Python is itself written in "C", so integrating by either "gluing" C to python (extending python), or embedding the python interpreter right in your C source code. ;-) , is relatively speaking, extremely easy. And if you happen to be one of those c++, java .net, etc, etc.. guys/gals, well that's available too. Right now I'm working on finding/writing a C source code generator (not for gluing) in python, so I don't have to write all that damn C verbosity by hand.
Btw, In case you've missed it from my older postings, I highly recommend Pytables for your data base. Especially if you plan to use NumPy or Numarray Nd array/table interfaces. Don't forget about SciPy for a whole other conglomeration of mathematical computation.
Does anyone have a favorite IDE? I have used idle so far, which has been adequate for my newbie needs, but I am wondering what I have been missing. I fired up Stani's Python Editor and it looks like there are some nice features.
Wow brokershopping your striking a major chord of mine... finding the most suitable ide. An IDE is probably the most personal thing when programming, because, at least for me, it's gotta fit like a glove. Therefore it's very hard to find one that suits your personal preferences. This IDE has this fantastic feature, but it's lacking the fantastic feature that this other ide has. DAMN. etc.... SPE is very nice but it's not my glove. Eric3 is a very nice ide, but is not my glove. One that is very simple (light weight on the bells and whistles) but is very robust (autocompletion/calltips/run command and integrated shell) in pythonWin (Win32). But that's out for me because it's windoz only. The IDE (python) that fits me like a glove (nearly) is DrPython. Why? because it's a plugin framework... allows you to plugin the features you want. These plugins can be very easily programmed by you as well. There are some constructive criticisms though as well (aren't there always). Naturally OpenSource developers are providing there work for free, so who can actually criticize. But my biggest peeve, is that I think that there should be a underlying sense of responsibility by developers to provide well thought out documentation. It really bugs me generally. In fact at this very moment I'm working on improving PyTables docs, which relatively (OpenSource or Commercial docs) aren't bad at all. DrPython is by far not the worst offender of inadequate documentation and it's easy enough to figure out. So for the reasons mentioned, it's my recommendation. Lastly, from python's home page, you can get to the wiki which has a page listing all the IDEs by operating system and whether it's commercial or free. kt
Python is a script language, and would not run as fast as C/C++. Why would you want your programs to execute slower?...besides I hear that python has bad memory management..or something like that. - nathan
It might have something to do with the cost of programmer time vs cost of machine time. It entirely depends on what you are doing. If it is fast enough and likely to remain so with forseeable change of circumstances, then one should obviously use the tool that gets the job done sooner.
By the same argument, there are measurable performance differences between pure C and C++ (5-8% by some benchmarks), not to mention systems wrote in pure Assembler vs C (approximately 10-15 has been measured). I am not an advocate of Python by any means (my personal scripting language remains Tcl), but there are plenty of applications where a scripting language like Python / Perl / Tcl can be very useful, not having to manage memory can be a big plus in simplifying the code base. Just to clarify, I code mostly pure C. But it is mostly due to personal preference than anything else.
SAGE looks interesting. Has anyone used it or heard anything about it? It looks like a python-centered shell that pulls together a lot of major open-source projects and provides an interface to many commercial apps. http://modular.math.washington.edu/sage/features.html
Interesting topic on Python. Most of my rudimentary Python dabbling has been webscraping EOD quotes to store data into ascii files and some simple indicators analogous to using Excel formulas. Now I appreciate the discussion about Pytables, and that will be my next step. But can all experienced folks provide any more insight on charting software packages that work with Python, with a penchant for candlesticks, i.e. pros and cons of the various packages? I am aware of gnuplot, matplotlib, and chaco, but I am wondering if there are are others out there. Any feedback would be appreciated, before I arbitrarily commit to one. TIA, dsss27