It would be great if scraper could store the results of scraping without sending them to the Typesense server right away. In my scraping the Typesense communication takes the vast majority of time. Setting up a new HTTP connection to /documents/import for every single scraped website is inefficient. Caching the results and sending them in bulk at the end of scraping would speed up the process significantly and potentially make it more reliable by requiring fewer successful connections to Typesense.
This could also open new possibilities around the cache if it was stored in a local file. E.g. it could be re-submitted later if there's a network problem or submitted to another instance of Typesense.
It would be great if scraper could store the results of scraping without sending them to the Typesense server right away. In my scraping the Typesense communication takes the vast majority of time. Setting up a new HTTP connection to
/documents/importfor every single scraped website is inefficient. Caching the results and sending them in bulk at the end of scraping would speed up the process significantly and potentially make it more reliable by requiring fewer successful connections to Typesense.This could also open new possibilities around the cache if it was stored in a local file. E.g. it could be re-submitted later if there's a network problem or submitted to another instance of Typesense.