With cheapglk, transcript created in ISO8859 instead of utf-8 despite -u

[stormi]

This may be another obvious question, but I'm using glulxe+cheapglk+RegTest, and all is well, however when I tried to simplify my test writing by starting with a transcript produced by glulxe -u mygame.ulx, I discovered that despite the console output is utf8, the produced transcript is encoded in ISO8859.

Is it the expected behaviour?

[erkyrath]

This is a problem with the library/kit code, which calls glk_stream_open_file() rather than glk_stream_open_file_uni().

I thought this was planned to be updated in the next Inform release, but I'm not sure. I'll check.

[stormi]

It's possible I'm using too old a version of the compiler or the libs.

[erkyrath]

No, the current versions still use byte mode for transcripts (which means ISO8859).

I filed https://inform7.atlassian.net/jira/software/c/projects/I7/issues/I7-2516 .

[stormi]

Thanks. It's sad Jira won't even let us read the issue without creating an account first. That's where inform7 issues are managed?