featurebasedb / featurebase-docs Goto Github PK

This is a note-to-self because internet access is too flaky to load Jira.

Multiple have virtually identical content.
Need to get them into a single include if at all possible
Then need to explain why this is a bad idea to whomever put them together.

• timestamp-timeunit-table.md
• datetimeadd-function-timeunit-table.md
• datatimename-function-timeunit-table.md
• timeunit-ttl-args.md

In the Cloud Quick Start Guide, the Complex Group By Query SELECT hobbies, COUNT(*) as cnt FROM cseg GROUP BY hobbies HAVING COUNT(*) > 200000000 ORDER BY cnt DESC; returns an error. This only runs for me when I remove the last filter of ORDER BY cnt DESC

Please run the below sample,
create table account(_id id, username string, salary int, balance decimal(2), account_type stringset);

BULK INSERT INTO account(_id, username, salary, balance, account_type) MAP(0 id, 1 STRING, 2 int, 3 decimal(2), 4 stringset) FROM '/home/account.csv' WITH BATCHSIZE 100000 FORMAT 'CSV' INPUT 'FILE';

account.csv
12,erin,85000,334.43,"checking" 12,erin,85000,334.43,"savings" 12,erin,85000,334.43,"test" 13,will,90000,111.13,"savings" 14,john,35000,999.22,"savings" 15,paul,50000,444.53,"checking,savings" 16,mary,40000,555.63,"checking,savings" 17,jack,95000,888.73,"savings" 19,dogg,85000,777.83,"checking,savings" 21,zoom,15000,222.93,"checking"

expected output for decimal(am showing first row alone)
12 | erin | 85,000 | 334.43 | checking, savings

but getting below result(Please check the decimal point of balance column)
12 | erin | 85,000 | 3.344 | checking, savings

Run below insert command to check the stringset issue
INSERT into account (_id, username, salary, balance, account_type) values (19, 'dogg', 25000,2.24, ['checking']);

expected output for stringset
19 | dogg | 85,000 | 7.778 | checking,savings

But getting below result
19 | dogg | 85,000 | 7.778 | checking, checking,savings

Kindly let us know how to frame stringset in single row in CSV file, Also please share how to import "2019-05-22T13:44" CSV data in timestamp bulk import

Thanks in advance

Currently, the docs point to either use:

This should be --output rather. Since if you use the flag: -output it will backup to the utput directory.

I've run into a few problems getting Jekyll running locally and I thought it would probably make more sense long term to get a container running the docs for reviews and authoring locally. At the least, @ljsinclair has had some challenges with this and I know setting it up on Windows is a pain...

• build a docker container for deploying documentation
• container checks the code out (requiring this repo be public sooner than later)
• ensure it runs on all major OS platforms

Join the FeatureBase Discord channel to discuss issues and get support

On the Quick Start Guide for FeatureBase Community on Mac OS I had run into a few issues.
This command line to untar install files to FB directory does not seem to work, but it runs when I adjust it slightly:

• old: tar xvfz featurebase--community-.tar.gz -C ~/featurebase
• new: tar xvfz featurebase---.tar -C ~/featurebase

The command to authroize FB with gatekeeper system raises an error and I was wondering if I should try without the "community" in the command since that worked earlier:

• command: mv featurebase-*-community-darwin-arm64/ opt
• error raised: zsh: no matches found: featurebase-*-community-darwin-arm64/