Handle inproper \*\csN in body without braces

src/calibre/ebooks/rtf2xml/ParseRtf.py

@@ -375,7 +375,7 @@ class ParseRtf:
         old_rtf = old_rtf_obj.check_if_old_rtf()
         if old_rtf:
             if self.__run_level > 5:
-                msg = 'older RTF\n'
+                msg = 'Older RTF\n'
                 msg += 'self.__run_level is "%s"\n' % self.__run_level
                 raise RtfInvalidCodeException, msg
             if self.__run_level > 1:

src/calibre/ebooks/rtf2xml/delete_info.py

@@ -48,6 +48,7 @@ class DeleteInfo:
                             'cw<it<listtable_',
                             'cw<it<revi-table',
                             'cw<ls<list-lev-d',
+                            # Field allowed
                             'cw<fd<field-inst',
                             'cw<an<book-mk-st',
                             'cw<an<book-mk-en',
@@ -86,7 +87,7 @@ class DeleteInfo:
             self.__ob = line
             return False
         else:
-            # write previous bracket, since didn't fine asterisk
+            # write previous bracket, since didn't find asterisk
             if self.__ob:
                 self.__write_obj.write(self.__ob)
                 self.__ob = 0
@@ -109,7 +110,7 @@ class DeleteInfo:
         If you find that you are in a delete group, and the previous
         token in not an open bracket (self.__ob = 0), that means
         that the delete group is nested inside another acceptable
-        detination group. In this case, you have alrady written
+        detination group. In this case, you have already written
         the open bracket, so you will need to write the closed one
         as well.
         """

src/calibre/ebooks/rtf2xml/fields_small.py

@@ -16,7 +16,9 @@
 #                                                                       #
 #########################################################################
 import sys, os, tempfile, re
+
 from calibre.ebooks.rtf2xml import field_strings, copy
+
 class FieldsSmall:
     """
 =================
@@ -24,7 +26,7 @@ Purpose
 =================
 Write tags for bookmarks, index and toc entry fields in a tokenized file.
 This module does not handle toc or index tables.  (This module won't be any
-use to use to you unless you use it as part of the other modules.)
+use to you unless you use it as part of the other modules.)
 -----------
 Method
 -----------
@@ -55,6 +57,7 @@ file.
         self.__copy = copy
         self.__write_to = tempfile.mktemp()
         self.__run_level = run_level
+
     def __initiate_values(self):
         """
         Initiate all values.
@@ -81,6 +84,7 @@ file.
         tx = 'tx<nu<__________<(.*?)'
         reg_st = ob + bk_st + tx + cb
         self.__book_start = re.compile(r'%s' % reg_st)
+
     def __before_body_func(self, line):
         """
         Requires:
@@ -94,6 +98,7 @@ file.
         if self.__token_info == 'mi<mk<body-open_':
             self.__state = 'body'
         self.__write_obj.write(line)
+
     def __body_func(self, line):
         """
         Requires:
@@ -110,6 +115,7 @@ file.
             action(line, tag)
         else:
             self.__write_obj.write(line)
+
     def __found_bookmark_func(self, line, tag):
         """
         Requires:
@@ -125,6 +131,7 @@ file.
         self.__cb_count = 0
         self.__state = 'bookmark'
         self.__type_of_bookmark = tag
+
     def __bookmark_func(self, line):
         """
         Requires:
@@ -153,6 +160,7 @@ file.
             self.__write_obj.write(line)
         elif line[0:2] == 'tx':
             self.__text_string += line[17:-1]
+
     def __parse_index_func(self, my_string):
         """
         Requires:
@@ -201,6 +209,7 @@ file.
             my_changed_string += '<sub-entry>%s' % sub_entry
         my_changed_string += '\n'
         return my_changed_string
+
     def __index_see_func(self, my_string):
         in_see = 0
         bracket_count = 0
@@ -226,6 +235,7 @@ file.
                     in_see = 1
                 changed_string += '%s\n' % line
         return changed_string, see_string
+
     def __index_bookmark_func(self, my_string):
         """
         Requries:
@@ -262,6 +272,7 @@ file.
                     in_bookmark = 1
                 index_string += '%s\n' % line
         return index_string, bookmark_string
+
     def __index__format_func(self, my_string):
         italics = 0
         bold =0
@@ -273,6 +284,7 @@ file.
             if token_info == 'cw<in<index-ital':
                 italics = 1
         return italics, bold
+
     def __parse_toc_func(self, my_string):
         """
         Requires:
@@ -308,6 +320,7 @@ file.
         my_changed_string += '<main-entry>%s' % main_entry
         my_changed_string += '\n'
         return my_changed_string
+
     def __parse_bookmark_for_toc(self, my_string):
         """
         Requires:
@@ -353,6 +366,7 @@ file.
                     in_bookmark = 1
                 toc_string += '%s\n' % line
         return toc_string, book_start_string, book_end_string
+
     def __parse_bookmark_func(self, my_string, type):
         """
         Requires:
@@ -367,6 +381,7 @@ file.
         my_changed_string = ('mi<tg<empty-att_<field<type>%s'
         '<number>%s<update>none\n' % (type, my_string))
         return my_changed_string
+
     def __found_toc_index_func(self, line, tag):
         """
         Requires:
@@ -382,6 +397,7 @@ file.
         self.__cb_count = 0
         self.__state = 'toc_index'
         self.__tag = tag
+
     def __toc_index_func(self, line):
         """
         Requires:
@@ -409,6 +425,7 @@ file.
             self.__write_obj.write(line)
         else:
             self.__text_string += line
+
     def fix_fields(self):
         """
         Requires:
@@ -423,24 +440,19 @@ file.
            bookmark.
         """
         self.__initiate_values()
-        read_obj = open(self.__file)
+        with open(self.__file, 'r') as read_obj:
-        self.__write_obj = open(self.__write_to, 'w')
+            with open(self.__write_to, 'w') as self.__write_obj:
-        line_to_read = '1'
+                for line in read_obj:
-        while line_to_read:
-            line_to_read = read_obj.readline()
-            line = line_to_read
                     self.__token_info = line[:16]
                     if self.__token_info == 'ob<nu<open-brack':
                         self.__ob_count = line[-5:-1]
                     if self.__token_info == 'cb<nu<clos-brack':
                         self.__cb_count = line[-5:-1]
                     action = self.__state_dict.get(self.__state)
-            if action == None:
+                    if action is None:
-                sys.stderr.write('no no matching state in module fields_small.py\n')
+                        sys.stderr.write('No matching state in module fields_small.py\n')
                         sys.stderr.write(self.__state + '\n')
                     action(line)
-        read_obj.close()
-        self.__write_obj.close()
         copy_obj = copy.Copy(bug_handler = self.__bug_handler)
         if self.__copy:
             copy_obj.copy_file(self.__write_to, "fields_small.data")

src/calibre/ebooks/rtf2xml/tokenize.py

@@ -115,8 +115,8 @@ class Tokenize:
 
     def __sub_reg_split(self,input_file):
         input_file = self.__replace_spchar.mreplace(input_file)
-        # this is for older RTF
         input_file = self.__par_exp.sub('\n\\par \n', input_file)
+        input_file = self.__cs_ast.sub("\g<1>", input_file)
         input_file = self.__ms_hex_exp.sub("\\mshex0\g<1> ", input_file)
         input_file = self.__utf_ud.sub("\\{\\uc0 \g<1>\\}", input_file)
         #remove \n in bin data
@@ -172,6 +172,8 @@ class Tokenize:
         self.__splitexp = re.compile(r"(\\[{}]|\n|\\[^\s\\{}&]+(?:[ \t\r\f\v])?)")
         #this is for old RTF
         self.__par_exp = re.compile(r'\\\n+')
+        #handle improper cs char-style with \* before without {
+        self.__cs_ast = re.compile(r'\\\*([\n ]*\\cs\d+[\n \\]+)')
         # self.__par_exp = re.compile(r'\\$')
         #self.__bin_exp = re.compile(r"\\bin(-?\d{1,8}) {0,1}")
         #self.__utf_exp = re.compile(r"^\\u(-?\d{3,6})")