Select the product you need help with
How To Move Files That Are Currently in UseArticle ID: 140570 - View products that this article applies to. This article was previously published under Q140570 On This PageSUMMARY
Sometimes Win32 applications need to delete, rename, or move files that are
currently being used by the system. One common example is that setup
programs need to remove themselves from the user's hard disk when they are
finished setting up a software package. Sometimes, they also need to move
device drivers that are currently being used by the system. Applications
need help from the operating system to delete or move these files.
Windows 95 and Windows NT each provide a unique method for helping applications to remove, replace, or rename files and directories that are in use. Although the two platforms differ in how they implement these methods, both share an overall strategy where the application specifies which files to process, and the system processes them when it reboots. This article explains how applications can use the method provided by each Windows platform. MORE INFORMATIONMoving Files in Windows NTWin32-based applications running on Windows NT should use MoveFileEx() with the MOVEFILE_DELAY_UNTIL_REBOOT flag to move, replace, or delete files and directories currently being used. The next time the system is rebooted, the Windows NT bootup program will move, replace, or delete the specified files and directories.To move or replace a file or directory that is in use, an application must specify both a source and destination path on the same volume (for example, drive C:). If the destination path is an existing file, it will be overwritten. If the destination path is an existing directory, it will not be overwritten and both the source and destination paths will remain unchanged. Here is an example call to move or replace a file or move a directory: Moving Files in Windows 95Windows 95 does not implement MoveFileEx(), but does provide an alternate way for all Win32-based, 16-bit Windows-based, and MS-DOS-based applications to move, replace, or delete files (but not directories) that are currently in use. This capability is implemented through the [rename] section of a file named Wininit.ini. If Wininit.ini is present in the Windows directory, Wininit.exe processes it when the system boots. Once Wininit.ini has been processed, Wininit.exe renames it to Wininit.bak.The syntax of the [rename] section is: The [rename] section can have multiple lines with one file per line. To delete a file, specify NUL as the DestinationFileName. Here are some entry examples: Applications should not use WritePrivateProfileString() to write entries to the [rename] section because there can be multiple lines with the same DestinationFileName, especially if DestinationFileName is "NUL." Instead, they should add entries by parsing Wininit.ini and appending the entries to the end of the [rename] section. NOTE: Always use a case-insensitive search to parse Wininit.ini because the title of the [rename] section and the file names inside it may have any combination of uppercase and lowercase letters. Applications that use Wininit.ini should check for its existence in the Windows directory. If Wininit.ini is present, then another application has written to it since the system was last restarted. Therefore, the application should open it and add entries to the [rename] section. If Wininit.ini isn't present, the application should create it and add to the [rename] section. Doing so ensures that entries from other applications won't be deleted accidentally by your application. To undo a file rename operation before the system is rebooted, you must remove the corresponding line from the [rename] section of the Wininit.ini file. PropertiesArticle ID: 140570 - Last Review: November 21, 2006 - Revision: 4.3 APPLIES TO
| Article Translations
|


Back to the top








