I divorced with OpenIndiana

General discussion for topics related to the FreeBASIC project or its community.
Post Reply
Cretin Ho
Posts: 182
Joined: Feb 04, 2021 13:01

I divorced with OpenIndiana

Post by Cretin Ho »

This also means the Solaris port will be unmaintained, unless the fbc developers continue working on it.

https://illumos.topicbox.com/groups/dis ... his-system
fatman2021
Posts: 215
Joined: Dec 14, 2013 0:43

Re: I divorced with OpenIndiana

Post by fatman2021 »

UPDATE: ICU version 66.1
This update is ABI-incompatible , make sure to recompile any local application against the new libraries.
https://twitter.com/OpenIndiana/status/ ... 9562814466
caseih
Posts: 2157
Joined: Feb 26, 2007 5:32

Re: I divorced with OpenIndiana

Post by caseih »

FreeBASIC does not depend on libICU in any way.
Cretin Ho
Posts: 182
Joined: Feb 04, 2021 13:01

Re: I divorced with OpenIndiana

Post by Cretin Ho »

fatman2021 wrote:
UPDATE: ICU version 66.1
This update is ABI-incompatible , make sure to recompile any local application against the new libraries.
https://twitter.com/OpenIndiana/status/ ... 9562814466
I left because of the stupid memory model of that OS, nothing to do with the thing you mentioned. I always on the latest update of OpenIndiana Hipster.
Cretin Ho
Posts: 182
Joined: Feb 04, 2021 13:01

Re: I divorced with OpenIndiana

Post by Cretin Ho »

caseih wrote:FreeBASIC does not depend on libICU in any way.
I found the same thing when building FBC, too. So how does FBC deal with unicode support?

The only things I found so far are:

viewtopic.php?f=8&t=26856

viewtopic.php?f=8&t=29169
caseih
Posts: 2157
Joined: Feb 26, 2007 5:32

Re: I divorced with OpenIndiana

Post by caseih »

The compiler source code itself directly decodes UTF-8, 16, and 32 encodings when processing a source code file. The runtime itself only deals with ANSI or wide characters in memory. A WSTRING is 4-bytes per character on Linux, 2 bytes per character on Windows. Technically it's UTF-16 on Windows, but I don't think the runtime library has any concept of things like surrogate pairs. As near as I can tell the runtime library has zero support for unicode encoding and decoding, for example when reading and writing data to a file. FB does ship with a utility .bi file that implements UTF-8 encoding and decoding. But generally, speaking, FB has very little intrinsic understanding and support for unicode, much like other classic languages. At least compared to Unicode-native languages like Go, Python, etc.
Post Reply