Mohit,
I use Hive 0.7.1 and am able to access the file from distributed cache just by
filename. Did you try that?
Mark
----- Original Message -----
From: "Chinna Rao Lalam 72745" <chinna...@huawei.com>
To: user@hive.apache.org
Sent: Wednesday, October 19, 2011 6:56:38 AM
Subject: Re: Accessing distributed cache in transform scripts
Hi,
Can u post some more details like for the "list file" what command u have
used.
----- Original Message -----
From: Mohit Gupta <success.mohit.gu...@gmail.com>
Date: Wednesday, October 19, 2011 3:16 pm
Subject: Re: Accessing distributed cache in transform scripts
To: user@hive.apache.org
> Plz help...Any pointers!!
>
> On 10/19/11, Mohit Gupta <success.mohit.gu...@gmail.com> wrote:
> > Hi All,
> >
> > I want some read-only data to be available at the reducers /
> transform> scripts. I am trying to use distributed cache to
> achieve this using
> > the following steps:
> > 1. add file s3://bucket_name/prefix/testfile
> > then
> > 2. "list file" to find out the location of local copy of testfile.
> > it shows the locally file is stored at
> > "/mnt/var/lib/hive_05/downloaded_resources/testfile"
> >
> > 3. reduce script code ( in php)
> > #!/usr/bin/php
> > <?php
> >
> $fin=fopen("/mnt/var/lib/hive_05/downloaded_resources/testfile","r");>
> $line=fgets($fin);
> > ?>
> >
> > 4. the reduce script throws an error "no such file or directory
> > /mnt/var/lib/hive_05/downloaded_resources/testfile ".
> >
> > 5. On ssh, I found that the file was indeed present at
> > "/mnt/var/lib/hive_05/downloaded_resources/testfile".
> >
> > Plz help. Is this the right way to pass some info to the reduce
> > scripts. If yes, what I could be missing here? Else, any other
> > possible way(s)!!
> >
> > Thanks
> >
> > Mohit
> >
>
>
> --
> Best Regards,
>
> Mohit Gupta
> Software Engineer at Vdopia Inc.
>
