Hey /prog/, I'm a biochemist building a website for a Biology project, and I'm going to have to include NCBI's BLAST+ tools in it eventually. So I was just wondering, would it be better to use the c++ tools that they offer, or would it be better to try to re-write everything in python, since that's what the rest of the backend is written in?
What >>3 said. There are more suitable tools than SEPPLES AND FIOC, but if you don't know any and aren't willing to learn anything, you might as well pick what you're more comfortable with.
Alright, thanks, I was just wondering if there will be any real significant differences in the speed, errors, ease of use on the user end, etc. But if both are equally terrible that answers my question too.
>>5
Obviously the answer is because I'm not a very good biologist.
>>7
Of course SEPPLES will be faster than FIOC in most implementations, but SEPPLES comes with extra complexity which can be tiring. If you only know those languages, you could for example write the resource intensive parts in the lower-level language (SEPPLES), and the scripting parts in the higher-level language (FIOC). It's all about the trade-offs.
>>7 But if both are equally terrible that answers my question too.
They aren't. Python is light beer terrible. C++ is urine after eating asparagus and drinking coffee terrible. That may sound like the same thing, but I assure you it's not.
Nothing significantly wrong with Python. But it should go without saying that if you have to ask, you're not even going to be able to port the provided tools, and even if you were, it would almost certainly be a terrible idea.
Python is a poor man's Scheme with shitty syntax but a very rich man's library and support. It's not a bad choice for getting things done. If you're almost as good at Python as you are at C++, listen to >>12 and do Python. It'll be far more productive and powerful abstraction-wise, and far less full of shit.
Actually, Python doesn't follow all trends (though parts of the standard library did/do follow some Java disease), and it's not really like RoR. The main problem with Python is that instead of getting the best possible language (with such an awesome library and support, I have to admit), you get the most Pythonic language, for an arbitrary, stupid definition of Pythonic only Guido and his gang of idiots know and care for.