1
76%
master: 76%

Ran 23 Nov 2018 09:27AM UTC

Files 1016

Run time 1min

Badge

Embed ▾

Committed 23 Nov 2018 09:27AM UTC coverage: 79.572% (-0.03%) from 79.599%

Job # 1306.1

Build Type

push

jenkins

Committed by ravipesala

Commit Message

[CARBONDATA-3101] Fixed dataload failure when a column is dropped and added in partition table

Problem: Data load is failing when a column in partition table is dropped and recreated because fieldConverters are created based on schemaOrdinal and this may be different than the order in which data is read from the csv.
For example --> columns = nonparcolumn string, parcolumn int
Now if the user drops and adds column the nonparcolumn then the fieldConverters would be created in the following order
fieldConverters[0] = MeasureFieldConverter
fieldConverters[1] = NonDictionaryFieldConverterImpl

The data read from the csv would be 'columndata', 1. The conversion for nonPartition column would fail i this case.

Solution: Add the partition column to the last while doing alter add column and creating fieldConverters.

This closes #2923

Coverage Stats

61550 of 77351 relevant lines covered (79.57%)

1.05 hits per line

apache / carbondata / 1306 / 1
76%
master: 76%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 1306.1

apache / carbondata / 1306 / 1 76% master: 76%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 1306.1

apache / carbondata / 1306 / 1
76%
master: 76%

README BADGES
x