Different output between Powershell ToBase64String & Linux base64

Question

SGVsbG8sIHdvcmxkIQ== I have to write two scripts, one for Windows Server and another for Ubuntu Server. To illustrate, if my bash script runs:

echo -n 'BASE64' | base64

the result is QkFTRTY0. If my PowerShell Script runs:

[System.Convert]::ToBase64String([System.Text.Encoding]::Unicode.GetBytes('BASE64'))

the result is QgBBAFMARQA2ADQA. Is important get the same PowerShell string in Linux. Anybody knows why this difference and some solution to my problem? I think that the problem is related to Unix's UTF-8 encoding, but I can't find a solution. I can decode the PowerShell output in Linux, but I can't encode a string in Linux and get the same result as PowerShell. A bash solution is optimal, but a C/C++ code solution or guide works too.

What about if you do [System.Convert]::ToBase64String([System.Text.Encoding]::UTF8.GetBytes('BASE64'))? — Scott Chamberlain
– Scott Chamberlain, Commented Mar 31, 2014 at 17:19
@ScottChamberlain, I have to get the same result of Unicode.GetBytes in Linux. — 1e100
– 1e100, Commented Mar 31, 2014 at 17:21
You can't get it to be the same because Linux uses utf8 and you are forcing Unicode and therefor have two different byte arrays. Why do you have the requirement to force Unicode encoding, using UTF8.GetBytes returns the correct result. — Scott Chamberlain
– Scott Chamberlain, Commented Mar 31, 2014 at 17:23
@ScottChamberlain I'm forced to use Unicode.GetBytes, because I have to exchange the result of my Bash script with a propietary Windows software that seems use Unicode.GetBytes. I'm not sure, because I don't have access to the source code, but if I use the strings generated by PowerShell i can interact with it in Windows without problem. But I need to do the same on my Ubuntu Server. — 1e100
– 1e100, Commented Mar 31, 2014 at 17:32
OK, so your question is actually: "how do I convert a string to UTF-16 on Ubuntu, and how do I get the base64 of that?" — Roger Lipscombe
– Roger Lipscombe, Commented Mar 31, 2014 at 17:47

Roger Lipscombe · Accepted Answer · 2018-04-18 08:45:42Z

13

You need to use iconv to convert from UTF-8 to UTF-16 without a byte order mark:

$ echo -n 'BASE64' | iconv -f UTF8 -t UTF16LE | base64
QgBBAFMARQA2ADQA

The UTF16LE causes it to omit the BOM.

See https://superuser.com/q/381056/4206, where someone asks about forcibly including the BOM, but you want the opposite, which -- conveniently -- is in the question.

edited Apr 18, 2018 at 8:45

answered Mar 31, 2014 at 17:50

Roger Lipscombe

92.5k59 gold badges255 silver badges402 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Dɑvïd Over a year ago

With OSX 'iconv' the supported encoding switch is UTF-16LE. FWIW!

Collectives™ on Stack Overflow

Different output between Powershell ToBase64String & Linux base64

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related