Techmail, Languages
Oracle B2C Service

The behavior is not related to the fact that the customer entry was submitted in Hebrew. It is related to the type of encoding used by the customer in their entry.

Techmail does not support the implicit mode for iso-8859-8-i encoding when parsing bidirectional text. The same message sent with the default iso-8859-8 encoding will be parsed correctly and presented in the threads. Therefore, in order to address the issue going forward, you will need to use the raw message attached to the incident for any message using this encoding.
 
Even if UTF-8 encoding includes Hebrew, this language is stored in the body of the email as a hexadecimal code. And UTF-8 uses different hexadecimal values for Hebrew, thus, characters become unreadable. Charset=ISO-8859-8 is accepted and causes no problems as opposed to charset=ISO-8859-8-i. By looking at "-i" addition, it appears to be the same charset but "charset=ISO-8859-8-i" is a kind of addition or extension (charset with a parameter) to the original one "charset=ISO-8859-8". It is possibly not in general use or not popular. That is why it cannot be recognized by the utility.

 

This behavior is as designed. The product supports all major encoding charsets such as charset=ISO-8859-8 but there are some limitations for a SUB charsets or an extension.