@lpl Hi, in my opinion (my opinion tries always to be in favor of democracy) it is better if Pythonista is not so powerful, because otherwise there would be very little chances for other developers. In this way other people can develop apps that try to be closer to users needs (let's say that this creates a good and lively market).
The main vocation of Pythonista is not to be a scientific tool (we all understood it), but an advanced IDE with a lot of features.
omz, the developer is very perfectionist in the things he does (it is my idea) and I suppose he uses a lot of time to plan very well all his development ideas about his app. In other words, I think that to add a c based python library to Pythonista is not so difficult for him, but it is not the main target in his idea about Pythonista and, also, adding c libraries in Pythonista would inevitably give the developer extra responsibility for keeping these libraries up-to-date (it is not to be taken for granted that the updated version of a c-based python library can be easily compiled like the previous one).
The best thing for now could be to install different python IDEs available on Apple Store (everyone has to decide his/her budget: 3$, 15$, 300$, ...) that perform different and specific tasks, and even better if all these apps/IDEs can access to user's iCloud folder, in your device you could run what you want with the IDE that best meets your needs.
If you don't want to install other python apps/IDEs, try to develop something about the sage interface in order to allow user to use pandas in Pythonista console using the remote server but in a way that is as if pandas were really installed in Pythonista, I mean, in order to be able to execute in console each pandas command with any argument and each single pandas command returns the output in console.
I would be grateful to you, because I do not know where to start doing something like that.
Example of session in Pythonista console with pandas available in sage remote server:
>>> data = pandas.read_csv("mydata.csv") ## here Pythonista sends the pandas command to server with the argument (automatically, the argument 'mydata.csv' is obviously in Pythonista folder, in this case it is in main Documents folder, but you could add relative path for other files) and wait for answer in order to save in data the output
>>> data.head() ## here Pythonista print data content