tablib
tablib copied to clipboard
Memory usage of Dataset
We have to build and load the dataset in memory every time, which seems to occupy a lot of memory if data is really huge. Is there any way of improving this?
No, for now tablib is not intended to large datasets.
I agree it is an issue. I'd like to support it for a subset of tablib API, mostly format conversions.
#207 is somewhat related to this issue.
With the respect for tablib, the streaming feature is pending. For now, I would suggest giving it a go with pyexcel. Here is the tutorial on streaming data and the information on the choice of plugins.