{ "WorkItem": { "AffectedComponent": { "Name": "", "DisplayName": "" }, "ClosedComment": "Not a bug; possibly user error.", "ClosedDate": "2009-05-30T07:38:55.51-07:00", "CommentCount": 0, "Custom": null, "Description": "- Im trying to compress a folder with multiply files. Some of these file names include Danish characters. \n- Those files with Danish characters have been renamed because the characters cannot be resolved.\n- It is in particular the character 'ø' that fails ø becomes o. Characters 'æ' and 'å' are fine.", "LastUpdatedDate": "2013-05-16T05:32:23.593-07:00", "PlannedForRelease": "", "ReleaseVisibleToPublic": false, "Priority": { "Name": "Low", "Severity": 50, "Id": 1 }, "ProjectName": "DotNetZip", "ReportedDate": "2009-05-20T06:58:16.517-07:00", "Status": { "Name": "Closed", "Id": 4 }, "ReasonClosed": { "Name": "Unassigned" }, "Summary": "Zip/Unzip not working correctly when using Danish characters.", "Type": { "Name": "Issue", "Id": 3 }, "VoteCount": 1, "Id": 7791 }, "FileAttachments": [ { "FileId": 1968, "FileName": "example.txt", "DownloadUrl": ".\\1968" } ], "Comments": [ { "Message": "What encoding did you specify for the zip file? IBM437, the default encoding for zipfiles, is not able to encode all characters in the Danish alphabet. What is your default system code page? Have you tried to set the encoding on the zipfile to that value? See the documentation for the ZipFile.ProvisionalAlternateEncoding property for more information. ", "PostedDate": "2009-05-21T08:41:28.427-07:00", "Id": -2147483648 }, { "Message": "Hello, i have tried setting the ProvisionalAlternateEncoding to (ASCII, UTF8 and Unicode) but the only one that comes close is the default encoding (IBM437). \r\n- Im not sure what the default system code page is.", "PostedDate": "2009-05-25T01:25:04.043-07:00", "Id": -2147483648 }, { "Message": "Can you show me your code? I have test cases that include files that use the Danish language alphabet, incuding the ø character. These tests succeed on my system.", "PostedDate": "2009-05-25T02:38:33.28-07:00", "Id": -2147483648 }, { "Message": "I have attached a txt file with the function and the arguments i use when i call the function. \r\n1. The folder that i zip contains 2 txt files one of them contains ø in the filename. When zipped ø becomes o.\r\n\r\nIf you need more information please let me know.", "PostedDate": "2009-05-25T04:00:14.297-07:00", "Id": -2147483648 }, { "Message": "", "PostedDate": "2009-05-25T04:00:18.28-07:00", "Id": -2147483648 }, { "Message": "Thanks for the code. It code works for me, if I uncomment the line that reads zipfile.UseUnicodeAsNecessary = True . What version of the library are you using? How do you determine that the character conversion has occurred? \r\n", "PostedDate": "2009-05-25T08:36:57.157-07:00", "Id": -2147483648 }, { "Message": "Strange because it fails for me. The version is 1.8.3.7. The last part about the character conversion, do you mean the zip file i create? i open it in winrar.", "PostedDate": "2009-05-26T00:07:28.393-07:00", "Id": -2147483648 }, { "Message": "The reason I asked about \"how can you determine the char conversion has occurred?\" is this: if I write a .NET application to read a zip file that contains an entry with the Danish characters, encoded in UTF-8, and then print out the filename in the console window, it will show incorrectly. In fact the ø becomes o, in accordance with the \"best fit\" conversion approach. But, if I extract the entry to the filesystem, and then view the extracted folder in Windows Explorer, Explorer shows the correct filename. My point is that it is possible to jump to incorrect conclusions about filename conversion, depending on where you view the filename. \r\n\r\nWinRar supports code pages, but I am not sure if it supports zip files with the UTF-8 filename encoding. Also I imagine that code page support depends on the version of WinRAR. I am not an expert on WinRAR, but I believe the default behavior of WinRAR is to create zip files using the system default codepage. This is in violation of the zip specification, which says zip tools ought to use only IBM437 or UTF-8. Even so, I have built DotNetZip to create and read zip files using any code page, as a way to get better interoperability. \r\n\r\nI believe these combinations work successfully (no conversion of characters): \r\n - zip up with DotNetZip using UTF-8, and unzip with DotNetZip. \r\n - zip up with DotNetZip using the System.Text.Encoding.Default code page, unzip with WinRAR. \r\n - zip up with DotNetZip using the System.Text.Encoding.Default code page, unzip with Windows Explorer. \r\n - zip up with WinRAR (which I believe implicitly uses the Encoding.Default page), and unzip with DotNetZip, explicitly specifying the Encoding.Default code page\r\n\r\nI believe these combinations do not work (in other words I *believe* you will get char conversions):\r\n - zip up with DotNetZip using UTF-8, and unzip with WinRAR\r\n - zip up with DotNetZip using UTF-8, and unzip with Windows Explorer. I believe Windows Explorer does not support UTF-8 zip files.\r\n - zip up with DotNetZip using a code page other than the default or IBM437, and unzip with WinRAR\r\n\r\nI am certain you will get character conversions if you use Danish-alphabet characters, and this approach: \r\n - zip up with DotNetZip using IBM437 (UseUnicode=false), and unzip with anything\r\n\r\nPlease let me know what combination of tools and code pages you are using. \r\n", "PostedDate": "2009-05-26T10:45:19.103-07:00", "Id": -2147483648 }, { "Message": "Hey i have tried all suggestions and still no luck zipping the files, i have tried various programs to view the zip file in and so on. \r\nBut finally i have had a little break, if i user System.Text.Encoding.GetEncoding(850) for encoding it will zip the files correct, and i can view it correct from various programs. \r\nHowever when i try to unzip it with dotnetzip the Ø i transformed to a weird character ¢. So i can zip it but i cant unzip. But if i unzip it with winrar it willl unzip correct.", "PostedDate": "2009-05-27T00:41:20.26-07:00", "Id": -2147483648 }, { "Message": "Ive tried to use System.Text.Encoding.GetEncoding(1252) this works for unzipping and this works well", "PostedDate": "2009-05-27T05:51:29.28-07:00", "Id": -2147483648 }, { "Message": "I'm glad you figured out a way to get it to work. \r\nI don't know why you had so much trouble. If I had a look at the code you were using I might be able to provide some insight. \r\n\r\nI'm going to close this workitem. If you want to continue the discussion to try to solve any remaining mysteries, let's move the dialogue bto the discussion forums, rather than use the workitem list. Thanks.", "PostedDate": "2009-05-30T07:38:35.137-07:00", "Id": -2147483648 }, { "Message": "", "PostedDate": "2009-05-30T07:38:55.51-07:00", "Id": -2147483648 }, { "Message": "", "PostedDate": "2013-02-21T18:44:20.72-08:00", "Id": -2147483648 }, { "Message": "", "PostedDate": "2013-05-16T05:32:23.593-07:00", "Id": -2147483648 } ] }