Howdy; I'm working with the

tap-gitlab

extractor (

meltanolabs

variant) and I'm interested in specifying different replication methods for different GitLab selectors. Something along the lines of "load the projects table via truncation, but load the pipelines and jobs tables via an incremental approach." How would I configure this / is this even possible?

Hi @warren_ersly! It might be possible, depending on which target you're using. target-bigquery supports a replication method setting, which you could configure differently for different instances of tap-gitlab using inheritance

Hi @edgar_ramirez_mondragon - thank you! This worked very well on handling the table that needed to be truncated. The "incremental" tables are appending also, which is partially the desired behavior. However, to really make things incremental in the true sense of the word (ie, append only the latest records since the last run), I think I'm going to need to do some manipulation of the state file alongside each run, incrementing the time in the file to that of the run...otherwise I'm duplicating records, not to mention making a lot of useless GitLab API calls. Does this sound like a reasonable approach to you? Or is there some other feature I'm not aware of that can handle this case?