Comments on: Encoding C# strings as Byte[] (Byte Arrays) and back again

By: Martijn

Martijn — Sun, 15 Feb 2009 04:58:10 +0000

This is a shot in the dark without seeing the original files of course. You will need to establish the source encoding first. If it isn’t unicode then its probably a Japanese ISO-2022-JP or SHIFT-JIS encoded file. In Shift-JIS certain combinations of multiple high ASCII characters form a single Japanese character.

Did you try something like:

srInputFile = new StreamReader(filename,Encoding.GetEncoding(”iso-2022-jp”))

That should make sure the text is imported from ISO-2022 and correctly converted to Unicode on reading.

By: Laith

Laith — Fri, 13 Feb 2009 05:37:41 +0000

Hello

I am writing an app in C# which reads in a c/h (c-language) files, and removing certain lines of comments that meets a criteria. These c/h files contain both English and Japanese comments. such as this line, if it will show up…
u1_ret = (U1)SOME_CONST; /* 仮に未確定とする */

The problem is that when writing the file back, the the Japanese comments are corrupted and show up as squares like this…
u1_ret = (U1)SOME_CONST; /* ��ɖ��m��Ƃ��*/

I am creating the streams as follows:
srInputFile = new StreamReader(filename);
swOutputFile = new StreamWriter(outputDir + “\\_” + outputFileName, false, srInputFile.CurrentEncoding);

Here is how i am writing to file
while (!srInputFile.EndOfStream)
{
LineIn = srInputFile.ReadLine();
if (IsTextMatch(LineIn) == false)
{
swOutputFile.WriteLine(LineIn.Normalize());
}
}

I tried different encodings but the problem is still the same…
I appreciate your help…

regards
ld

By: Tyler

Tyler — Fri, 30 Jan 2009 16:55:16 +0000

I needed a quick reminder on encoding byte arrays, and ended up rethinking my approach because of the estra detail in your post. appreciate it.