fix umlaut handling

identify.py

@@ -13,6 +13,7 @@ The information stored for the fields are:
 
 # todo: Button Varianten sind Off, On, sollte anpassbar/ automatisch
 # angepasst werden
+# todo: disable focus on pdf  - apparently not simple :(
 
 import argparse
 import subprocess as cmd
@@ -34,8 +35,13 @@ class FormField():
         standard value (["Stdvalue"]) todo: how to implement that?
         something about a special treatment (["Special"])
 
+    Constants:
+        LISTINFOS: those infos can appear several times, hence store a list.
+
     """
 
+    LISTINFOS = ["FieldStateOption"]
+
     def convertToBool(self, possBool):
         """Convert value to True or or False if this field is a Checkbox.
 
@@ -62,12 +68,34 @@ class FormField():
             # maybe something more intelligent necessary
             return possBool
 
+    @staticmethod
+    def convertUmlauts(value):
+        """Convert all html-Syntax-Umlauts to the correct symbols.
+
+        Necessary to make fdfgen generate files that are read correctly
+        by pdftk.
+
+        Returns:
+            value, &#number replaced by the correct unicode
+
+        """
+        for entity, uni in [("Ä", "Ä"),
+                            ("Ö", "Ö"),
+                            ("Ü", "Ü"),
+                            ("ß", "ß"),
+                            ("ä", "ä"),
+                            ("ö", "ö"),
+                            ("ü", "ü")]:
+            value = value.replace(entity, uni)
+        return value
+
     def __init__(self):
         """
         Initialise a FormField.
 
         With an empty dictionary for pdfinfo
         and None for the other values.
+
         """
         self.__pdfinfo = {}
 
@@ -204,6 +232,7 @@ class FormField():
                     """Show a special message for buttons.
 
                     Explains Button options.
+
                     """
                     if self["FieldType"] == "Button":
                         print("It is a butten.",
@@ -363,18 +392,23 @@ def listFields(pdf_file, fieldListFile):  # NOQA
             field = FormField()  # start new Field
         else:
             line = line.split(": ", maxsplit=1)
-            if len(line) != 2:
+            try:
+                line[1] = FormField.convertUmlauts(line[1])
+            except IndexError:
                 raise ValueError("The line '" + str(line) +
                                  "' cannot be parsed. " +
                                  "Apparently there is no ': '.")
             if line[0] in field:  # several entries
-                try:  # works if it is already a list
+                if line[0] in FormField.LISTINFOS:
                     field[line[0]].append(line[1])
-                except AttributeError:  # append does not exist for str :)
-                    field[line[0]] = [field[line[0]], line[1]]
-                    # include existing info
-            else:  # information is new
-                field[line[0]] = line[1]
+                else:
+                    raise ValueError("The information " + line[0] +
+                                     "appeared twice in one field.")
+            else:
+                if line[0] in FormField.LISTINFOS:
+                    field[line[0]] = [line[1]]
+                else:
+                    field[line[0]] = line[1]
 
     return fields