@Nirav, I tried your query on the sample dataset you sent, with the latest Drill master branch. The query is successful, either as a standalone query, or embedded in a CTAS.
Can you please double check if you have JDK install on "every" node in your drill cluster? The run-time code generation and compilation happen in execution phase. This means any drillbit assigned to execute the Project operator would have to go through code compilation, requiring to have JDK to compile the large generated code for your query. It's not sufficient to only have JDK installed on the foreman node. On Wed, Dec 9, 2015 at 2:17 AM, Boris Chmiel <boris.chm...@yahoo.com.invalid> wrote: > Thanks Nirav,I did the same. > I think the limitation, in my case, is due to the number of cast / > regex_replace / substr etc to handle in the query, not the number of Kline of > code > Boris > > > Le Mercredi 9 décembre 2015 6h41, Nirav Shah <nirav.s...@games24x7.com> a > écrit : > > > Hi Boris, > > As a workaround I broke my query and rejoin results. In case need any help > on this let me know. > > Best, > Nirav > > On Sat, Dec 5, 2015 at 9:17 PM, Boris Chmiel <boris.chm...@yahoo.com.invalid >> wrote: > >> Hello >> I am getting the same error on huge queries. My query is about 10K >> characters. This error is a regression and is directly related to 1.3 >> since I have already ran queries more than twice biggers (nearly 30K >> characters) with 1.1 and 1.2 I'm downgrading to 1.2 >> >> Boris >> >> >> >> >> Le Vendredi 4 décembre 2015 12h42, Nirav Shah < >> nirav.s...@games24x7.com> a écrit : >> >> >> Hello, >> >> I am getting below error while running big query. >> >> ======================================= >> >> Error: SYSTEM ERROR: CompileException: File >> 'org.apache.drill.exec.compile.DrillJavaFileObject[ProjectorGen2825.java]', >> Line 5799, Column 17: ProjectorGen2825.java:5799: error: code too large >> public void doEval(int inIndex, int outIndex) >> ^ (compiler.err.limit.code) >> >> >> Regards, >> Nirav >> >> >> >> > >
