search

tphinney / font-tools

Font-related tools such as FontLab Studio Python scripts, FontLab .enc files for character sets, etc.
Apache License 2.0
31 stars 5 forks source link

Possible error in AC-3-exp.enc #4

Open LibertyType opened 9 years ago

LibertyType commented 9 years ago

Thomas: One of the unencoded entries in AC-3-exp.enc actually has a Unicode value. uni0438_uni0300 == uni045D, iigrave-cy. The only noticeable difference is the naming. uni0438_uni0300 translates as ii_gravecomb-cy using OS X Services. The same situation exists for the cap version, uni0418_uni0300 == uni040D, Iigrave-cy.

I'm wondering why this duplication exists.

Thanks, George Thomas

tphinney commented 9 years ago

Sounds like a bug! I'll look into it.

On Tue, Nov 17, 2015 at 8:03 PM, George Thomas notifications@github.com wrote:

Thomas: One of the unencoded entries in AC-3-exp.enc actually has a Unicode value. uni0438_uni0300 == uni045D, iigrave-cy. The only noticeable difference is the naming. uni0438_uni0300 translates as ii_gravecomb-cy using OS X Services. The same situation exists for the cap version, uni0418_uni0300 == uni040D, Iigrave-cy.

I'm wondering why this duplication exists.

Thanks, George Thomas

— Reply to this email directly or view it on GitHub https://github.com/tphinney/font-tools/issues/4.

“If friendship is magic, then any sufficiently advanced technology is indistinguishable from friendship.” —Shanna Germain (after Arthur C. Clarke)

tphinney commented 9 years ago

Yes, this is a bug in the source data I got from Adobe. I've alerted them to the bug, and will adjust my files appropriately.

(Although, as they use combining marks for non-Unicode combinations, it's not clear the bug would really hurt them any.)

On Wed, Nov 18, 2015 at 8:15 AM, Thomas Phinney tphinney@cal.berkeley.edu wrote:

Sounds like a bug! I'll look into it.

On Tue, Nov 17, 2015 at 8:03 PM, George Thomas notifications@github.com wrote:

Thomas: One of the unencoded entries in AC-3-exp.enc actually has a Unicode value. uni0438_uni0300 == uni045D, iigrave-cy. The only noticeable difference is the naming. uni0438_uni0300 translates as ii_gravecomb-cy using OS X Services. The same situation exists for the cap version, uni0418_uni0300 == uni040D, Iigrave-cy.

I'm wondering why this duplication exists.

Thanks, George Thomas

— Reply to this email directly or view it on GitHub https://github.com/tphinney/font-tools/issues/4.

“If friendship is magic, then any sufficiently advanced technology is indistinguishable from friendship.” —Shanna Germain (after Arthur C. Clarke)

“If friendship is magic, then any sufficiently advanced technology is indistinguishable from friendship.” —Shanna Germain (after Arthur C. Clarke)

LibertyType commented 9 years ago

Thomas, I found one more in AC-3-exp.enc.

uni0415_uni0300, Ie_gravecomb-cy == Iegrave-cy, uni0400 and the lowercase, uni0435_uni0300, ie_gravecomb-cy == iegrave-cy, uni0450.

It is kind of odd, but as you said, it probably wouldn't cause any problems.

I also spot-checked the Latin 5 unencoded but found nothing.

Best, George Thomas

tphinney commented 9 years ago

Thanks!

I've updated the file accordingly.

On Wed, Nov 18, 2015 at 3:42 PM, George Thomas notifications@github.com wrote:

Thomas, I found one more in AC-3-exp.enc.

uni0415_uni0300, Ie_gravecomb-cy == Iegrave-cy, uni0400 and the lowercase, uni0435_uni0300, ie_gravecomb-cy == iegrave-cy, uni0450.

It is kind of odd, but as you said, it probably wouldn't cause any problems.

I also spot-checked the Latin 5 unencoded but found nothing.

Best, George Thomas

— Reply to this email directly or view it on GitHub https://github.com/tphinney/font-tools/issues/4#issuecomment-157900100.

“If friendship is magic, then any sufficiently advanced technology is indistinguishable from friendship.” —Shanna Germain (after Arthur C. Clarke)