incorrect _span ordering inside paragraph'
last = int(argres)
for wordnum in xrange(first, last):
result.append(('ocr', wordnum))
line += 1
elif name.endswith('word.class'):
(cname, space) = argres.split('-',1)
if cname == 'spaceafter':
word_class = 'sa'
elif name.endswith('word.img.src'):
result.append(('img' + word_class, int(argres)))
word_class = ''
elif name.endswith('word_semantic.firstWord'):
first = int(argres)
(name, argres) = self.lineinDoc(line+1)
if not name.endswith('word_semantic.lastWord'):
print 'Error: - incorrect word_semantic ordering inside paragraph'
last = int(argres)
for wordnum in xrange(first, last):
result.append(('ocr', wordnum))
line += 1
line += 1
return pclass, result
def buildParagraph(self, cname, first, last, type, regtype) :
def buildParagraph(self, cname, pdesc, type, regtype) :
parares = ''
sep =''
br_lb = False
if (regtype == 'fixed') or (regtype == 'chapterheading') :
br_lb = True
handle_links = False
if len(self.link_id) > 0:
handle_links = True
if (type == 'full') or (type == 'begin') :
parares += '<p class="' + cname + '">'
if (type == 'end'):
parares += ' '
for j in xrange(first, last) :
word = self.ocrtext[j]
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[j]
link = self.link_id[num]
if (link > 0):
title = self.link_title[link-1]
if title == "": title='_link_'
@ -128,7 +207,7 @@ class DocParser(object):
if word == '_link_' : word = ''
if word == '_lb_':
if (j-1) in self.dehyphen_rootid :
if (num-1) in self.dehyphen_rootid :
word = ''
sep = ''
elif handle_links :
@ -141,12 +220,22 @@ class DocParser(object):
word = '\n'
sep = ''
if j in self.dehyphen_rootid :
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
if len(sep) > 0 : parares = parares[0:-1]
if (type == 'full') or (type == 'end') :
parares += '</p>'
@ -222,7 +311,7 @@ class DocParser(object):
htmlpage += '<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc)
elif regtype == 'chapterheading' :
(pclass, first, last) = self.getParaDescription(start,end)
(pclass, pdesc) = self.getParaDescription(start,end)
if not breakSet:
htmlpage += '<div style="page-break-after: always;">&nbsp;</div>\n'
breakSet = True
@ -234,7 +323,7 @@ class DocParser(object):
if pclass[3:7] == 'ch2-' : tag = 'h2'
if pclass[3:7] == 'ch3-' : tag = 'h3'
htmlpage += '<' + tag + ' class="' + pclass + '">'
htmlpage += self.buildParagraph(pclass,first,last,'middle', regtype)
htmlpage += self.buildParagraph(pclass, pdesc, 'middle', regtype)
htmlpage += '</' + tag + '>'
elif (regtype == 'text') or (regtype == 'fixed') or (regtype == 'insert') :
@ -247,17 +336,17 @@ class DocParser(object):
if not anchorSet:
htmlpage += '<div id="' + self.id + '" class="page_' + pagetype + '">&nbsp</div>\n'
anchorSet = True
(pclass, first, last) = self.getParaDescription(start,end)
(pclass, pdesc) = self.getParaDescription(start,end)
if 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, first, last, 'middle', regtype)
htmlpage += self.buildParagraph(pclass, pdesc, 'middle', regtype)
htmlpage += '</' + tag + '>'
else :
htmlpage += self.buildParagraph(pclass, first, last, ptype, regtype)
htmlpage += self.buildParagraph(pclass, pdesc, ptype, regtype)
elif (regtype == 'tocentry') :
@ -271,12 +360,43 @@ class DocParser(object):
if not anchorSet:
htmlpage += '<div id="' + self.id + '" class="page_' + pagetype + '">&nbsp</div>\n'
anchorSet = True
(pclass, first, last) = self.getParaDescription(start,end)
htmlpage += self.buildParagraph(pclass, first, last, ptype, regtype)
(pclass, pdesc) = self.getParaDescription(start,end)
htmlpage += self.buildParagraph(pclass, pdesc, ptype, regtype)
elif regtype == 'synth_fcvr.center' :
if not anchorSet:
htmlpage += '<div id="' + self.id + '" class="page_' + pagetype + '">&nbsp</div>\n'
anchorSet = True
(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 'Unknown region type', regtype
print 'Warning: skipping this region'
print 'Warning: Unknown region type', regtype
print 'Treating this like a "fixed" region'
regtype = 'fixed'
ptype = 'full'
# check to see if this is a continution from the previous page
if (len(self.parastems_stemid) > 0):
ptype = 'end'
self.parastems_stemid=[]
else:
if not anchorSet:
htmlpage += '<div id="' + self.id + '" class="page_' + pagetype + '">&nbsp</div>\n'
anchorSet = True
(pclass, desc) = self.getParaDescription(start,end)
if 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)
if len(self.paracont_stemid) > 0 :
if htmlpage[-4:] == '</p>':
@ -289,10 +409,10 @@ class DocParser(object):
def convert2HTML(flatxml, fileid):
def convert2HTML(flatxml, classlst, fileid):
# create a document parser
dp = DocParser(flatxml, fileid)
dp = DocParser(flatxml, classlst, fileid)
htmlpage = dp.process()

13
Topaz_Tools/lib/genhtml.py
View File

@ -95,22 +95,27 @@ def main(argv):
htmlstr += '<meta name="Author" content="' + meta_array['Authors'] + '" />\n'
htmlstr += '<meta name="Title" content="' + meta_array['Title'] + '" />\n'
# get some scaling info from metadata to use while processing styles
fontsize = '135'
if 'fontSize' in meta_array:
fontsize = meta_array['fontSize']
print ' ', 'other0000.dat'
fname = os.path.join(bookDir,'other0000.dat')
xname = os.path.join(bookDir, 'style.css')
xmlstr = convert2xml.main('convert2xml.py --flat-xml ' + dictFile + ' ' + fname)
cssstr = '<style>\n'
cssstr += stylexml2css.convert2CSS(xmlstr)
cssstr += '</style>\n'
htmlstr += '<style>\n'
cssstr , classlst = stylexml2css.convert2CSS(xmlstr, fontsize)
file(xname, 'wb').write(cssstr)
htmlstr += cssstr
htmlstr += '</style>\n'
htmlstr += '</head>\n<body>\n'
for filename in filenames:
print ' ', filename
fname = os.path.join(pageDir,filename)
flat_xml = convert2xml.main('convert2xml.py --flat-xml ' + dictFile + ' ' + fname)
htmlstr += flatxml2html.convert2HTML(flat_xml, fname)
htmlstr += flatxml2html.convert2HTML(flat_xml, classlst, fname)
htmlstr += '</body>\n</html>\n'

37
Topaz_Tools/lib/gensvg.py
View File

@ -60,11 +60,24 @@ class GParser(object):
zx = tx[self.vlen[k-1]+1:self.vlen[k]+1]
zy = ty[self.vlen[k-1]+1:self.vlen[k]+1]
p += 1
for j in xrange(0, len(zx)):
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])
else:
path += 'L %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
@ -72,8 +85,10 @@ class PParser(object):
def __init__(self, flatxml):
self.flatdoc = flatxml.split('\n')
self.temp = []
self.ph = self.getData('page.h')[0]
self.pw = self.getData('page.w')[0]
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')
@ -120,12 +135,12 @@ class PParser(object):
def getImages(self):
result = []
self.temp = self.flatdoc
while (self.getDataTemp('region.img') != None):
h = self.getDataTemp('region.img.h')[0]
w = self.getDataTemp('region.img.w')[0]
x = self.getDataTemp('region.img.x')[0]
y = self.getDataTemp('region.img.y')[0]
src = self.getDataTemp('region.img.src')[0]
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

10
Topaz_Tools/lib/readme.txt
View File

@ -1,3 +1,13 @@
Contributors:
cmbtc - removal of drm which made all of this possible
clarknova - for all of the svg and glyph generation and many other bug fixes and improvements
skindle - for figuing out the general case for the mode loops
some updates - for conversion to xml, basic html
DiapDealer - for extensive testing and feeback
and others for posting, feedback and testing
This is experimental and it will probably not work for you but...
ALSO: Please do not use any of this to steal. Theft is wrong.

48
Topaz_Tools/lib/stylexml2css.py
View File

@ -11,8 +11,9 @@ from struct import unpack
class DocParser(object):
def __init__(self, flatxml):
def __init__(self, flatxml, fontsize):
self.flatdoc = flatxml.split('\n')
self.fontsize = int(fontsize)
stags = {
'paragraph' : 'p',
@ -20,14 +21,14 @@ class DocParser(object):
}
attr_val_map = {
'hang' : ('text-indent: ', 135),
'indent' : ('text-indent: ', 135),
'line-space' : ('line-height: ', 190),
'margin-bottom' : ('margin-bottom: ', 135),
'margin-left' : ('margin-left: ', 135),
'margin-right' : ('margin-right: ', 135),
'margin-top' : ('margin-top: ', 135),
'space-after' : ('padding-bottom: ', 135),
'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 = {
@ -55,7 +56,7 @@ class DocParser(object):
for j in xrange(pos, end):
item = docList[j]
if item.find('=') >= 0:
(name, argres) = item.split('=')
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
@ -81,6 +82,7 @@ class DocParser(object):
def process(self):
classlst = ''
csspage = ''
# generate a list of each <style> starting point in the stylesheet
@ -132,23 +134,19 @@ class DocParser(object):
else :
# handle value based attributes
if attr in self.attr_val_map :
(name, scale) = self.attr_val_map[attr]
name = self.attr_val_map[attr]
scale = self.fontsize
if attr == 'line-space': scale = scale * 1.41
if not ((attr == 'hang') and (int(val) == 0)) :
ems = int(val)/scale
cssargs[attr] = (self.attr_val_map[attr][0], ems)
cssargs[attr] = (self.attr_val_map[attr], ems)
keep = True
start = pos + 1
# disable all of the after class tags until I figure out how to handle them
# remove all numerals after the "reclustered"
if aftclass != "" : keep = False
p = sclass.find('reclustered')
if p >= 0:
sclass = sclass[0:p+11]
if keep :
# make sure line-space does not go below 1em
if 'line-space' in cssargs:
@ -156,7 +154,7 @@ class DocParser(object):
val = cssargs['line-space'][1]
if val < 1.0: val = 1.0
del cssargs['line-space']
cssargs['line-space'] = (self.attr_val_map['line-space'][0], val)
cssargs['line-space'] = (self.attr_val_map['line-space'], val)
@ -165,7 +163,7 @@ class DocParser(object):
hseg = cssargs['hang'][0]
hval = cssargs['hang'][1]
del cssargs['hang']
cssargs['hang'] = (self.attr_val_map['hang'][0], -hval)
cssargs['hang'] = (self.attr_val_map['hang'], -hval)
mval = 0
mseg = 'margin-left: '
if 'margin-left' in cssargs:
@ -188,6 +186,9 @@ class DocParser(object):
cssline += '}'
if sclass != '' :
classlst += sclass + '\n'
# handle special case of paragraph class used inside chapter heading
# and non-chapter headings
if sclass != '' :
@ -207,14 +208,15 @@ class DocParser(object):
csspage += self.stags[tag] + cssline + '\n'
return csspage
return csspage, classlst
def convert2CSS(flatxml):
def convert2CSS(flatxml, fontsize):
# create a document parser
dp = DocParser(flatxml)
dp = DocParser(flatxml, fontsize)
csspage = dp.process()