Request limit and lapply

Asked Apr 18 '19 at 11:27

Active Oct 09 '20 at 23:49

Viewed 141 times

I want to make thousands of entrez requests. I know that you are not allowed to make more than 3 requests per second unless you have an API key. The rentrez introduction says that rentrez enforces this limit.

I just want to make sure that NCBI does not block me, so the performance is not the issue here.

https://cran.r-project.org/web/packages/rentrez/vignettes/rentrez_tutorial.html

But what if I combine lapply() with entrez_search()?

authors <- c("Uhelski ML", "Manjavachi MN")
authors_pubs <- lapply(authors, function(x) entrez_search(db = "pubmed",term = paste0(x, "[AUTH]"), use_history = TRUE))

Does rentrez still enforces the request limit in this case or do I have to introduce stops into my code?

edited Oct 09 '20 at 23:49

Werner Hertzog

2,002
3
24
36

asked Apr 18 '19 at 11:27

Martin Graßhoff

What happens when you run the above code? – user 123342 Apr 18 '19 at 11:32
It returns a list with a entrez_search result for each author. – Martin Graßhoff Apr 18 '19 at 11:44
1

If any API provider is putting restriction on `rate limit`, you need to respect that. You can do that in your code by introducing some form of `sleep`? – Sonny Apr 18 '19 at 11:44
1

Yes, if I have to I would use Sys.sleep(0.4). The question is do I have to? If rentrez already does this, I would just slow down the code by introducing stops. – Martin Graßhoff Apr 18 '19 at 11:48
2

The answer is yes, you need to add the Sys.sleep function. `lapply` is just a compact loop, thus you are calling the `entrez_search` many many times. – Dave2e Apr 18 '19 at 15:36
OK, but then the question is: What does the vignette mean when it states that rentrez enforces the limit? – Martin Graßhoff Apr 23 '19 at 13:35

Request limit and lapply

0 Answers0