francois/DeDRM_tools
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

ereader2html 0.09 by brutusbum, also splitting out into erdr2pml and xpml2xhtml as well, so that we get some history

Browse Source
This commit is contained in:
brutusbum
2009-12-01 20:59:14 +00:00
committed by Apprentice Alf
parent 5115a4aed6
commit e4a0f92846
3 changed files with 1900 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files

194
eReader_Tools/lib/ereader2html.py
View File

@@ -44,10 +44,10 @@
# Using that with Calibre works a lot better than the HTML
# conversion in this code.
# 0.07 - Further Improved type 272 support for sidebars with all earlier fixes
# 0.07a - Fixed some typos
# 0.08 - fixed typos, removed extraneous things
# 0.09 - tried to greatly improve html conversion especially with \t tags
__version__='0.07'
__version__='0.09'
# Import Psyco if available
try:
@@ -80,7 +80,6 @@ import logging
logging.basicConfig()
#logging.basicConfig(level=logging.DEBUG)
write_pml=False
ECB = 0
CBC = 1
@@ -503,17 +502,25 @@ class EreaderProcessor(object):
# the remaining records of the footnote sections need to be decoded with the content_key and zlib inflated
des = Des(fixKey(self.content_key))
r += '\\w="100%"'
r += 'Footnotes\p'
r += '\\pFootnotes:\n\n'
for i in xrange(1,self.num_footnote_pages):
logging.debug('get footnotepage %d', i)
id_len = ord(fnote_ids[2])
id = fnote_ids[3:3+id_len]
fmarker='\Q="%s"' % id
fmarker='\\t\\Q="footnote-%s"' % id
r+=fmarker
r += zlib.decompress(des.decrypt(self.section_reader(self.first_footnote_page + i)))
r += '\n'
r += '\\t\n\n'
fnote_ids = fnote_ids[id_len+4:]
# according to ereader pml spec we should be outputing the following xml for each footnote - but then we would have to handle
# parsing it back in to convert it since that xml is not valid xhtml
# fmarker = '<footnote id="footnote-%s">\n' % id
# fmarker += zlib.decompress(des.decrypt(self.section_reader(self.first_footnote_page + i)))
# fmarker += '\n</footnote>\n'
# r += fmarker
# now handle sidebar pages
if self.num_sidebar_pages > 0:
# the record 0 of the sidebar section must pass through the Xor Table to make it useful
@@ -522,20 +529,70 @@ class EreaderProcessor(object):
# the remaining records of the sidebar sections need to be decoded with the content_key and zlib inflated
des = Des(fixKey(self.content_key))
r += '\\w="100%"'
r += 'Sidebars\p'
r += '\\pSidebars:\n\n'
for i in xrange(1,self.num_sidebar_pages):
id_len = ord(sbar_ids[2])
id = sbar_ids[3:3+id_len]
fmarker='\Q="%s"' % id
r+=fmarker
smarker='\\t\\Q="sidebar-%s"' % id
r+=smarker
r += zlib.decompress(des.decrypt(self.section_reader(self.first_sidebar_page + i)))
r += '\n'
r += '\\t\n\n'
sbar_ids = sbar_ids[id_len+4:]
# according to ereader pml spec we should be outputing the following xml for each sidebar - but then we would have to handle
# parsing it back in to convert it since that xml is not valid xhtml
# smarker = '<sidebar id="sidebar-%s">\n' % id
# smarker += zlib.decompress(des.decrypt(self.section_reader(self.first_footnote_page + i)))
# smarker += '\n</sidebar>\n'
# r += smarker
return r
class PmlConverter(object):
def __init__(self, s, bkinfo):
self.s = s
def findPrevStartofLine(src,p,n):
# find last end of previous line in substring from p to n
b1 = src.rfind('\n',p,n)
b2 = src.rfind('\\c',p,n)
b3 = src.rfind('\\r',p,n)
b4 = src.rfind('\\x',p,n)
b5 = src.rfind('\\p',p,n)
b = max(b1, b2, b3, b4, b5)
if b == -1:
return n
if b == b1:
return b + 1
return b + 2
def markHangingIndents(src):
r = ''
p = 0
while True:
if p > len(src):
return r
n = src.find('\\t', p)
if n == -1:
r += src[p:]
return r
pc = findPrevStartofLine(src,p,n)
if pc == n :
# \t tag is at start of line so indent block will work
end = src.find('\\t',n+2)
if end == -1:
end = n
r += src[p:end+2]
p = end + 2
else :
# \t tag not at start of line so hanging indent case
# recode \t to pseudo \h tags and move it to start of this line
# and recode its close as well
r += src[p:pc] + '\\h' + src[pc:n]
end = src.find('\\t',n+2)
if end == -1:
end = n+2
r += src[n+2:end] + '\\h'
p = end + 2
self.s = markHangingIndents(s)
# file(os.path.join("./pseudo.pml"), 'wb').write(self.s)
self.pos = 0
self.bkinfo = bkinfo
def nextOptAttr(self):
@@ -560,7 +617,8 @@ class PmlConverter(object):
self.pos = res
return self.s[p : res], None, None
c = self.s[p+1]
if c in 'pxcriuovtnsblBk-lI\\d':
# add in support for new pseudo tag \\h
if c in 'pxcriuovthnsblBk-lI\\d':
self.pos = p + 2
return None, c, None
if c in 'TwmqQ':
@@ -583,48 +641,36 @@ class PmlConverter(object):
self.pos = p + 1
return None, None, None
def LinePrinter(link):
return '<hr width="%s" />' % link
return '<hr width="%s" />\n' % link
def LinkPrinter(link):
return '<a href="%s">' % link
def InternalLinkPrinter(link):
return '<a href="#%s">' % link
def FootnoteLinkPrinter(link):
return '<a href="#footnote-%s">' % link
def SidebarLinkPrinter(link):
return '<a href="#sidebar-%s">' % link
def NotSupported(link):
raise NotImplemented()
def NewChapterNewPage(link):
raise NotImplemented()
def ChapterTitle(link):
print "Nonfatal Error: ChapterTitle not implemented."
return '<!-- ChapterTitle %s -->' %link
def IndentPercent(link):
print "Nonfatal Error: IndentPercent not implemented."
return '<!-- IndentPercent: %s -->' %link
return '<span style="padding-left: %s%%;"></span>' %link
def NormalFont(link):
print "Nonfatal Error: NormalFont not implemented."
return '<!-- NormalFont %s -->' %link
def StdFont(link):
print "Nonfatal Error: StdFont not implemented."
return '<!-- StdFont: %s -->' %link
def SingleBackslash(link):
print "Nonfatal Error: SingleBackslash not implemented."
return '<!-- SingleBackslash: %s -->' %link
def SoftHyphen(link):
print "Nonfatal Error: SoftHyphen not implemented."
return '<!-- SoftHyphen: %s -->' %link
def ReferenceIndexItem(link):
print "Nonfatal Error: ReferenceIndexItem not implemented."
return '<!-- CReferenceIndexItem: %s -->' %link
# See http://wiki.mobileread.com/wiki/PML#Palm_Markup_Language
html_tags = {
'c' : ('<div class="center">', '</div>'),
'r' : ('<div class="right">', '</div>'),
'i' : ('<i>', '</i>'),
'u' : ('<u>', '</u>'),
'u' : ('<span class="under">', '</span>'),
'b' : ('<strong>', '</strong>'),
'B' : ('<strong>', '</strong>'),
'o' : ('<strike>', '</strike>'),
'o' : ('<del>', '</del>'),
'v' : ('<!-- ', ' -->'),
't' : ('', ''),
't' : ('<div class="indent">','</div>'),
'h' : ('<div class="hang">','</div>'), # pseudo-tag created to handle hanging indent cases
'Sb' : ('<sub>', '</sub>'),
'Sp' : ('<sup>', '</sup>'),
'X0' : ('<h1>', '</h1>'),
@@ -632,28 +678,31 @@ class PmlConverter(object):
'X2' : ('<h3>', '</h3>'),
'X3' : ('<h4>', '</h4>'),
'X4' : ('<h5>', '</h5>'),
'l' : ('<font size="+2">', '</font>'),
'l' : ('<span class="big">', '</span>'),
'q' : (LinkPrinter, '</a>'),
'Fn' : (InternalLinkPrinter, '</a>'),
'Sd' : (InternalLinkPrinter, '</a>'),
'Fn' : (FootnoteLinkPrinter, '</a>'),
'Sd' : (SidebarLinkPrinter, '</a>'),
'w' : (LinePrinter, ''),
#'m' : handled in if block,
#'Q' : handled in if block,
#'a' : handled in if block,
#'U' : handled in if block,
'x' : ('<h1> class="breakbefore">', '</h1>'),
'Cn' : (ChapterTitle, ''),
'x' : ('<h1 class="breakbefore">', '</h1>'),
#'C0' : handled in if block,
#'C1' : handled in if block,
#'C2' : handled in if block,
#'C3' : handled in if block,
#'C4' : handled in if block,
'T' : (IndentPercent, ''),
'n' : (NormalFont, ''),
#'s' : (StdFont, ''),
's' : ('', ''),
'k' : ('<span style="font-variant: small-caps;">', '</span>'), # NOTE some pdb's then go ahead and use uppercase letters - which doesn't format the way one would expect (perhaps post process the output with html dom and lower ase is only upper case letters are found?)
'\\' : (SingleBackslash, ''),
'-' : (SoftHyphen, ''),
'I' : (ReferenceIndexItem, ''),
'k' : ('<span class="small">', '</span>'),
'I' : ('<i>', '</i>'), # according to calibre - all ereader does is italicize the index entries
}
html_one_tags = {
'p' : '<p class="breakafter">&nbsp;</p>\n'
'p' : '<p class="breakafter">&nbsp;</p>\n',
'\\': '\\',
'-' : '&shy;',
}
pml_chars = {
160 : '&nbsp;',130 : '&#8212;',131: '&#402;',132: '&#8222;',
@@ -674,7 +723,17 @@ class PmlConverter(object):
final += '<meta name="Copyright" content="%s"/>\n' % copyright
final += '<meta name="Publisher" content="%s"/>\n' % publisher
final += '<meta name="ISBN" content="%s"/>\n' % isbn
final += '<style type="text/css">\ndiv.center { text-align:center; }\ndiv.right { text-align:right; }\n.breakbefore { page-break-before: always; }\n.breakafter { page-break-after: always; }\n</style>\n'
final += '<style type="text/css">\n'
final += 'div.center { text-align:center; }\n'
final += 'div.right { text-align:right; }\n'
final += 'div.indent { margin-left: 5%; }\n'
final += 'div.hang { text-indent: -5%; margin-left: 5%; }\n'
final += 'span.big { font-size: 175%; }\n'
final += 'span.small { font-size: 50%; }\n'
final += 'span.under { text-decoration: underline; }\n'
final += '.breakbefore { page-break-before: always; }\n'
final += '.breakafter { page-break-after: always; }\n'
final += '</style>\n'
final += '</head><body>\n'
in_tags = []
def makeText(s):
@@ -682,7 +741,7 @@ class PmlConverter(object):
#s = s.replace('"', '&quot;')
s = s.replace('<', '&lt;')
s = s.replace('>', '&gt;')
s = s.replace('\n', '<br>\n')
s = s.replace('\n', '<br />\n')
return s
while True:
r = self.next()
@@ -698,7 +757,6 @@ class PmlConverter(object):
if type(r) != str:
r = r(attr)
return r
if cmd in self.html_tags:
pair = (cmd, attr)
if cmd not in [a for (a,b) in in_tags]:
@@ -719,21 +777,33 @@ class PmlConverter(object):
if cmd in self.html_one_tags:
final += self.html_one_tags[cmd]
if cmd == 'm':
unquotedimagepath = bookname + "_img/" + attr
unquotedimagepath = "images/" + attr
imagepath = urllib.quote( unquotedimagepath )
final += '<img src="%s" alt="">' % imagepath
if cmd == 'Q':
final += '<a name="%s" id="%s"> </a>' % (attr, attr)
final += '<span id="%s"> </span>' % attr
if cmd == 'C0':
final += '<!-- ContentsList "%s" -->' % attr
if cmd == 'C1':
final += '<!-- ContentsList " %s" -->' % attr
if cmd == 'C2':
final += '<!-- ContentsList " %s" -->' % attr
if cmd == 'C3':
final += '<!-- ContentsList " %s" -->' % attr
if cmd == 'C4':
final += '<!-- ContentsList " %s" -->' % attr
if cmd == 'a':
final += self.pml_chars.get(attr, '&#%d;' % attr)
if cmd == 'U':
final += '&#%d;' % attr
final += '</body></html>\n'
while True:
s = final.replace('<br>\n<br>\n<br>\n', '<br>\n<br>\n')
if s == final:
break
final = s
# while True:
# s = final.replace('<br />\n<br />\n<br />\n', '<br />\n<br />\n')
# if s == final:
# break
# final = s
s = final.replace('</div><br />','</div>\n')
final = s
return final
def convertEreaderToHtml(infile, name, cc, outdir):
@@ -743,7 +813,7 @@ def convertEreaderToHtml(infile, name, cc, outdir):
er = EreaderProcessor(sect.loadSection, name, cc)
if er.getNumImages() > 0:
imagedir = bookname + "_img"
imagedir = "images/"
imagedirpath = os.path.join(outdir,imagedir)
if not os.path.exists(imagedirpath):
os.makedirs(imagedirpath)
@@ -761,14 +831,14 @@ def convertEreaderToHtml(infile, name, cc, outdir):
htmlfilename = bookname + ".html"
file(os.path.join(outdir, htmlfilename),'wb').write(pml.process())
ts = er.getExpandedTextSizesData()
file(os.path.join(outdir, 'xtextsizes.dat'), 'wb').write(ts)
# ts = er.getExpandedTextSizesData()
# file(os.path.join(outdir, 'xtextsizes.dat'), 'wb').write(ts)
cv = er.getChapterNamePMLOffsetData()
file(os.path.join(outdir, 'chapters.dat'), 'wb').write(cv)
lv = er.getLinkNamePMLOffsetData()
file(os.path.join(outdir, 'links.dat'), 'wb').write(lv)
# lv = er.getLinkNamePMLOffsetData()
# file(os.path.join(outdir, 'links.dat'), 'wb').write(lv)
def main(argv=None):
@@ -811,4 +881,4 @@ if __name__ == "__main__":
#command = """sys.exit(main())"""
#cProfile.runctx( command, globals(), locals(), filename="cprofile.profile" )
sys.exit(main())
sys.exit(main())