[Cyberduck-trac] [Cyberduck] #9677: Recursive Searching only works down one level
Cyberduck
trac at trac.cyberduck.io
Tue Sep 6 08:29:25 UTC 2016
#9677: Recursive Searching only works down one level
---------------------+-------------------------
Reporter: Scoth42 | Owner: dkocher
Type: defect | Status: assigned
Priority: normal | Milestone: 5.2
Component: s3 | Version: 5.1
Severity: normal | Resolution:
Keywords: | Architecture:
Platform: |
---------------------+-------------------------
Changes (by dkocher):
* owner: => dkocher
* status: new => assigned
* milestone: => 5.2
Old description:
> I did some searching around and couldn't decide if this was intentional
> or not, so feel free to close this as such if it is.
>
> Recursive searching only works one level deep. For example, if I have a
> path with some files that looks like:
>
> /Bucket/Collections/Bob/Date1/SomeRecording1.wav
> /Bucket/Collections/Bob/Date1/SomeRecording2.wav
> /Bucket/Collections/Bob/Date1/SomeRecording3.wav
> /Bucket/Collections/Bob/Date2/SomeRecording1.wav
> /Bucket/Collections/Bob/Date2/SomeRecording2.wav
> /Bucket/Collections/Bob/Date2/SomeRecording3.wav
> /Bucket/Collections/Steve/Date1/SomeRecording1.wav
> /Bucket/Collections/Steve/Date1/SomeRecording2.wav
> /Bucket/Collections/Steve/Date1/SomeRecording3.wav
> /Bucket/Collections/Steve/Date2/SomeRecording1.wav
> /Bucket/Collections/Steve/Date2/SomeRecording2.wav
> /Bucket/Collections/Steve/Date2/SomeRecording3.wav
>
> If I drill down to the person's name and try to search, I'll get the
> expected resulting date folders that have contents that match. This is
> fine.
>
> If I back up to a lower directory, however, and try searching no results
> will be found. Based on the time, logs, and status bars it seems like
> it's poking through the directories, but never returns any paths. This is
> preventing me from searching for matches among multiple directories. I
> couldn't sit in /Bucket and search on "SomeRecording1" and have it return
> all the directories that have that. I've tried expanding out the
> directories as much as I can in case that's related somehow, but I have
> too many files to practically expand it that far.
>
> Here's a snippet of the logs. They're all basically like this as it walks
> through the tree. Apologies for the weird formatting, it wasn't including
> newlines properly.
>
> GET /?max-
> keys=1000&prefix=recordings%2FCollections%20-%20Broken%20Promise%2F7_1_2016%2F&delimiter=%2F
> HTTP/1.1
>
> Date: Wed, 31 Aug 2016 15:23:37 GMT
>
> x-amz-request-payer: requester
>
> x-amz-content-sha256: sha256 key
>
> Host: bucket.s3.amazonaws.com
>
> x-amz-date: 20160831T152337Z
>
> Authorization: auth stuff
>
> Connection: Keep-Alive
>
> User-Agent: Cyberduck/5.2.0.20906 (Windows 10/10.0) (x86)
>
> HTTP/1.1 200 OK
>
> x-amz-id-2: amz id info
>
> x-amz-request-id: 1CDB3FEB58747DCA
>
> Date: Wed, 31 Aug 2016 15:23:38 GMT
>
> x-amz-bucket-region: us-east-1
>
> Content-Type: application/xml
>
> Transfer-Encoding: chunked
>
> Server: AmazonS3
New description:
I did some searching around and couldn't decide if this was intentional or
not, so feel free to close this as such if it is.
Recursive searching only works one level deep. For example, if I have a
path with some files that looks like:
{{{
/Bucket/Collections/Bob/Date1/SomeRecording1.wav
/Bucket/Collections/Bob/Date1/SomeRecording2.wav
/Bucket/Collections/Bob/Date1/SomeRecording3.wav
/Bucket/Collections/Bob/Date2/SomeRecording1.wav
/Bucket/Collections/Bob/Date2/SomeRecording2.wav
/Bucket/Collections/Bob/Date2/SomeRecording3.wav
/Bucket/Collections/Steve/Date1/SomeRecording1.wav
/Bucket/Collections/Steve/Date1/SomeRecording2.wav
/Bucket/Collections/Steve/Date1/SomeRecording3.wav
/Bucket/Collections/Steve/Date2/SomeRecording1.wav
/Bucket/Collections/Steve/Date2/SomeRecording2.wav
/Bucket/Collections/Steve/Date2/SomeRecording3.wav
}}}
If I drill down to the person's name and try to search, I'll get the
expected resulting date folders that have contents that match. This is
fine.
If I back up to a lower directory, however, and try searching no results
will be found. Based on the time, logs, and status bars it seems like it's
poking through the directories, but never returns any paths. This is
preventing me from searching for matches among multiple directories. I
couldn't sit in /Bucket and search on "SomeRecording1" and have it return
all the directories that have that. I've tried expanding out the
directories as much as I can in case that's related somehow, but I have
too many files to practically expand it that far.
Here's a snippet of the logs. They're all basically like this as it walks
through the tree. Apologies for the weird formatting, it wasn't including
newlines properly.
{{{
GET /?max-
keys=1000&prefix=recordings%2FCollections%20-%20Broken%20Promise%2F7_1_2016%2F&delimiter=%2F
HTTP/1.1
Date: Wed, 31 Aug 2016 15:23:37 GMT
x-amz-request-payer: requester
x-amz-content-sha256: sha256 key
Host: bucket.s3.amazonaws.com
x-amz-date: 20160831T152337Z
Authorization: auth stuff
Connection: Keep-Alive
User-Agent: Cyberduck/5.2.0.20906 (Windows 10/10.0) (x86)
HTTP/1.1 200 OK
x-amz-id-2: amz id info
x-amz-request-id: 1CDB3FEB58747DCA
Date: Wed, 31 Aug 2016 15:23:38 GMT
x-amz-bucket-region: us-east-1
Content-Type: application/xml
Transfer-Encoding: chunked
Server: AmazonS3
}}}
--
--
Ticket URL: <https://trac.cyberduck.io/ticket/9677#comment:1>
Cyberduck <https://cyberduck.io>
Libre FTP, SFTP, WebDAV, S3 & OpenStack Swift browser for Mac and Windows
More information about the Cyberduck-trac
mailing list