expand set of codepoints returned

[Ichimonji10]

When asked to generate UTF8 characters, the FauxFactory.generate_string class method returns only characters in the CJK characterset. Although CJK characters are valid UTF8 characters, they do not represent the entire range of valid UTF8 characters. By the same logic, generate_string could return only ASCII characters when asked for UTF8 characters, and given that ASCII characters are a subset of UTF8, it would technically be acting correctly.

It would be better if the generate_string method returned a fuller set of UTF8 characters when asked to generate UTF8 characters.

The lower limit of valid UTF8 code points is 0x0, and I'm not sure what the upper limit is. According to RFC 3629, several ranges of characters are also off-limits: 0xC0, 0xC1, 0xF5–0xFF and 0xD800–0xDFFF.

[Ichimonji10]

The upper limit (inclusive) of UTF8 is 0x10FFFF.

[omaciel]

The tricky part is determining all the valid ranges and avoiding characters that are not "real" but I totally agree with you that ultimately generate_string can be improved. I need to spend a bit more time researching the topic so that I can take a stab at it.

[Ichimonji10]

I'm working on this. You'll have a pull request soon.

[omaciel]

Sweet!!!

[Ichimonji10]

The current pull request fixes this issue.

[omaciel]

Merged! You should update README.rst, HISTORY.rst and add yourself to AUTHORS.rst :)