Forged email detection - dictionary diacritics ignored?

[lukaklepic]

Hi,

we have custom content filter  - forged email detection - which includes dictionary with some internal company names.

This filter generaly works fine, detects and sends forged mail to quarantine.

But today false email passed through this filter, containing first and last name which is included in dictionary.

Only thing that comes in mind is that problem could be in diacritics which this dicrionary name contains:

ří č ý

What could be the reason ESA-a didn't detect this forged email, based on our dictionary, has anyone else had problem like this?

BR,

Luka

[hemohemoh · Accepted solution]

hi @lukaklepic ,

My understanding is, ESA uses ASCII encoding for dictionary matching, which does not support diacritics such as ř, í, č, and ý. Therefore, the dictionary entry with these characters is not recognized by ESA as a valid match.