In #410 we implemented a throttling mechanism for dialing back the number of GitHub requests that Rosetta makes when it gets close to max per-hour API interaction limit. One thing we did not do was investigate whether we could do fewer of these API calls in general and thus make it so that we would hit this limit less often. I have seen the warning message that is shown to the user when the throttling is happing twice in the last couple of days, one because of testing that I was doing, but the other probably just due to normal translator interactions. We should probably take some time to see if the interaction can be made more efficient so that the throttling is less likely to kick in.
In #410 we implemented a throttling mechanism for dialing back the number of GitHub requests that Rosetta makes when it gets close to max per-hour API interaction limit. One thing we did not do was investigate whether we could do fewer of these API calls in general and thus make it so that we would hit this limit less often. I have seen the warning message that is shown to the user when the throttling is happing twice in the last couple of days, one because of testing that I was doing, but the other probably just due to normal translator interactions. We should probably take some time to see if the interaction can be made more efficient so that the throttling is less likely to kick in.