Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Load additional fields and/or tables into QV
  • Prepare the data so it is ready for Squirro
  • Store the required data into CSV file(s) using the STORE command
  • Drop the data not needed for the QV Dashboard using the DROP FIELDS command, to keep RAM consumption to a minimum.
  • Prepare the data so it is ready for the QV Dashboard
  • Store the finished data in QVD file(s) using the STORE command
  • Import the CSV file(s) into Squirro using an executable provided by Squirro (not available yet) with the EXECUTE command.
  • Optionally delete the CSV filsfiles(s) from the disk

Note

The actual bulk load loader command line tool is documented here: Squirro Bulk Loader Guideon the page /wiki/spaces/WIP/pages/2949365.

Working Example Script

(warning) There are still some encoding issues that need to be taken care of. QlikView actually writes a BOM!

Code Block
SET DecimalSep='.';
SET MoneyThousandSep=',';
SET MoneyDecimalSep='.';
SET MoneyFormat='$#,##0.00;($#,##0.00)';
SET TimeFormat='h:mm:ss TT';
SET DateFormat='M/D/YYYY';
SET TimestampFormat='M/D/YYYY h:mm:ss[.fff] TT';
SET MonthNames='Jan;Feb;Mar;Apr;May;Jun;Jul;Aug;Sep;Oct;Nov;Dec';
SET DayNames='Mon;Tue;Wed;Thu;Fri;Sat;Sun';


//load both structured and unstructured data from our sources
Storm:
LOAD BEGIN_YEARMONTH, 
     BEGIN_DAY,
      BEGIN_TIME, 
     END_YEARMONTH, 
     END_DAY,
      END_TIME, 
     EPISODE_ID, 
     EVENT_ID,
      STATE,
      STATE_FIPS,
      YEAR,
      MONTH_NAME, 
     EVENT_TYPE, 
     CZ_TYPE,
      CZ_FIPS,
      CZ_NAME,
      WFO,
      BEGIN_DATE_TIME,
      CZ_TIMEZONE,
      END_DATE_TIME, 
     INJURIES_DIRECT, 
     INJURIES_INDIRECT,
      DEATHS_DIRECT, 
     DEATHS_INDIRECT, 
     DAMAGE_PROPERTY,
      DAMAGE_CROPS, 
     SOURCE, 
     MAGNITUDE,
      MAGNITUDE_TYPE, 
     FLOOD_CAUSE, 
     CATEGORY,
      TOR_F_SCALE,
      TOR_LENGTH,
      TOR_WIDTH,
      TOR_OTHER_WFO, 
     TOR_OTHER_CZ_STATE, 
     TOR_OTHER_CZ_FIPS,
      TOR_OTHER_CZ_NAME,
      BEGIN_RANGE,
      BEGIN_AZIMUTH,
      BEGIN_LOCATION,
      END_RANGE,
      END_AZIMUTH, 
     END_LOCATION, 
     BEGIN_LAT,
      BEGIN_LON, 
     END_LAT, 
     END_LON,
      EPISODE_TITLE, 
     EPISODE_NARRATIVE, 
     EVENT_NARRATIVE,
      LAST_MOD_DATE, 
     LAST_MOD_TIME, 
     LAST_CERT_DATE,
      LAST_CERT_TIME,
      LAST_MOD,
      LAST_CERT,
      ADDCORR_FLG, 
     ADDCORR_DATE
FROM
Z:\Desktop\squirro_bulk_loader\storm.xls
(biff, embedded labels);


//store fields required by squirroSquirro into a CSV file
STORE EPISODE_ID, EVENT_ID, EVENT_TYPE, BEGIN_LOCATION, END_LOCATION, EVENT_NARRATIVE, EPISODE_NARRATIVE from Storm into squirro.csv (txt);


//remove unstructured data fields, do they don't take up space in ram and on disk
DROP FIELDS EVENT_NARRATIVE,EPISODE_NARRATIVE;


//store the remaining data into QVD for use in QV dashboards
STORE * from @2 into qlikview.qvd (qvd);


//call a squirroSquirro helper tool to push the CSV file into the Squirro index.
EXECUTE cmd.exe /C "C:\Users\Alexander Sennhauser\Downloads\bulk_loader\squirro_bulk_loader.exe" --local --user-api-url http://172.16.148.1:20009 --topic-api-url http://172.16.148.1:20000 --bulk-api-url http://172.16.148.1:20021 -vv --log-file c:\tmp\squirro.log -t cbc5e25fffa21bf68d3383bd84e8764fd2804c75e73149e4317691dcfd2b5b4680905089060264ecc3af6f9379afcce015197692593a7ed9e50938c1f981c csv --infile squirro.csv;