Cleaning and Romanization

The bot processes user input for convenience and romanization for users with English keyboards.

The bot first performs a "clean" step that removes all whitespace characters and then further removes the following characters:

• -
• .
• '
• ·
• .

Afterwards, it also performs a romanization step. This starts with the cleaned strings of the en localization. For ships of the nations of Japan and Germany, the bot will convert characters with the following table:

From	To
ō	ou
ū	uu
ä	ae
ö	oe
ü	uu

Regardless of nationality, it will also run Unidecode against the cleaned strings. The bot will check cleaned user input against the cleaned strings in the user's language and romanizations. For example, after cleaning, the bot would expect

• any of zaō, zaou, or zao from an English user.
• any of 蔵王, zaou, or zao from a Japanese user.
• any of gkurfürst, grosserkurfürst, gkurfurst, grosserkurfurst, grosserkurfuerst, gkurfuerst from an English user.

In addition, kreml is a hardcoded romanization for Кремль/Kremlin.

If you have found an issue with this process, feel free to open an Issue.