def readepub(zf):
global mw, basefile, ht, conf
names = []; allnames = []
mw.waitcursor(True)
for aa in zf.infolist():
#print aa.filename, aa.file_size,
# Extract
if conf.extract:
try:
dd = conf.data_dir + "/" + basefile + "/" + os.path.split(aa.filename)[0]
if not os.path.isdir(dd):
os.makedirs(dd)
fh = zf.open(aa.filename)
fff = dd + "/" + os.path.basename(aa.filename)
# Only extract once:
if not os.path.isfile(fff):
mw.prog.set_text("Extracting '" + aa.filename + "'")
usleep(1)
fh2 = open(fff, "w+")
while 1:
sss = fh.readline()
if sss == "": break
fh2.write(sss)
fh2.close()
except:
print_exception("Cannot extract file")
if os.path.splitext(aa.filename)[1].find("htm") >= 0:
names.append(aa.filename)
allnames.append(aa.filename)
#print
#names.sort(cmp)
#print names
found = False
for aa in allnames:
if "toc.ncx" in aa:
#print "Found toc", aa
found = True
fh = zf.open(aa)
ht = TocHTMLParser(mw)
while 1:
sss = fh.readline()
if sss == "": break
ht.feed(sss)
mw.add_text(sss, True)
break
# No TOC Open and parse all the [x]htm[l] files
if not found:
nnn = "000"; mw.update_tree("Start", nnn); mw.add_text_mark(nnn)
for bbb in names:
openHTML(zf, bbb)
nnn = "999"; mw.update_tree("End", nnn); mw.add_text_mark(nnn)
mw.waitcursor(False)
def html_tick(fname):
global conf, mw
conf.recurse = False
try:
fh = open(fname)
ht = pubhtml.HTML_Parser(conf)
ht.mw = mw
ht.fh = fh
ht.getimagecb = getimagecb
ln = 0
mw.waitcursor(True)
while 1:
if conf.recurse:
break
sss = fh.readline()
if ln % 10 == 0:
mw.prog.set_text("Reading line: %d" % ln)
usleep(1)
ln += 1
if sss == "": break
if conf.recurse:
break
if ht.feed(sss):
break
mw.add_text(sss, True)
mw.fname = fname
mw.waitcursor(False)
except:
print_exception("Opening HTML")
mw.prog.set_text("Done Loading.")
mw.hpaned.set_position(0)
mw.gohome()
def reader_tick():
global conf
try:
readepub(zf)
if not conf.nogui:
# Seen this ebook before?
seb = mw.conf.sql.get(conf.basefile)
if seb and seb != "":
#print "opening remembered file", seb
openHTMLcb(seb)
# Seen this html before? Go to offset.
comp = conf.basefile + "/" + seb
offs = mw.conf.sql.get(comp)
if offs != None:
#print "seen", mw.fname, offs
iter = mw.buffer_1.get_iter_at_offset(int(offs))
mw.buffer_1.place_cursor(iter)
mw.view.scroll_to_iter(iter, 0.2) #, True, 0, 0)
# Load last selection
comp = conf.basefile + "_sel"
lsel = mw.conf.sql.get(comp)
mw.sel_tree(lsel)
#print "lastsel", comp, lsel
else:
#print "opening first link", conf.ht.firstlink
openHTMLcb(conf.ht.firstlink)
# Requested command line read
if conf.read:
mw.read_tts(None)
except:
print_exception("Reader tick")
def stop_tts(self):
self.stopspeak = True
try:
pubutil.withps(self.stop_tts2)
except:
pubutil.print_exception("Cannot kill")
#self.speech_pid.pid
self.speech_pid = None
self.butt4.set_label("_Read")
return
def openHTML(zf, fname, mark = True):
# Is it a tagged link?
ppp = fname.find("#")
spl = []
if spl >= 0:
spl = fname.split("#")
fname = spl[0]
# Not loaded, load
if mw.fname != fname:
mw.clear(); mw.clear(True); found = False; fh = None
try:
fh = zf.open(fname)
found = True
except:
#print "searching for", fname, sys.exc_info()
# Search for it ...
for aa in zf.infolist():
#print aa.filename, fname
if os.path.basename(aa.filename) == fname:
try:
fh = zf.open(aa.filename)
except:
print_exception("No Zip File")
return
found = True
break
if not found: return False
try:
#ht = MyHTMLParser(mw)
ht = pubhtml.HTML_Parser()
ht.mw = mw; ht.mark = mark; ht.fh = fh
mw.waitcursor(True)
while 1:
sss = fh.readline()
if sss == "": break
if ht.feed(sss):
break
mw.add_text(sss, True)
mw.fname = fname
mw.waitcursor(False)
except:
print_exception("Parsing HTML")
# Jump to tag, if any
if len(spl) > 1:
#print "Jumping to", "'" + spl[1] + "'"
mm = mw.buffer_1.get_mark(spl[1])
if mm:
mw.view.scroll_to_mark(mm, 0.0, True, 0, 0)
return True
def speak(self, cstr):
fname = self.conf.data_dir + "/festival.txt"
try:
fh = open(fname, "w")
fh.write(cstr)
except:
pubutil.print_exception("Cannot create festival file")
return
self.speech_pid = subprocess.Popen(["festival", "--tts", fname])
#print "started", self.speech_pid.pid
gobject.timeout_add(100, self.check_speak)
def openHTMLcb(fname, addback = True):
global gzf, backlist
ret = False
#print "openHTMLcb", fname
try:
if addback:
backlist.append(fname)
ret = openHTML(gzf, fname, False)
gobject.timeout_add(100, size_tick)
except:
print_exception("Cannot load HTML")
return ret
def sel_next(self, dir=False):
#print "sel_next", self.lastsel
try:
# Move tree selection
sel = self.treeview.get_selection()
xmodel, xiter = sel.get_selected()
next = xmodel.iter_next(xiter)
if next:
sel.select_iter(next)
self.lastsel = xmodel.get_value(next, 0)
self.tree_sel_row(self.treeview)
except:
pubutil.print_exception("sel_next")
pass
def stop_tts(self):
try:
dl = os.listdir("/proc")
for aa in dl:
fname = "/proc/" + aa + "/stat"
if os.path.isfile(fname):
ff = open(fname, "r").read().split()
if self.speech_pid.pid == int(ff[3]):
os.kill(int(ff[0]), signal.SIGKILL)
except:
pubutil.print_exception("Cannot kill", self.speech_pid.pid)
self.speech_pid = None
self.butt4.set_label("Read")
return
def openHTMLcb(fname, addback = True):
global conf, gzf, backlist
ret = False
print "openHTMLcb", fname
conf.recurse = True
usleep(10)
try:
if addback:
backlist.append(fname)
ret = openHTML(gzf, fname, False)
gobject.timeout_add(100, size_tick)
except:
print_exception("Cannot load HTML")
conf.recurse = False
return ret
def locateroot():
global gzf
rootfile = ""
try:
fh = gzf.open("META-INF/container.xml")
ht = RootHTMLParser()
while 1:
sss = fh.readline()
if sss == "": break
ht.feed(sss)
if conf.rootf:
print sss,
rootfile = ht.rootfile
except:
print_exception("No meta data")
return rootfile
def sel_tree(self, xstr):
#print "sel_tree", self.lastsel
try:
sel = self.treeview.get_selection()
xmodel, xiter = sel.get_selected()
iter = xmodel.get_iter_root()
while 1:
if not iter: break
xstr2 = xmodel.get_value(iter, 0)
if xstr == xstr2:
sel.select_iter(iter)
break
iter = xmodel.iter_next(iter)
except:
pubutil.print_exception("sel_tree")
pass
def parse_ncx(fname):
global conf
ret = "", ""
try:
fh = zf.open(fname)
ht = NcxHTMLParser(contentcb)
conf.ht = ht
while 1:
sss = fh.readline()
if sss == "": break
ht.feed(sss)
if conf.header:
print sss,
if not conf.nogui:
mw.add_text(sss, True)
ret = ht.title, ht.auth
except:
print_exception("Cannot parse .ncx file")
return ret
def getimagecb(fname):
global gzf, conf
fullname = ""
#print "Image callback", "'" + fname + "'"
fh = None
try:
fh = gzf.open(fname)
fullname = fname
except:
# Search for it ...
for aa in zf.infolist():
if os.path.basename(aa.filename) == fname:
fullname = aa.filename
try:
fh = gzf.open(aa.filename)
except:
print_exception("No image File")
return
break
# Extract image file
dd = conf.data_dir + "/" + conf.basefile + "/" + os.path.split(fullname)[0]
if not os.path.isdir(dd):
os.makedirs(dd)
fname2 = dd + "/" + os.path.basename(fullname)
if not os.path.isfile(fname2):
try:
fh2 = open(fname2, "w")
while True:
buff = fh.read()
if len(buff) == 0: break
fh2.write(buff)
fh.close()
fh2.close()
except:
print_exception("Cannot load image", fname)
return
pixbuf = gtk.gdk.pixbuf_new_from_file(fname2)
return pixbuf
def parse_opf(fname):
global conf
ret = ("", "")
try:
fh = zf.open(fname)
ht = OpfHTMLParser(content2cb)
conf.ht = ht
while 1:
sss = fh.readline()
if sss == "": break
if conf.header:
print sss,
ht.feed(sss)
if conf.header:
print sss,
if not conf.nogui:
mw.add_text(sss, True)
ret = (ht.title, ht.auth)
except:
print_exception("Cannot parse .opf file")
return ret
def speak(self, cstr):
fname = self.conf.data_dir + "/festival.txt"
try:
fh = open(fname, "w")
fh.write(cstr)
except:
pubutil.print_exception("Cannot create festival file")
return
try:
if FESTIVAL:
self.speech_pid = subprocess.Popen(
["festival", "--tts", fname])
else:
# This is more configurable, add option as it is displayed below.
#self.speech_pid = subprocess.Popen(["espeak", "-f", fname, "-s", "100"])
self.speech_pid = subprocess.Popen(
["espeak", "-f", fname, "-v", "english-us"])
except:
print "Cannot start TTS", sys.exc_info()
#print "started", self.speech_pid.pid
gobject.timeout_add(100, self.check_speak)
def getimagecb(fname):
global gzf, conf
print "Image callback", "'" + fname + "'"
fh = None
try:
fh = gzf.open(fname)
except:
#sys.exc_info()
#print "searching for", "'" + fname + "'"
# Search for it ...
for aa in zf.infolist():
#print "search", aa.filename, fname
if os.path.basename(aa.filename) == fname:
#print "found:", "'" + aa.filename + "'"
try:
fh = gzf.open(aa.filename)
except:
print_exception("No image File")
return
break
# Extract image file
fname2 = conf.data_dir + "/" + fname
try:
fh2 = open(fname2, "w")
while True:
buff = fh.read()
if len(buff) == 0: break
fh2.write(buff)
fh.close()
fh2.close()
except:
print_exception("Cannot load image", fname)
return
pixbuf = gtk.gdk.pixbuf_new_from_file(fname2)
return pixbuf
def main(zf, fname=None):
global mw, conf
if not conf.nogui:
mw = pubdisp.PubView(conf)
mw.fname = ""
mw.callback = openHTMLcb
mw.bscallback = backcb
if conf.nogui:
reader_tick()
else:
if zf:
gobject.timeout_add(100, reader_tick)
else:
gobject.timeout_add(100, html_tick, fname)
if not conf.nogui:
try:
gtk.main()
except:
print_exception("gtk_main")
def sel_prev(self, dir=False):
#print "sel_prev", self.lastsel
cnt = 0
try:
for aa in self.stags:
if aa == self.lastsel and cnt > 0:
#print "got:", self.xtags[cnt], self.stags[cnt]
#print "prev:", self.xtags[cnt-1], self.stags[cnt-1]
# Move tree selection
sel = self.treeview.get_selection()
xmodel, xiter = sel.get_selected()
iter = xmodel.get_iter_root()
old = None
while 1:
if not iter: break
if xmodel.get_path(iter) == xmodel.get_path(xiter):
break
old = iter
iter = xmodel.iter_next(iter)
if old:
sel.select_iter(old)
self.lastsel = xmodel.get_value(old, 0)
# Move in buffer
mm = self.buffer_1.get_mark(self.xtags[cnt - 1])
if mm:
# Internal tag
self.view.scroll_to_mark(mm, 0.0, True, 0, 0)
else:
# This is an external link
self.callback(self.xtags[cnt - 1])
self.view.grab_focus()
self.gohome()
break
cnt += 1
except:
pubutil.print_exception("sel_prev")
pass
def size_tick():
#print "size_tick"
try:
mw.apply_size()
except:
print_exception("Reader tick")
args = conf.comline(sys.argv[1:])
if len(args) < 1:
print "Usage: pyepub.py ebookfile"
sys.exit(1)
basefile = os.path.basename(args[0])
basefile = os.path.splitext(basefile)[0]
basefile = basefile.translate(string.maketrans(" {}()", "_____"))
conf.data_dir = os.path.expanduser("~/.pyepub")
try: os.mkdir(conf.data_dir)
except: pass
try:
conf.fname = args[0]
zf = zipfile.ZipFile(args[0])
except:
zf = None
print_exception("Cannot open zipfile")
gzf = zf
main(zf)
def openHTML(zf, fname, mark=True):
global conf
found = False
fh = None
spl = []
ht = None
conf.recurse = False
# Is it a link with a tag?
spl = fname.split("#")
# Not loaded, load
if mw.fname != spl[0]:
mw.clearall()
found = False
fh = None
usleep(1)
try:
fh = zf.open(spl[0])
found = True
except:
for aa in zf.infolist():
if os.path.basename(aa.filename) == os.path.basename(spl[0]):
#print "found", aa.filename
try:
fh = zf.open(aa.filename)
except:
print_exception("No Zip File")
return
found = True
break
if not found: return False
try:
ht = pubhtml.HTML_Parser(conf)
ht.mw = mw
ht.mark = mark
ht.fh = fh
ht.getimagecb = getimagecb
ln = 0
mw.waitcursor(True)
while 1:
if conf.recurse:
break
if mw.stopload:
break
sss = fh.readline()
if mw.stopload:
break
if ln % 10 == 0:
mw.prog.set_text("Reading line: %d" % ln)
usleep(1)
ln += 1
if sss == "": break
if conf.recurse:
break
if ht.feed(sss):
break
mw.add_text(sss, True)
mw.fname = spl[0]
mw.waitcursor(False)
except:
print_exception("Parsing HTML")
#if mw.stopload:
# return
mw.prog.set_text("Done Loading.")
usleep(10)
if mw.stopload:
return
# Jump to tag, if any
if len(spl) > 1:
#print "Jumping to", "'" + spl[1] + "'"
mm = mw.buffer_1.get_mark(spl[1])
if mm:
ii = mw.buffer_1.get_iter_at_mark(mm)
mw.buffer_1.place_cursor(ii)
mw.view.scroll_to_mark(mm, 0.0, True, 0, 0)
else:
mw.gohome()
return True
def readepub(zf):
global mw, conf
names = []
allnames = []
if not conf.nogui:
mw.waitcursor(True)
if conf.extract:
print "Extracting epub to:", conf.data_dir + "/" + conf.basefile + "/"
for aa in zf.infolist():
if conf.list:
print aa.filename, aa.file_size
# Extract
if conf.extract:
try:
dd = conf.data_dir + "/" + conf.basefile + "/" + os.path.split(
aa.filename)[0]
if not os.path.isdir(dd):
os.makedirs(dd)
fh = zf.open(aa.filename)
fff = dd + "/" + os.path.basename(aa.filename)
# Only extract once:
if not os.path.isfile(fff):
if not conf.nogui:
tt = "Extracting '" + aa.filename + "'"
mw.prog.set_text(tt[-18])
usleep(1)
fh2 = open(fff, "w+")
while 1:
sss = fh.readline()
if sss == "": break
fh2.write(sss)
fh2.close()
except:
print_exception("Cannot extract file")
if os.path.splitext(aa.filename)[1].find("htm") >= 0:
names.append(aa.filename)
allnames.append(aa.filename)
found = False
ret1 = None
ret2 = None
'''rootfile = locateroot()
print "rootfile:", rootfile
if rootfile != "":
found = True
if ".ncx" in rootfile:
ret1 = parse_ncs(rootfile)
if ".opf" in rootfile:
ret2 = parse_opf(rootfile)'''
# Hack: to simplify parsing we search for an .ncx and .opf file
if not found:
for aa in allnames:
if ".ncx" in aa:
#print "Found toc", aa
#found = True
ret1 = parse_ncx(aa)
break
if not conf.nogui:
mw.update_tree(" --------- ", "")
# Fallback
if not found:
for aa in allnames:
if ".opf" in aa:
#print "Found opf", aa
found = True
ret2 = parse_opf(aa)
break
# Pick from the two optional places
auth = ret1[0]
try:
if auth == "": auth = ret2[0]
except:
pass
title = ret1[1]
try:
if title == "": title = ret2[1]
except:
pass
if conf.title:
print "'" + auth + "' '" + title + "'"
if not conf.nogui:
old = mw.get_title()
mw.set_title(old + " '" + auth + "' '" + title + "'")
if conf.nogui:
return
# No TOC Open and parse all the [x]htm[l] files
if not found:
nnn = "000"
mw.update_tree("Start of Book", nnn)
mw.add_text_mark(nnn)
for bbb in names:
openHTML(zf, bbb)
nnn = "999"
mw.update_tree("End", nnn)
mw.add_text_mark(nnn)
mw.waitcursor(False)
def readepub(zf):
global mw, conf
names = []
allnames = []
if not conf.nogui:
mw.waitcursor(True)
if conf.extract:
print "Extracting epub to:", conf.data_dir + "/" + conf.basefile + "/"
# Locate root meta data
try:
md = zf.open("META-INF/container.xml")
if conf.rootf:
print md.read()
except:
print "No meta data, guessing"
for aa in zf.infolist():
if conf.list:
print aa.filename, aa.file_size
# Extract
if conf.extract:
try:
dd = conf.data_dir + "/" + conf.basefile + "/" + os.path.split(
aa.filename)[0]
if not os.path.isdir(dd):
os.makedirs(dd)
fh = zf.open(aa.filename)
fff = dd + "/" + os.path.basename(aa.filename)
# Only extract once:
if not os.path.isfile(fff):
if not conf.nogui:
tt = "Extracting '" + aa.filename + "'"
mw.prog.set_text(tt[-18])
usleep(1)
fh2 = open(fff, "w+")
while 1:
sss = fh.readline()
if sss == "": break
fh2.write(sss)
fh2.close()
except:
print_exception("Cannot extract file")
if os.path.splitext(aa.filename)[1].find("htm") >= 0:
names.append(aa.filename)
allnames.append(aa.filename)
found = False
ht = None
ht2 = None
for aa in allnames:
if "toc.ncx" in aa:
#print "Found toc", aa
found = True
fh = zf.open(aa)
ht = TocHTMLParser(contentcb)
conf.ht = ht
while 1:
sss = fh.readline()
if sss == "": break
ht.feed(sss)
if conf.header:
print sss,
if not conf.nogui:
mw.add_text(sss, True)
break
if 1:
for aa in allnames:
if "content.opf" in aa:
if not conf.nogui:
mw.update_tree("Legacy Entries:", [])
#print "Found content", aa
found = True
fh = zf.open(aa)
ht2 = ConHTMLParser(content2cb)
while 1:
sss = fh.readline()
if sss == "": break
if conf.header:
print sss,
ht2.feed(sss)
if conf.header:
print sss,
if not conf.nogui:
mw.add_text(sss, True)
break
auth = ""
title = ""
try:
auth = ht.auth
except:
pass
try:
if auth == "": auth = ht2.auth
except:
pass
try:
title = ht.title
except:
pass
try:
if title == "": title = ht2.title
except:
pass
if conf.title:
print "'" + auth + "' '" + title + "'"
if not conf.nogui:
old = mw.get_title()
mw.set_title(old + " '" + auth + "' '" + title + "'")
if conf.nogui:
return
# No TOC Open and parse all the [x]htm[l] files
if not found:
nnn = "000"
mw.update_tree("Start of Book", nnn)
mw.add_text_mark(nnn)
for bbb in names:
openHTML(zf, bbb)
nnn = "999"
mw.update_tree("End", nnn)
mw.add_text_mark(nnn)
mw.waitcursor(False)
return ht
def openHTML(zf, fname, mark=True):
global conf
# Is it a tagged link?
ppp = fname.find("#")
spl = []
if spl >= 0:
spl = fname.split("#")
fname = spl[0]
# Not loaded, load
if mw.fname != fname:
#print "loading", fname
mw.clear()
mw.clear(True)
found = False
fh = None
try:
fh = zf.open(fname)
found = True
except:
#print "searching for", fname, sys.exc_info()
# Search for it ...
for aa in zf.infolist():
#print aa.filename, fname
if os.path.basename(aa.filename) == fname:
try:
fh = zf.open(aa.filename)
except:
print_exception("No Zip File")
return
found = True
break
if not found: return False
try:
ht = pubhtml.HTML_Parser(conf)
ht.mw = mw
ht.mark = mark
ht.fh = fh
ht.getimagecb = getimagecb
ln = 0
mw.waitcursor(True)
while 1:
sss = fh.readline()
if ln % 10 == 0:
mw.prog.set_text("Reading line: %d" % ln)
usleep(1)
ln += 1
if sss == "": break
if ht.feed(sss):
break
mw.add_text(sss, True)
mw.fname = fname
mw.waitcursor(False)
except:
print_exception("Parsing HTML")
mw.prog.set_text("Done Reading.")
usleep(10)
# Jump to tag, if any
if len(spl) > 1:
#print "Jumping to", "'" + spl[1] + "'"
mm = mw.buffer_1.get_mark(spl[1])
ii = mw.buffer_1.get_iter_at_mark(mm)
if mm:
mw.view.scroll_to_mark(mm, 0.0, True, 0, 0)
else:
mw.gohome()
return True
