Hi Jose,
Okay, I understand the problem now. In KT2, we left it up to PHP to set the character set, which means that most people probably ended up entering things in ISO8859-1.
Now that we've changed everything to UTF-8, this sort of thing will be a problem. When written to the database in ISO8859-1, and then byte-converted to UTF-8, we're going to have a problem. This is why the folders and documents have different names on the filesystem now.
We're not going to try convert UTF-8 to a specific character set to store things on the filesystem - for one thing it will break all current KT 3.0.0 installs.
Basically, what's needed is a script to:
a) Read every string in the database in, do a character set conversion to UTF-8, and write them back to the database. I imagine we'll focus on document names, folder names, user and group names, and workflow names.
b) Read every file and folder name on the file system, do a character set conversion to UTF-8, and rename the files and folders on the filesystem.
In terms of files and folders, it may be better to convert to the new storage provider that doesn't emulate folder and file names onto the filesystem (to avoid problems encoding the file and folder name to the host operating system's filesystem).
To write and test the script, you will probably need to send us an example KnowledgeTree instance (DB and documents archive) generated from 2.0.2 that exhibits these problems. We can then run the upgrade and then get the script to fix the problem.
You can send that to me directly on
nbm@ktdms.com if you want to keep the contents relatively private.
Regards,
Neil
Please provide us with the information from the "Admin>Miscellanous>Server Information" section.