The patch looks great for me. Thank you for nice contribution. There are a couple of issues that we need to discuss.
Firstly, I wonder whether 'Download to CSV feature' reads data on HDFS or only outputs the data of HTML page? Unless the output table is explicitly given, the result data stored on HDFS will be automatically removed after a while. If Download feature is based on HDFS, it is only available for a while. Next and previous buttons also have the same problem. So, I think that Tajo web UI should show some part whose volume can be shown in the web page in the case where output data is too large.
Second, google js works only online. However, The environments where Tajo usually is used have local networks that cannot access to Internet. So, the chart included in the patch won't work in those environments.
When it comes to code, there are some bugs. 'Download feature' leads to wrong results. I've just experienced the number of results are fewer than that of the correct result. I'm expecting that the first row is omitted.
In addition, two jsapi.js files are included in the patch.
Thank you for your contribution.