englefly commented on PR #10633: URL: https://github.com/apache/doris/pull/10633#issuecomment-1176975418
> 1. I think it's a really bad idea to initialize a look-up table from an explicit-defined data file. Could we do this job in BE process initialization phase? (e.g. doris_main.cpp) A: using data file is better - it save launch time, meanwhile we have a bigger BE binary (80kb) - Since we do not need to add more source code to generate the table, saves maintain effort. > 2. Using two look-up table with same keys also need to be re-thinking. Maybe a look-up table with mapping a key to a pair of value is enough? A: the table dimension from high to low:mode->year->month->day is better than that of year->month->day->mode, because former is more locality friendly. > 3. I noticed there still are much unused codes. It seems using in your developing phase. Before merging to mater, it should be removed. A: these codes are used to generate look-up table. As mention in comments, they are attached as comments to help us update the pre-calculated years. For example, if we want to pre-calculate year from 1970 to 2050. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org For additional commands, e-mail: commits-h...@doris.apache.org
