[FIXED] Bug recognizing non-assigned code points

Post your bug reports here. Include information that helps us to understand and reproduce the bug.
Post Reply
ClintGoss
Posts: 128
Joined: Mon Nov 06, 2017 2:18 pm
Location: Westport, CT
Contact:

[FIXED] Bug recognizing non-assigned code points

Post by ClintGoss »

I think this is a bug, although I'm pretty new to this stuff ... I'm using FC Pro 11.0.0.2408, which seems to be the latest, on Win7x64.

It appears that a number of the Unicode 10.0.0 blocks have non-assigned characters which are not recognized as such by FCP. In particular (this is cut-pasted from my XLS, so excuse the repetition). I believe that this list is a complete roster of these type of issues:

3400..4DBF; CJK Unified Ideographs Extension A: FCP 11.0 does not identify non-assigned code points in this block of 6,592 characters. The standard calls for only 6,582 characters. The last 10 code points, from U+4DB6 through U+4DBF, are non-assigned code points.

4E00..9FFF; CJK Unified Ideographs: FCP 11.0 does not identify non-assigned code points in this block of 20,992 characters. The standard calls for only 20,971 characters. The last 21 code points, from U+9FEB through U+9FFF, are non-assigned code points.

AC00..D7AF; Hangul Syllables: FCP 11.0 does not identify non-assigned code points in this block of 11,184 characters. The standard calls for only 11,172 characters. The last 12 code points, from U+D7A4 through U+D7AF, are non-assigned code points.

16FE0..16FFF; Ideographic Symbols and Punctuation: FCP 11.0 does not identify non-assigned code points in this block of 32 characters. The standard calls for only 2 characters. The last 30 code points, from U+16FE2 through U+16FFF, are non-assigned code points.

17000..187FF; Tangut: FCP 11.0 does not identify non-assigned code points in this block of 6,144 characters. The standard calls for only 6,125 characters. The last 19 code points, from U+187ED through U+187FF, are non-assigned code points.

20000..2A6DF; CJK Unified Ideographs Extension B: FCP 11.0 does not identify non-assigned code points in this block of 42,720 characters. The standard calls for only 42,711 characters. The last 9 code points, from U+2A6D7 through U+2A6DF, are non-assigned code points.

2A700..2B73F; CJK Unified Ideographs Extension C: FCP 11.0 does not identify non-assigned code points in this block of 4,160 characters. The standard calls for only 4,149 characters. The last 11 code points, from U+2B735 through U+2B73F, are non-assigned code points.

2B740..2B81F; CJK Unified Ideographs Extension D: FCP 11.0 does not identify non-assigned code points in this block of 224 characters. The standard calls for only 222 characters. The last 2 code points, from U+2B81E through U+2B81F, are non-assigned code points.

2B820..2CEAF; CJK Unified Ideographs Extension E: FCP 11.0 does not identify non-assigned code points in this block of 5,776 characters. The standard calls for only 5,762 characters. The last 14 code points, from U+2CEA2 through U+2CEAF, are non-assigned code points.

2CEB0..2EBEF; CJK Unified Ideographs Extension F: FCP 11.0 does not identify non-assigned code points in this block of 7,488 characters. The standard calls for only 7,473 characters. The last 15 code points, from U+2EBE1 through U+2EBEF, are non-assigned code points.

--Clint Goss
Erwin Denissen
Moderator
Moderator
Posts: 11107
Joined: Fri Oct 04, 2002 12:41 am
Location: Bilthoven, The Netherlands
Contact:

Re: Bug recognizing non-assigned code points

Post by Erwin Denissen »

Yes, this is indeed an issue. We'll try to find a solution which we'll try to include in the next upcoming update.
Erwin Denissen
High-Logic
Proven Font Technology
ClintGoss
Posts: 128
Joined: Mon Nov 06, 2017 2:18 pm
Location: Westport, CT
Contact:

Re: Bug recognizing non-assigned code points

Post by ClintGoss »

Thanks Erwin! This is, of course, a pretty low-priority issue, but one that might cause heartburn here and there for a few folks.

I believe that those are the only issues of this type, because with those corrections, the number of total defined Unicode 10.0 code points that are not private use and not surrogates (136,690) shown in FCP would seem to match that same number given by the standard (Core Spec, top of page 3 and also in table D-5, p923) (this is a calc from my spreadsheet of the blocks).

One niggly might be - and this is really beyond my experience level - but the 65 control codes in the first two blocks seem to be treated as assigned code points in the Standard, but not by FCP. I'm guessing that's for a very good reason, and not an oversight.

BTW, *thanks* for a *very* nicely designed application (!!!) I've been using FCP since version 3.1.1.
Erwin Denissen
Moderator
Moderator
Posts: 11107
Joined: Fri Oct 04, 2002 12:41 am
Location: Bilthoven, The Netherlands
Contact:

Re: Bug recognizing non-assigned code points

Post by Erwin Denissen »

ClintGoss wrote: Sun Jan 14, 2018 7:50 pm Thanks Erwin! This is, of course, a pretty low-priority issue, but one that might cause heartburn here and there for a few folks.
We've fixed the issue. The bug fix will be available with the next upcoming release.
ClintGoss wrote: Sun Jan 14, 2018 7:50 pm I believe that those are the only issues of this type, because with those corrections, the number of total defined Unicode 10.0 code points that are not private use and not surrogates (136,690) shown in FCP would seem to match that same number given by the standard (Core Spec, top of page 3 and also in table D-5, p923) (this is a calc from my spreadsheet of the blocks).

One niggly might be - and this is really beyond my experience level - but the 65 control codes in the first two blocks seem to be treated as assigned code points in the Standard, but not by FCP. I'm guessing that's for a very good reason, and not an oversight.
Control characters are not intended to be used within fonts, so in general we try to hide or avoid them.
ClintGoss wrote: Sun Jan 14, 2018 7:50 pm BTW, *thanks* for a *very* nicely designed application (!!!) I've been using FCP since version 3.1.1.
That was over 15 years ago. FontCreator is still the most popular font editor for Windows, so enjoy the software!
Erwin Denissen
High-Logic
Proven Font Technology
Post Reply