none
robocopy unicode output jibberish log file

    Question

  • When I use the unicode option for a log file or even redirect unicode output from robocopy then try to open the resuting file in notepad.exe or whatever, it just looks like jibberish. How can I make this work or when will Microsoft fix it? Since Microsoft put that option in there, one supposes that it works with something. What is the expected usage for this option?

    Yes, I have file names with non-ASCII characters and I want to be able to view them correctly. Without unicode support robocopy just converts such characters into a '?'. It does, however, actually copy the file over correctly, thankfully. I have tried running robocopy from PowerShell and from cmd /u. Neither makes any difference. Also, one odd thing is that if I use the /unicode switch, the output to screen does properly show the non-ASCII characters, except that it doesn't show all of them, such as the oe ligature used in French 'œ'. That was just converted into an 'o' (not even an oe as is usually the case). Again, it does properly make a copy of the file. This just makes it not quite possible to search log results.
    Let's see if this post has those non-ASCII characters transmuted when this gets posted even though everything looks fine as I type it. âéèïöùœ☺♥♪
    Saturday, May 23, 2009 12:09 AM

Answers

All replies

  • When I use the unicode option for a log file or even redirect unicode output from robocopy then try to open the resuting file in notepad.exe or whatever, it just looks like jibberish. How can I make this work or when will Microsoft fix it? Since Microsoft put that option in there, one supposes that it works with something. What is the expected usage for this option?

    Yes, I have file names with non-ASCII characters and I want to be able to view them correctly. Without unicode support robocopy just converts such characters into a '?'. It does, however, actually copy the file over correctly, thankfully. I have tried running robocopy from PowerShell and from cmd /u. Neither makes any difference. Also, one odd thing is that if I use the /unicode switch, the output to screen does properly show the non-ASCII characters, except that it doesn't show all of them, such as the oe ligature used in French 'œ'. That was just converted into an 'o' (not even an oe as is usually the case). Again, it does properly make a copy of the file. This just makes it not quite possible to search log results.
    Let's see if this post has those non-ASCII characters transmuted when this gets posted even though everything looks fine as I type it. âéèïöùœ☺♥♪
    Uses /UNILOG:logfile   instead of /LOG:logfile


    Wednesday, August 5, 2009 9:01 AM
  • "Uses /UNILOG:logfile   instead of /LOG:logfile" ok, what's that supposed to mean? Yes, I am using the /UNILOG option as I said, "
    When I use the unicode option for a log file"
    Wednesday, August 5, 2009 4:46 PM
  • Same issue for me (/unilog give garbage) very visible in french

    /unicode and >> to a file don t help

    Sunday, December 12, 2010 1:29 PM
  • I have the same issue.  However, if I used the XP026 (5.1.2600.26) version it produce a readable unicode log file.  The XP027 version i have on my system32 directory is 5.1.10.1027.
    Wednesday, December 29, 2010 2:24 AM
  • Yeap same issue

    + 1 "very visible in french"

    Friday, March 22, 2013 7:01 AM
  • It seems like you have wrong code page in the command line. Try use "chcp 65001" command before the robocopy.
    Saturday, August 9, 2014 3:36 PM
  • It doesn't work for me, even at the prompt command window i can't read the unicode properly...nothing to do in the log file generated by robocopy :S
    Monday, April 20, 2015 12:42 PM
  • It seems like you have wrong code page in the command line. Try use "chcp 65001" command before the robocopy.

    thanks.

    chcp 1255 - for Hebrew

    Tuesday, December 4, 2018 8:57 AM