Full UTF-8 awareness (close #74)

Tested only on Windows. But he had the biggest problems with it.

kcc.py

@@ -77,7 +77,7 @@ elif sys.platform.startswith('win'):
 
 # Implementing detection of already running KCC instance and forwarding argv to it
 class QApplicationMessaging(QtWidgets.QApplication):
-    messageFromOtherInstance = QtCore.pyqtSignal(str)
+    messageFromOtherInstance = QtCore.pyqtSignal(bytes)
 
     def __init__(self, argv):
         QtWidgets.QApplication.__init__(self, argv)
@@ -102,7 +102,7 @@ class QApplicationMessaging(QtWidgets.QApplication):
     def handleMessage(self):
         socket = self._server.nextPendingConnection()
         if socket.waitForReadyRead(self._timeout):
-            self.messageFromOtherInstance.emit(socket.readAll().data().decode('utf8'))
+            self.messageFromOtherInstance.emit(socket.readAll().data())
 
     def sendMessage(self, message):
         if self.isRunning():
@@ -110,7 +110,8 @@ class QApplicationMessaging(QtWidgets.QApplication):
             socket.connectToServer(self._key, QtCore.QIODevice.WriteOnly)
             if not socket.waitForConnected(self._timeout):
                 return False
-            socket.write(message.encode('utf8'))
+            # noinspection PyArgumentList
+            socket.write(bytes(message, 'UTF-8'))
             if not socket.waitForBytesWritten(self._timeout):
                 return False
             socket.disconnectFromServer()
@@ -133,7 +134,7 @@ if __name__ == "__main__":
     KCCAplication = QApplicationMessaging(sys.argv)
     if KCCAplication.isRunning():
         if len(sys.argv) > 1:
-            KCCAplication.sendMessage(sys.argv[1].decode(sys.getfilesystemencoding()))
+            KCCAplication.sendMessage(sys.argv[1])
             sys.exit(0)
         else:
             messageBox = QtWidgets.QMessageBox()
@@ -146,5 +147,5 @@ if __name__ == "__main__":
     KCCWindow = QMainWindowKCC()
     KCCUI = KCC_gui.KCCGUI(KCCAplication, KCCWindow)
     if len(sys.argv) > 1:
-        KCCUI.handleMessage(sys.argv[1].decode(sys.getfilesystemencoding()))
+        KCCUI.handleMessage(sys.argv[1])
     sys.exit(KCCAplication.exec_())

kcc/KCC_gui.py

@@ -129,7 +129,7 @@ class WebServerHandler(BaseHTTPRequestHandler):
                                  '</body>\n'
                                  '</html>\n', 'UTF-8'))
             elif sendReply:
-                outputFile = GUI.completedWork[urllib.parse.unquote(self.path[1:])].decode('utf-8')
+                outputFile = GUI.completedWork[urllib.parse.unquote(self.path[1:])]
                 fp = open(outputFile, 'rb')
                 self.send_response(200)
                 self.send_header('Content-type', mimetype)
@@ -243,8 +243,7 @@ class KindleGenThread(QtCore.QRunnable):
         kindlegenError = ''
         try:
             if os.path.getsize(self.work) < 367001600:
-                output = Popen('kindlegen -locale en "' + self.work.encode(sys.getfilesystemencoding()).decode('utf-8')
+                output = Popen('kindlegen -locale en "' + self.work + '"', stdout=PIPE, stderr=STDOUT, shell=True)
-                               + '"', stdout=PIPE, stderr=STDOUT, shell=True)
                 for line in output.stdout:
                     line = line.decode('utf-8')
                     # ERROR: Generic error
@@ -486,7 +485,7 @@ class WorkerThread(QtCore.QThread):
                                 GUI.progress.content = ''
                                 mobiPath = item.replace('.epub', '.mobi')
                                 os.remove(mobiPath + '_toclean')
-                                GUI.completedWork[os.path.basename(mobiPath).encode('utf-8')] = mobiPath.encode('utf-8')
+                                GUI.completedWork[os.path.basename(mobiPath)] = mobiPath
                             MW.addMessage.emit('Cleaning MOBI files... <b>Done!</b>', 'info', True)
                         else:
                             GUI.progress.content = ''
@@ -516,7 +515,7 @@ class WorkerThread(QtCore.QThread):
                                                False)
                 else:
                     for item in outputPath:
-                        GUI.completedWork[os.path.basename(item).encode('utf-8')] = item.encode('utf-8')
+                        GUI.completedWork[os.path.basename(item)] = item
         GUI.progress.content = ''
         GUI.progress.stop()
         MW.hideProgressBar.emit()

kcc/cbxarchive.py

@@ -21,7 +21,6 @@ __docformat__ = 'restructuredtext en'
 
 import os
 import zipfile
-import locale
 from subprocess import STDOUT, PIPE
 from psutil import Popen
 from shutil import move
@@ -59,7 +58,7 @@ class CBxArchive:
         cbzFile.extractall(targetdir, filelist)
 
     def extractCBR(self, targetdir):
-        cbrFile = rarfile.RarFile(self.origFileName.encode(locale.getpreferredencoding()))
+        cbrFile = rarfile.RarFile(self.origFileName)
         filelist = []
         for f in cbrFile.namelist():
             if f.startswith('__MACOSX') or f.endswith('.DS_Store') or f.endswith('thumbs.db'):
@@ -70,16 +69,15 @@ class CBxArchive:
                 except Exception:
                     pass  # the dir exists so we are going to extract the images only.
             else:
-                filelist.append(f.encode(locale.getpreferredencoding()))
+                filelist.append(f)
         cbrFile.extractall(targetdir, filelist)
 
     def extractCB7(self, targetdir):
-        output = Popen('7za x "' + self.origFileName.encode(locale.getpreferredencoding()) +
+        output = Popen('7za x "' + self.origFileName + '" -xr!__MACOSX -xr!.DS_Store -xr!thumbs.db -o"'
-                       '" -xr!__MACOSX -xr!.DS_Store -xr!thumbs.db -o"' + targetdir + '"',
+                       + targetdir + '"', stdout=PIPE, stderr=STDOUT, shell=True)
-                       stdout=PIPE, stderr=STDOUT, shell=True)
         extracted = False
         for line in output.stdout:
-            if "Everything is Ok" in line:
+            if b"Everything is Ok" in line:
                 extracted = True
         if not extracted:
             raise OSError
@@ -97,6 +95,10 @@ class CBxArchive:
             adir.remove('ComicInfo.xml')
         if len(adir) == 1 and os.path.isdir(os.path.join(targetdir, adir[0])):
             for f in os.listdir(os.path.join(targetdir, adir[0])):
+                # If directory names contain UTF-8 chars shutil.move can't clean up the mess alone
+                if os.path.isdir(os.path.join(targetdir, f)):
+                    os.rename(os.path.join(targetdir, adir[0], f), os.path.join(targetdir, adir[0], f + '-A'))
+                    f += '-A'
                 move(os.path.join(targetdir, adir[0], f), targetdir)
             os.rmdir(os.path.join(targetdir, adir[0]))
         return targetdir

kcc/comic2ebook.py

@@ -78,7 +78,7 @@ def buildHTML(path, imgfile):
         if not os.path.exists(htmlpath):
             os.makedirs(htmlpath)
         htmlfile = os.path.join(htmlpath, filename[0] + '.html')
-        f = open(htmlfile, "w")
+        f = open(htmlfile, "w", encoding='UTF-8')
         f.writelines(["<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.1//EN\" ",
                       "\"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd\">\n",
                       "<html xmlns=\"http://www.w3.org/1999/xhtml\">\n",
@@ -189,9 +189,8 @@ def buildHTML(path, imgfile):
 
 def buildNCX(dstdir, title, chapters):
     options.uuid = str(uuid4())
-    #options.uuid = options.uuid.encode('utf-8')
     ncxfile = os.path.join(dstdir, 'OEBPS', 'toc.ncx')
-    f = open(ncxfile, "w")
+    f = open(ncxfile, "w", encoding='UTF-8')
     f.writelines(["<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n",
                   "<!DOCTYPE ncx PUBLIC \"-//NISO//DTD ncx 2005-1//EN\" ",
                   "\"http://www.daisy.org/z3986/2005/ncx-2005-1.dtd\">\n",
@@ -227,7 +226,7 @@ def buildOPF(dstdir, title, filelist, cover=None):
         writingmode = "horizontal-rl"
     else:
         writingmode = "horizontal-lr"
-    f = open(opffile, "w")
+    f = open(opffile, "w", encoding='UTF-8')
     f.writelines(["<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n",
                   "<package version=\"2.0\" unique-identifier=\"BookID\" ",
                   "xmlns=\"http://www.idpf.org/2007/opf\">\n",
@@ -284,7 +283,7 @@ def buildOPF(dstdir, title, filelist, cover=None):
     f.write("</spine>\n<guide>\n</guide>\n</package>\n")
     f.close()
     os.mkdir(os.path.join(dstdir, 'META-INF'))
-    f = open(os.path.join(dstdir, 'META-INF', 'container.xml'), 'w')
+    f = open(os.path.join(dstdir, 'META-INF', 'container.xml'), 'w', encoding='UTF-8')
     f.writelines(["<?xml version=\"1.0\"?>\n",
                   "<container version=\"1.0\" xmlns=\"urn:oasis:names:tc:opendocument:xmlns:container\">\n",
                   "<rootfiles>\n",
@@ -426,7 +425,7 @@ def genEpubStruct(path):
     cover = None
     _, deviceres, _, _, panelviewsize = options.profileData
     os.mkdir(os.path.join(path, 'OEBPS', 'Text'))
-    f = open(os.path.join(path, 'OEBPS', 'Text', 'style.css'), 'w')
+    f = open(os.path.join(path, 'OEBPS', 'Text', 'style.css'), 'w', encoding='UTF-8')
     # DON'T COMPRESS CSS. KINDLE WILL FAIL TO PARSE IT.
     # Generic Panel View support + Margins fix for Non-Kindle devices.
     f.writelines(["@page {\n",