Documentation error: wrong term for font encoding in usrguide.pdf

[gmilde]

Brief outline of the bug

The new version of usrguide.pdf adds documentation of the case changing commands. Following sentence is not fully correct:

The full range of Unicode UTF-8 input can be supported, with the proviso that at present the characters set up with 8-bit engines match those available in standard input encodings (T1, T2, LGR).

• T1, T2, and LGR are font encodings, not input encodings.
• LGR it not a standard font encoding. (The "encguide" reserves the term "standard font encoding" and provides a strict definitions of compliance.)

Suggestion: change to

The full range of Unicode UTF-8 input can be supported, with the proviso that at present the characters set up with 8-bit engines match those available in the font encodings T1, T2, or LGR (cf. encguide.pdf) .

Maybe explain for users that T1, T2, and LGR comprise characters for the Latin, Cyrillic, and Greek script. Caveat: I did not test, which subset of Cyrillic (T2A, T2B, T2C, X2) is actually supported.

[FrankMittelbach]

correct is that all characters from any "font encoding" (standard or otherwise :-)) that has a corresponding .dfu file is supported (provided the font encoding is actually loaded in the document).

[josephwright]

I recall now that the switch to full-Unicode for 8-bit engines means that at the input side there's (almost) no restriction, so I've adjusted again to remove the caveat line.

[gmilde]

I can confirm, that upcasing Greek literals works fine also with the iso-8859-7 8-bit encoding (where lgrenc.dfu is not loaded).