Ticket #4740 (new enhancement)

SVN Diffs for #4740

 

Opened 3 years ago

Last modified 1 year ago

Detect Illegal Mark Combinations

Reported by: emader(at)icu-project.org Assigned to: eric
Priority: minor Milestone: UNSCH
Component: layout Version: cvs
Keywords: Cc:
Load: Xref: 2963
Java Version: Operating System: all
Project (C/J): ICU4C Weeks: 2
Review:

Description (Last modified by grhoten)

Some of the layout engines don't detect illegal mark combinations. This can mean that marks are over-struck, or that non-canonically identical strings will render the same way. This could be exploited for spoofing.

This need to be done for Arabic, Hebrew, and the default OpenType engine. Also the Thai and Lao OpenType engines.

It might be possible to develope a generic algorithm that will work for all scripts.

Also, the TR 2 default accent stacking should be done if there's no GSUB / GPOS tables. See JB#2963.

Attachments

Change History

12/31/69 17:40:55 changed by auditor

  • Mon Oct 16 21:27:01 2006 emader changed notes2: target: "UNSCH" to "3.8 candidate",
  • Tue Nov 14 00:12:54 2006 srl changed notes2: target: "3.8 candidate" to "3.8",

06/27/07 15:49:30 changed by grhoten

  • load changed.
  • java changed.
  • description changed.
  • milestone changed from 3.8 to UNSCH.
  • keywords deleted.
  • revw changed.

10/03/07 10:56:35 changed by eric

  • summary changed from RFE: Detect Illegal Mark Combinations to Detect Illegal Mark Combinations.

11/12/07 09:54:41 changed by eric

  • weeks changed from 1 to 2.

Add/Change #4740 (Detect Illegal Mark Combinations)




Anti spam check: