Implement byteLength method.

[reklatsmasters]

This PR implements byteLength method. API is the same as in Buffer.byteLength.

[ashtuchkin]

Interesting, is it much faster than iconv.encode(..).length? What's the use case here?

[reklatsmasters]

@ashtuchkin It's faster for internal and single-byte encodings. The use cases are the same as for Buffer.byteLength - if you want to known size of string in bytes before encoding.

[ashtuchkin]

What is your use case, why do you need that api (vs using encode().length)? Why do you need these milliseconds? I'm still trying to wrap my head over why would you need the byte length without the content? You're not allocating memory for that are you? I'm wondering why there were no feature requests for that in the past 4 years.. did anything change recently? I'm in general reluctant to increase the API surface for something that is a bit faster for a small subset of encodings, so we need to really understand what that gives us.

…

On Jul 10, 2017 4:34 AM, "Dmitriy Tsvettsikh" ***@***.***> wrote: @ashtuchkin <https://github.com/ashtuchkin> It's faster for internal and single-byte encodings. The use cases are the same as for Buffer.byteLength - if you want to known size of string in bytes before encoding. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#158 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAmVHQ1G-FWuVRp1ByOHqhdfKjgdFsK5ks5sMgwygaJpZM4OSOvP> .