š shows as '?' in chm TOC for Czech

[lade]

I tried building some stuff in Czech in chm format. The character 'š' shows as a question mark in the TOC.

Removing 'š' and 'Š' from entities.properties fixed the problem, so it seems that chm TOC does not support this character as an html entity.

[jelovirt]

IIRC I had a similar problem with Polish with and the problem was that CHM has no way of storing non-ASCII characters in the TOC file. Closing this as a duplicate of issue #1151.

[lade]

The problem actually can be fixed if you use a correct charset (windows-1250) used by HTMLHelp Compiler instead of iso-8859-2, and š instead of š.

[lade]

I just submitted a case about this wrong charset used (Issue #1269).

[lade]

The problem with š and Š still exists in DITA OT 1.7.5 for Czech and Slovenian. Removing them from entities.properties and changing html output to 'windows-1250' in codepages.xml for these languages inside dost.jar fixes the problem.