drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From kbotzum <kbot...@maprtech.com>
Subject Re: [VOTE] Release Apache Drill version 0.5.0-incubating (rc2)
Date Mon, 08 Sep 2014 13:46:04 GMT
Here’s what I mean regarding multiple platforms:  https://cwiki.apache.org/confluence/display/DRILL/Apache+Drill+in+10+Minutes

Region File

If you followed the Apache Drill in 10 Minutes instructions to install Drill in embedded mode,
the path to the parquet file varies between operating systems. 

To view the data in the region.parquet file, issue the query appropriate for your operating
system:

Linux
SELECT * FROM dfs.`/opt/drill/apache-drill-0.4.0-incubating/sample-data/region.parquet`; 

Mac OS X
SELECT * FROM dfs.`/Users/max/drill/apache-drill-0.4.0-incubating/sample-data/region.parquet`;

Windows
SELECT * FROM dfs.`C:\drill\apache-drill-0.4.0-incubating\sample-data\region.parquet`;

Notice that all three platforms are exactly the same. The only difference is the path. And
since the path is configurable I think it would clearer if the docs simply indicated some
kind of path like INSTALL_ROOT. 

Keys
_______________________________
Keys Botzum 
Senior Principal Technologist
kbotzum@maprtech.com
443-718-0098 
MapR Technologies 
http://www.mapr.com



On Sep 6, 2014, at 5:24 PM, Bob Rumsby <brumsby@maprtech.com> wrote:

> The docs refer to the current (0.4) version of Drill. When 0.5 is actually
> released, the docs will be updated.
> 
> Regarding the donuts.json file, yes, it would be good to add this to the
> sample data.
> 
> Not sure about the "examples for multiple OS platforms"--please send a link
> to the page where you see this problem.
> 
> Thanks,
> Bob
> 
> 
> 
> On Sat, Sep 6, 2014 at 1:38 PM, kbotzum <kbotzum@maprtech.com> wrote:
> 
>> +1 (binding)
>> 
>> For variety I downloaded the release onto a Mac without Hadoop to focus on
>> standalone mode.
>> 
>> I then:
>> * ​Verified MD5 and SHA checksums on binary package
>> 
>> * Followed README examples
>> 
>> * followed drill in 10 minutes:
>> https://cwiki.apache.org/confluence/display/DRILL/Apache+Drill+in+10+Minutes
>> 
>> * observed that the documentation contains references to the wrong version
>> of Drill and contains examples for multiple OS platforms even though they
>> are exactly the same except for the install path. Why not just make the
>> install path a variable and define it in a platform and version specific
>> way at the top of the docs? I’ll open a JIRA if others agree.
>> 
>> * the documentation references donuts.json (
>> https://cwiki.apache.org/confluence/display/DRILL/Sample+Data%3A+Donuts)
>> but no such file ships with drill. The docs themselves mention no place to
>> obtain these files. I’ll open a JIRA unless folks point out the error of my
>> ways…. :^)
>> 
>> * used a random excel spreadsheet I happened to have, converted to CSV and
>> did queries against it.
>> 
>> * observed that a query of  a CSV file that returns nothing results in
>> this exception:
>> 0: jdbc:drill:zk=local> SELECT * FROM dfs.`/Users/kbotzum/Downloads/t.csv`
>> where columns[1] > 2;
>> Error: exception while executing query: DrillBuf(ridx: 0, widx: 0, cap:
>> 0/0, unwrapped: DrillBuf(ridx: 0, widx: 0, cap: 0/0, unwrapped:
>> EmptyByteBufLE)).slice(0, 4) (state=,code=0)
>> while this works:
>> 0: jdbc:drill:zk=local> SELECT * FROM dfs.`/Users/kbotzum/Downloads/t.csv`
>> where columns[1] < 2;
>> 
>> Is this a known issue? If not I’ll open a JIRA.
>> 
>> Keys
>> 
>> p.s. Here is the content of t.csv:
>> 
>> /output/foo/20131101,0,1,fooffx_20131101,0,2,0,6,,
>> /output/foo/20131102,0,1,fooffx_20131102,0,2,0,6,,
>> /output/foo/20131104,0,1,fooffx_20131104,0,2,0,6,,
>> /output/foo/20131105,0,1,fooffx_20131105,0,2,0,6,,
>> /output/foo/20131106,0,1,fooffx_20131106,0,2,0,6,,
>> /output/foo/20131107,0,1,fooffx_20131107,0,2,0,6,,
>> /output/foo/20131108,0,1,fooffx_20131108,0,2,0,6,,
>> /output/foo/20131109,0,1,fooffx_20131109,0,2,0,6,,
>> /output/foo/20131110,0,1,fooffx_20131110,0,2,0,6,,
>> 
>> Keys
>> _______________________________
>> Keys Botzum
>> Senior Principal Technologist
>> kbotzum@maprtech.com
>> 443-718-0098
>> MapR Technologies
>> http://www.mapr.com
>> 
>> 
>> 
>> On Sep 4, 2014, at 2:03 AM, Jacques Nadeau <jacques@apache.org> wrote:
>> 
>>> Hello All,
>>> 
>>> I'd like to propose our third release candidate for version
>>> 0.5.0-incubating.  This release includes >109 JIRAs closed [1].
>>> 
>>> You can find the release artifacts hosted at [2].  Please download and
>> try
>>> them out and cast your vote.
>>> 
>>> This release candidate now separates out class B dependencies in the
>> binary
>>> release into a separate directory and adds extra information in the
>> notices
>>> file to this regard.  Additionally, it includes one memory leak fix that
>>> was identified by one of our regression suites.
>>> 
>>> As always, vote will be open for 72 hours, ending 11PM Pacific, September
>>> 6, 2014.
>>> 
>>> [ ] +1
>>> [ ] +0
>>> [ ] -1
>>> 
>>> When making a vote, please make sure you state whether your vote is
>> binding
>>> or not.  This helps keep things simple.
>>> 
>>> Thanks,
>>> Jacques
>>> 
>>> [1] https://issues.apache.org/jira/browse/DRILL/fixforversion/12324880
>>> [2] http://people.apache.org/~jacques/apache-drill-0.5.0.rc2/
>> 
>> 


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message