Add Shift-JIS encoding support #29
Conversation
| default: throw Error._unsupported(encoding: encoding) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| extension Data { |
There was a problem hiding this comment.
Please, make the extension fileprivate.
| default: throw Error._unsupported(encoding: encoding) | ||
| } | ||
| } | ||
| } | ||
|
|
||
| extension Data { | ||
| var encodedHexadecimals: [UInt8]? { |
There was a problem hiding this comment.
private or fileprivate variables and functions are underscored in this library.
| var encodedHexadecimals: [UInt8]? { | ||
| let responseValues = self.withUnsafeBytes({ (pointer: UnsafeRawBufferPointer) -> [UInt8] in | ||
| let unsafeBufferPointer = pointer.bindMemory(to: UInt8.self) | ||
| let unsafePointer = unsafeBufferPointer.baseAddress! |
There was a problem hiding this comment.
Since this code will be repeated very often, I believe it is worth to change the force cast into a unsafelyUnwrapped.
|
Sigh, I guess it would. I will make the changes myself. |
|
Alright, @rkyymmt and @rikiya-yamamoto. I have taken a deeper look at your code and saw that you had an unnecessary array allocation that was being executed for every scalar. I removed it and access the underlying byte representation directly. Performance should now be better, but still it is far from ideal since we are performing a round trip from Unicode Scalar -> String -> to Data. I am not very knowledgable on ShiftJIS, but I am sure there must be a way to transform Unicode Scalar to UInt8 directly. In any case, the code should now work and be faster than what you sent. Please test it out with your data. Regards |
Description
Add Shift-JIS encoding support.
(Shift-JIS used in Japan.)