--- gforth/BUGS	1994/05/07 14:55:44	1.2
+++ gforth/BUGS	1994/08/19 17:47:18	1.6
@@ -1,23 +1,22 @@
-The system quits on exceptions like segmentation faults. anton 5oct93
-
-compile does not have its traditional meaning. anton 5oct93
-
-include does not read the last line if it does not end with \n.  anton 6oct93
-
-in decimal mode aaa is accepted as 1110  anton 4feb94
-
-abort" does something to the control flow stack  anton 2mar94
-
-the compiler does not complain about undefined words; instead, it
-complains about unstructured  anton 2mar94
-
-the compiler tries to reveal anonymous words ( -> redefined complaints )  anton 2mar94
-
-'ansforth "include xxx"' gives a segmentation fault after processing
-the file; No segmentation fault when including from the Forth prompt
-anton 2apr94
-
 name> does not take the same argument as e.g. .name. Remedy: add cell+
 before name>, but adapt all uses.  anton 23apr94
 
+revealing the same name several times (e.g., by using recursive)
+results in redefined messages.  anton 28jul94
+
+[IF] is case-sensitive.  anton 2aug94
 
+if blocks.fb does not exist, 1 block creates the file, but cannot
+read-file from it. Only if the file-id has been created with
+open-file, not create-file, read-file works. - anton 6aug94
+
+gforth.el: doing a forth-fill-paragraph on the following piece of code
+uncomments the first comment line. This may be a bug in
+fill-paragraph, but documentation says that a paragraph can start
+without prefix (hanging indentation), so I guess it's all right. -
+anton 19aug94
+
+: bb-shortest-paths recursive { bb bb-from start-path -- }
+    \ compute the
+    \ paths from bb-from through bb, start-path is the path from
+    \ bb-from to bb