[Question] What encoding names about iconv-lite

[ashtuchkin]

Hi!

I haven't checked it manually, but I think most names that are used in the real world will be interpreted correctly. First, names are not case sensitive in iconv-lite and all non-alphanumeric characters are stripped, as well as what seems like a year suffix (i.e. ISO_8859-1:1987 = ISO8859-1 = iso88591). Then, there are lots of aliases, for example win1251 = windows1251 = 1251 = Windows-1251.

If you have a limited set of encodings that you support, you can check them all with iconv.encodingExists(name) function. Then, you can either tell me to add an alias, or do it in your project with the following code:

var iconv = require('iconv-lite');
iconv.encode('utf8', ''); // load encoding definitions
iconv.encodings['<new alias name>'] = '<alias to what>'; // both lowercase, alphanumeric.

If, on the other hand, you want to support all IANA names and aliases, then, probably, iconv-lite is not an option because it doesn't support all the older encodings (neither do node-iconv AFAIK).

[btd]

Thanks, for response. My purpose is getting some possibly text content via http requests, and from headers get charset and decode them to utf-8 (i.e. that is what browsers do =), as header Content-Type: mime; charset=IANA_charset name or alias, thus i was need to know that question.
I do not want to support all encodings, it is enough most common one byte encoding and utf-8, utf-16*.
Thanks that point to . encodingExists it should be enough to see if will be any errors, just because it is missing, but i think with main encoding names, should no be problems (i have another module that resolve aliases).