I changed "Character" to "Codepoint" in some LazUTF8 function names in r56692. Please take a look.
"Codepoint" is more accurate and hopefully reduces confusion. It was discussed in forum and I agree with the importance of using correct terms.
http://forum.lazarus.freepascal.org/index.php/topic,39045.msg268541.html#msg268541 With Unicode the term "Character" is very ambiguous.

Existing code that uses those functions now throws "deprecated" warnings. Sorry for the inconvenience.

Juha

Juha Manninen wrote

I changed "Character" to "Codepoint" in some LazUTF8 function names in r56692. Please take a look. "Codepoint" is more accurate and hopefully reduces confusion.

I haven't looked at the code changes, but fully agree with the sentiment. Getting the terminology correct is already a great leap forward in the right direction.

Regards,
Graeme

Agreed, but I hope nobody tries to retrofit any more IBMisms like "octet" instead of "byte" :-)

Mark Morgan Lloyd wrote

Agreed, but I hope nobody tries to retrofit any more IBMisms like "octet" instead of "byte" :-)

Where did you get the connection for IBM?
AFAIK "codepoint" is not related to IBM but it is an essential part of Unicode.

Juha

Juha Manninen wrote

On Tue, Dec 12, 2017 at 11:51 AM, Mark Morgan Lloyd via Lazarus wrote:> Agreed, but I hope nobody tries to retrofit any more IBMisms like "octet"> instead of "byte" :-) Where did you get the connection for IBM?AFAIK "codepoint" is not related to IBM but it is an essential part of Unicode.

I think it was IBM introduced the term no later than the 80s when people started using codepages. It might have been related to CUA, and it was used in the context of both ASCII and EBCDIC.

Mark Morgan Lloyd wrote

I think it was IBM introduced the term no later than the 80s when people started using codepages. It might have been related to CUA, and it was used in the context of both ASCII and EBCDIC.

Ok, I learn new things every day. :)

Juha

Juha Manninen wrote

On Tue, Dec 12, 2017 at 5:10 PM, Mark Morgan Lloyd via Lazarus wrote:> I think it was IBM introduced the term no later than the 80s when people> started using codepages. It might have been related to CUA, and it was used> in the context of both ASCII and EBCDIC. Ok, I learn new things every day. :)

:-) It might go back at least as far as IBM's GDDM from 1979. In IBM's context, I think a codepoint would have represented a numeric value which was translated by lookup in a (single- or double-byte) codepage to a glyph, with the obvious gross differences depending on whether the codepage family was ASCII or EBCDIC. I think Unicode usage is fairly consistent with that, assuming that codepage lookup was brushed under the carpet as a manufacturer-specific detail.

Mark Morgan Lloyd wrote

It might go back at least as far as IBM's GDDM from 1979.

It's a real shame that generation of extremely talented developers disappeared without anybody of equal talent taking their place. IBM used to be a software power-house. Or maybe IBM is simply not that big in writing software I deal with any more - these days they might only stick to mainframe software only (which I have never dealt with).

Regards,
Graeme