I am using PyMuPDF, and it seems that there is a memory leak after opening and closing files.

[xiaolibuzai-ovo]

code:

As shown in the figure: I download a PDF file via HTTP and load it.
When I repeat the process continuously, the memory usage keeps increasing.

platform: linux
version: 1.25.5

thanks for u help！

[JorjMcKie]

Part of this is normal and intended: the base library caches data in n effort to increase speed. Please try the following two experiments to isolate the cause:

1. Do not open the downloaded stream as a pymupdf Document - i.e. comment out doc=pymupdf.open() and doc.close()
2. Add the following statement after doc.close(): pymupdf.mupdf.fz_empty_store(). This will empty this cache.

[JorjMcKie]

Just to clarify:
The two points in previous post are intended to isolate the problem. If experiment 1. also shows a memory increase, then we need to take it into account.
I also need to know whether Experiment 2. will resolve the problem.

[jamie-lemon]

@xiaolibuzai-ovo If required please continue this discussion on our forum at https://forum.pymupdf.com , we are trying to move all discussions there! :)