Re: [Gnumed-devel] simplified GUI <-> database interaction

[Horst Herb]

Hilmar Berger wrote:

> if we have to do all backend communication 'by hand'. Hiding
> implementation features in hierarchies of objects has certainly a
> perfomance cost but simplify coding a lot. If, however, the costs to

This was my initial thought. However, considering the time I spent 
designing abstraction layers I must say that in the same time I would 
have written the straightforward solutions twice.
The code didn't get simpler either - the opposite is the case.

The main problem is our high degree of normalisation and our wish to 
remain language independend on the backend.
The normalisation makes it almost impossible to autocreate the python 
objects from backend tables; putting rules into tables regarding this 
would make the backend structure even more awkward
Implementing persistent python objects in a straightforward way on the 
backend side makes it difficult to access the data from other languages 
(like web clients)

> lot I would rather suspect that Python is not the langugage of choice to
> code a project like gnumed.  

I still think it was and is the right choice. No other language allows 
that fast prototyping, and with no other language I spent that little 
time debugging (with Python I spend about 10% of the time in debugging - 
in C++ it is more than 50% of the time). Few languages are that 
portable, but probably no other one that easy to learn.

> How high are the performance hits you have found ? What is the part of the
> abstraction layers having the highest perfomance costs ?

Just as an example: the simple conversion of returned rows into 
dictionary-ready objects by PyPgSQL costs about 30% in performance. The 
individual performance penalties are small, but they do add up.
0.001 seconds is nothing, but 1000 times 0.001 seconds is a pain.

> Are you sure there is nothing we can do to keep at least a minimal
> abstraction layer ?

Oh yes - gmPG already provides a high degree of abstraction! It 
completely hides the implementation of the service concept. There will 
still be cached objects - but limited to the ones that are likely to be 
reused again and again during the "life cycle" of a patient record in 
the user interface

That said - I am open for suggestions and would welcome whoever proves 
me wrong.

Horst