This is strange... The length is not a problem. In unit test, TriDictinoaryTest.partOverflowTest(), it verifies string length over 1000 is no problem.
And I tested the problematic string ' http://www.baidu.com/link?url=ea-fwb14wpouqace1v0glqezcxpco43hdj4g8ms1q2hmrmqysk0ry3npmzubxd59m8mziai1f2ldlobwfpxo4a&wd=&eqid=b2d84a760003b4440000000256264354' on 1.0-incubating branch, it's also fine. 杨海乐, are you convenient to try the latest code on 1.x-staging? If not, then I might need some sample data to reproduce. On Fri, Dec 4, 2015 at 5:37 PM, 杨海乐 <[email protected]> wrote: > is the length of some REF's column too long ? > > > > -- > View this message in context: > http://apache-kylin-incubating.74782.x6.nabble.com/Build-Dimension-Dictionary-Step-Not-a-valid-value-tp2641p2675.html > Sent from the Apache Kylin (Incubating) mailing list archive at Nabble.com. >
