commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jens Scheffler (JIRA)" <>
Subject [jira] Updated: (VFS-225) File name parsing issues in layered file systems
Date Sun, 16 Nov 2008 09:53:50 GMT


Jens Scheffler updated VFS-225:


Please find a Unit Test attached which is generating the test data.

The test crates a temporary ZIP file with 2 files
- Dir/SomeFile.txt
- Dir/Important!!.txt
Then it uses commons-vfs to scan through the files.
During opening the ZIP file to function to resolve base URI is called and as of generating
the wrong base URI as described above the file is not added to the ZIP container during init.

Hope this helps.

> File name parsing issues in layered file systems
> ------------------------------------------------
>                 Key: VFS-225
>                 URL:
>             Project: Commons VFS
>          Issue Type: Bug
>    Affects Versions: 1.0
>         Environment: Windows XP SP2 - but also assuming in other environments
>            Reporter: Jens Scheffler
>         Attachments:
> The "!" character is used as delimiter for e.g. ZIP file access as VFS component.
> When trying to traverse a layered file, e.g. traversing a ZIP content and the ZIP contains
a file with a "!" in the file name itself then a FileSystemException appears - it seems that
the parsing routine for layered filenames is stumbling over the "!" character.
> Exception trace for a test VFS component:
> org.apache.commons.vfs.FileSystemException: Incorrect file system URI "syncdb:file:///C:/Temp/test.xml!/c/temp/Maps/Important
to Read!!/" in name "syncdb:file:///C:/Temp/test.xml!/c/temp/Maps/Important to Read!!.txt",
was expecting "syncdb:file:///C:/Temp/test.xml!/".
> 	at org.apache.commons.vfs.provider.AbstractFileSystem.resolveFile(
> 	at org.apache.commons.vfs.provider.AbstractFileSystem.resolveFile(
> 	at org.apache.commons.vfs.provider.AbstractFileObject.resolveFile(
> 	at de.jensscheffler.ftpsync.db.SyncFileSystemTest.treeCopy(
> 	(...)
> As the accessing code was just traversing a folder tree with no chance to handle this,
is there any alternative for traversing through these layered file systems when the content
contains special characters?
> Maybe the approach for parsing needs to be enhanced also, I could offer some help but
maybe need a hint to contribute a fix.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message