The performance of those two methods is really really bad when you have thousands of pages. In my case, with 100k pages, painting the paginator takes over 300 ms.
Pagination.iter_pages is slow by itself. Pagination.pages is not that slow but it's called once for every page from Pagination.iter_pages when you consume the iterator, and calling it 100k times really adds up. Python is a very slow language and calling the same method so many times is undesirable even if the method seems simple.
The low-hanging fruit here is caching the result of Pagination.pages. That fixes more than half of the problem (only the call to Pagination.iter_pages remains). I did so with this subclass, which might be helpful to someone. (I'm far from an expert in Python, so this may hurt your eyes)
class MyPagination(Pagination):
def __init__(self, query, page, per_page, total, items):
super().__init__(query, page, per_page, total, items)
self._pages = super().pages
@property
def pages(self):
return self._pagesBut of course a better solution is desirable.
The performance of those two methods is really really bad when you have thousands of pages. In my case, with 100k pages, painting the paginator takes over 300 ms.
Pagination.iter_pagesis slow by itself.Pagination.pagesis not that slow but it's called once for every page fromPagination.iter_pageswhen you consume the iterator, and calling it 100k times really adds up. Python is a very slow language and calling the same method so many times is undesirable even if the method seems simple.The low-hanging fruit here is caching the result of
Pagination.pages. That fixes more than half of the problem (only the call toPagination.iter_pagesremains). I did so with this subclass, which might be helpful to someone. (I'm far from an expert in Python, so this may hurt your eyes)But of course a better solution is desirable.