Skip to content
This repository has been archived by the owner on Sep 27, 2019. It is now read-only.

tounicode issue #102

Closed
dohyunkim opened this issue Jun 27, 2013 · 3 comments
Closed

tounicode issue #102

dohyunkim opened this issue Jun 27, 2013 · 3 comments

Comments

@dohyunkim
Copy link

\pdfcompresslevel=0
\pdfobjcompresslevel=0
\nopagenumbers
\input luaotfload.sty
\font\tenrm={linux libertine o:+hlig}
\tenrm
ct
\bye

The resulting pdf has wrong tounicode value:

<0953> <630074>

here 630074 should be 00630074. According to the luatex manual, value of tounicode must be UTF-16BE hexadecimal string. But luaotfload treated it as number in this case.

The cache file, linlibertine-r.lua, already has some problem.

[2387]=00630074,

that is, the value sould be quoted to be a string. On the other hand, the cache file created by ConTeXt has correct values.

@phi-gamma
Copy link
Member

The fontloader gets it right the first time a font is loaded, but when reading the same font from the cache the zeroes are dropped. Thanks for reporting, I’ll forward it to the Context list.

@phi-gamma
Copy link
Member

@phi-gamma
Copy link
Member

hot fix in v2.3 (no upstream fix yet because of some larger things going on with Context)

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

2 participants