tested grammar for reading entries (parsing quite slow)

code/grammar_entries.lark

@@ -1,15 +1,21 @@
-start: title description property+ _E note? _E? building
+start: title description property+ (_E note)? _E building
 
-title: "# " /(?! ).+(?<! )/ "\n" _E                 // not starting or ending with a space
+title: "#" /(?! ).+(?<! )/ "\n" _E                 // not starting or ending with a space
 
 description: "_" /(?! ).+(?<![ _])/ "_\n" _E        // single line not ending with underscore
 
-property: "- " _key ": " _value "\n"                // a property on a single line "- key: value"
-_key: /(?! ).+?(?=:)/                               // key: everything until next ":"
-_value: /.+(?<! )/                                  // everything until the end of the line
-
-building: "## Building\n" _E property+ _E? note     // the "building" section
-
 note: /(?![\-#]).*\n/*                              // Unstructured text, not starting with - or #
 
-_E: /^$\n/m                                         // empty new line
+building: "## Building\n" (_E property+)? (_E note)?     // the "building" section
+
+property: "-" _key ":" _values "\n"?                // a property on a single line "- key: value"
+_key: /(?! ).+?(?=:)(?<! )/                         // key: everything until next ":", not beginning or ending with a space
+_values: [_value ("," _value)*]                     // a comma separated list
+_value: quoted_value | unquoted_value               // quoted or unquoted values
+quoted_value: /\".+?\"/                             // with quotation marks, can contain commas
+unquoted_value: /(?![ \"])[^,\n]+(?<![ ])/        // cannot contain commas, cannot start or end with quotation mark
+
+_E: /^$\n/m                                         // empty new line
+
+%import common.WS
+%ignore WS

code/maintenance_collect_inspiration_infos.py

@@ -28,11 +28,12 @@ if __name__ == "__main__":
     print('{} inspirations in the inspirations database'.format(len(inspirations)))
     osg.write_inspirations_info(inspirations)  # write again just to check integrity
 
-    osg_ui.run_simple_button_app('Maintenance inspirations', (('Duplicate check', duplicate_check),))
+    #osg_ui.run_simple_button_app('Maintenance inspirations', (('Duplicate check', duplicate_check),))
 
 
 
     # assemble info
+    entries = osg.read_entries()
     entries = osg.assemble_infos()
 
     # assemble inspirations info from entries

code/utils/osg.py

@@ -58,11 +58,19 @@ class ListingTransformer(lark.Transformer):
 # transformer
 class EntryTransformer(lark.Transformer):
 
-    def start(self, x):
-        d = {}
-        for key, value in x:
-            d[key] = value
-        return d
+    def unquoted_value(self, x):
+        return x[0].value
+
+    def quoted_value(self, x):
+        return x[0].value[1:-1]  # remove quotation marks
+
+    def property(self, x):
+        """
+        The key of a property will be converted to lower case and the value part is the second part
+        :param x:
+        :return:
+        """
+        return x[0].lower(), x[1:]
 
     def title(self, x):
         return 'title', x[0].value
@@ -70,18 +78,32 @@ class EntryTransformer(lark.Transformer):
     def description(self, x):
         return 'description', x[0].value
 
-    def property(self, x):
-        return str.casefold(x[0].value), x[1].value
-
     def note(self, x):
+        """
+        Optional
+        :param x:
+        :return:
+        """
+        if not x:
+            raise lark.Discard
         return 'note', x[0].value
 
     def building(self, x):
         d = {}
         for key, value in x:
+            if key in d:
+                raise RuntimeError('Key in entry appears twice')
             d[key] = value
         return 'building', d
 
+    def start(self, x):
+        d = {}
+        for key, value in x:
+            if key in d:
+                raise RuntimeError('Key in entry appears twice')
+            d[key] = value
+        return d
+
 
 regex_sanitize_name = re.compile(r"[^A-Za-z 0-9-+]+")
 regex_sanitize_name_space_eater = re.compile(r" +")
@@ -535,6 +557,44 @@ def write_inspirations_info(inspirations):
     utils.write_text(inspirations_file, content)
 
 
+def read_entries():
+    """
+    Parses all entries and assembles interesting infos about them.
+    """
+
+    # setup parser
+    grammar_file = os.path.join(code_path, 'grammar_entries.lark')
+    grammar = utils.read_text(grammar_file)
+    parser = lark.Lark(grammar, debug=False)
+
+    # setup transformer
+    transformer = EntryTransformer()
+
+    # a database of all important infos about the entries
+    entries = []
+
+    # iterate over all entries
+    for file, _, content in entry_iterator():
+
+        print(file)
+
+        # parse and transform entry content
+        try:
+            tree = parser.parse(content)
+            entry = transformer.transform(tree)
+        except Exception as e:
+            print(e)
+            continue
+
+        # add file information
+        entry['file'] = file
+
+        # add to list
+        entries.append(entry)
+
+    return entries
+
+
 def compare_entries_developers(entries, developers):
     """
     Cross checks the game entries lists and the developers lists.