Currently, for each full import of data from Discourse into Graphryder, we have to adjust our settings for API rate limiting in Discourse because otherwise, the import will remain incomplete and consequently, connecting the different records (codes, annotations, users) will fail.
The proper solution would be one of these:
Making Graphryder understand and use the response given by Discourse, which includes an instruction to wait for so-and-so seconds before proceeding.
Providing a configuration parameter in Graphryder that allows to inform it about the Discourse API limits. Graphryder can then space out its requests evenly to avoid hitting this limit.
Currently, for each full import of data from Discourse into Graphryder, we have to adjust our settings for API rate limiting in Discourse because otherwise, the import will remain incomplete and consequently, connecting the different records (codes, annotations, users) will fail.
The proper solution would be one of these:
Making Graphryder understand and use the response given by Discourse, which includes an instruction to wait for so-and-so seconds before proceeding.
Providing a configuration parameter in Graphryder that allows to inform it about the Discourse API limits. Graphryder can then space out its requests evenly to avoid hitting this limit.