Dear Photran developers, and users,
It seems that I cannot use refactoring when there is a file in the
workspace containing comments or strings with accented characters é
(i.e. \'{e}) etc...)
I could not use the refactorings I tried (extract procedure and
introduce implicit none) in a project because there were all the times
out of bound errors or null character error in java and after that
windows popping every time I made a modification in one of the files.
I therefore tried to create a new workspace with only one subroutine
containing characters with accent and removed all the declarations. The
"introduce implicit non" refactoring did not work.
Then I erased all the comments with non ASCII characters and the
refactoring *worked*.
I reintroduced an accented character of each kind and it did not work.
However, when I tried to erase them once more and retry I got messages
begining by java.lang.ArrayIndexOutofBoundsException and
org.eclipse.runtime.CoreException
After disabling the refactorings, closing the workspace, enabling again
the refactorings and trying "introduce implicit none" I get
"An internal error occurred during: "Photran indexer".
java.lang.ArrayIndexOutOfBoundsException: 4095"
Am I right to suspect that non ASCII characters (and more specifically
letters with accents) are a problem for photran refactorings ?
Did someone else have the same experience ?
--
Sincerely yours,
Michel DEVEL
|