[Cyberduck-trac] [Cyberduck] #7284: Impossible to access a storage container when a file which name includes special caracters has previously been uploaded

[Cyberduck]

#7284: Impossible to access a storage container when a file which name includes
special caracters has previously been uploaded
-----------------------------------+------------------------
 Reporter:  benmauff               |         Owner:  dkocher
     Type:  defect                 |        Status:  new
 Priority:  high                   |     Milestone:  4.3.2
Component:  s3                     |       Version:  4.3.1
 Severity:  critical               |    Resolution:
 Keywords:  special caracters, S3  |  Architecture:
 Platform:  Windows 7              |
-----------------------------------+------------------------
Changes (by dkocher):

 * milestone:   => 4.3.2


Old description:

> Configuration: Swift + S3
>
> This issue has been encountered in the following scenario:
> 1 Upload the following file in a S3 container: Nom de fichier €%£ -
> Copie.xlsx
>
> 2 Access the folder where this file has been updated and check it has
> been added
>
> Result: it is now impossible to access the container; we get an Internal
> Server Error (see screenshots).
>
> Here is our first internal analysis:
>
> - Creating object with special characters works in Swift:
>
> ~$ curl -H 'Content-Length: 0' -i
> http://localhost:8080/v1/AUTH_05b7534319e849ba93c68462726057fb/newcont/€%£-Copie.xlsx
> -X PUT -H "X-Auth-Token: 2438e55a25a8407fa6318fea4e7bd10a"
>
> ~$ curl -i
> http://localhost:8080/v1/AUTH_05b7534319e849ba93c68462726057fb/newcont -X
> GET -H "X-Auth-Token: 2438e55a25a8407fa6318fea4e7bd10a"
> HTTP/1.1 200 OK
> Content-Length: 29
> X-Container-Object-Count: 2
> Accept-Ranges: bytes
> X-Timestamp: 1369816325.76530
> X-Container-Bytes-Used: 1196
> Content-Type: text/plain; charset=utf-8
> X-Trans-Id: tx16fe028810ba4be39506b-0051a5cc5d
> Date: Wed, 29 May 2013 09:37:33 GMT
>
> etc/passwd
> €%£-Copie.xlsx
>
> but not in  Cyberduck.
>
> - this works as well with the S3 API, the script is here :
>
> http://p.chmouel.com/s3testutf8.py
>
> and the output :
>
> ParseResult(scheme='http', netloc='46.231.128.140:8080', path='',
> params='', query='', fragment='')
> Delete key: €%£ - Copie.xlsx
> Deleting bucket: cont_s3
> Creating bucket: cont_s3
> uploading: €%£ - Copie.xlsx
> uploaded object
> €%£ - Copie.xlsx

New description:

 Configuration: Swift + S3

 This issue has been encountered in the following scenario:
 1 Upload the following file in a S3 container: Nom de fichier €%£ -
 Copie.xlsx

 2 Access the folder where this file has been updated and check it has been
 added

 Result: it is now impossible to access the container; we get an Internal
 Server Error (see screenshots).

 Here is our first internal analysis:

  * Creating object with special characters works in Swift:


 {{{
 ~$ curl -H 'Content-Length: 0' -i
 http://localhost:8080/v1/AUTH_05b7534319e849ba93c68462726057fb/newcont/€%£-Copie.xlsx
 -X PUT -H "X-Auth-Token: 2438e55a25a8407fa6318fea4e7bd10a"
 }}}



 {{{
 ~$ curl -i
 http://localhost:8080/v1/AUTH_05b7534319e849ba93c68462726057fb/newcont -X
 GET -H "X-Auth-Token: 2438e55a25a8407fa6318fea4e7bd10a"
 HTTP/1.1 200 OK
 Content-Length: 29
 X-Container-Object-Count: 2
 Accept-Ranges: bytes
 X-Timestamp: 1369816325.76530
 X-Container-Bytes-Used: 1196
 Content-Type: text/plain; charset=utf-8
 X-Trans-Id: tx16fe028810ba4be39506b-0051a5cc5d
 Date: Wed, 29 May 2013 09:37:33 GMT
 }}}



 {{{
 etc/passwd
 €%£-Copie.xlsx
 }}}


 but not in  Cyberduck.

  * this works as well with the S3 API, the script is here :

 [http://p.chmouel.com/s3testutf8.py]

 and the output :


 {{{
 ParseResult(scheme='http', netloc='46.231.128.140:8080', path='',
 params='', query='', fragment='')
 Delete key: €%£ - Copie.xlsx
 Deleting bucket: cont_s3
 Creating bucket: cont_s3
 uploading: €%£ - Copie.xlsx
 uploaded object
 €%£ - Copie.xlsx
 }}}

--

-- 
Ticket URL: <http://trac.cyberduck.ch/ticket/7284#comment:1>
Cyberduck <http://cyberduck.ch>
Open source FTP, SFTP, WebDAV, Cloud Files, Google Docs & Amazon S3 Browser for Mac & Windows.