As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
What is better way to install python3.3 on Mac OS X?
via homebrew
or via official installer in http://python.org/download/
Most people working in the Mac world will recommend homebrew or macports, I think.
I personally prefer homebrew. It doesn't install anything other than what you need, and it allows you an easy uninstall option, too. (brew uninstall blah). It also allows you to keep up with updates more easily.
Plus, (and I'm not sure if this is true with macports), you get pip for free.
That being said, you may have to get your hands dirty in your bash_profile and possibly your /usr directory---if you want to type python at the command line, and get your installed version of Python, you'll probably have to change some symbolic links.
I HATE/am not very good at fooling around with system settings, but I had my complete Python installation up and running in less than an hour, and it has been running without any issues for quite a while.
Related
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
I need a noSQL solution that is fast, small, embedded (because I don't have root privilege) and supports Python. For now I just have two names: DyBASE, CodernityDB. Please help me find others and choose the best one. Thank you!!!
ZODB - It is not "small", I guess - but it is fast, it is native to Python - and it can certainly be installed without root privileges using buildout or even only virtualenv -
Here is a blog post showing how ZODB compares against sqlite - (and sqlite, in its turn, for single threaded code, performes quite well against larger SQL database systems - just look for benchmarks):
http://pyinsci.blogspot.com.br/2007/09/zodb-vs-relational-database-simple.html
Here is a link to the tutorial -
for being able to install it as in the tutorial, without adm privileges on the system,
first create a virtualenv and activate it:
http://www.zodb.org/en/latest/documentation/tutorial.html
Here is the link for ZODB and install instructions:
https://pypi.python.org/pypi/ZODB3/3.8.2
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
I am wondering and pondering what python setup to use. Portable in sense of Portable Python in Windows (but i need such for Linux).
I am working on python project with a lot of dependencies , including C/C++ libraries , reuses some perl scripts , and even Java Tika lib to parse multiple document types. I want to put them all in local folder and bundle it so that they do not have to re-install when i am going to set it up on different servers (which have different Linux Distro , Gentoo , Ubuntu and Redhat ).
I wont use python provided by distro because it causes many headaches.
From studying around , these are choices I got :
1 - VirtualENV (But it defeat my purpose as it is not designed to relocate).
2 - Statically Compiling Python (Not sure how it will work, I do not see many success stories on it)
3 - Local (Manually) Compiling Python (Wont work at different Libc version right?)
Any other choices ? Please let me know..
Thanks in advance.
virtualenv is problematic, because it needs the libraries from the computer. Because nowadays most linux distros have python installed by default, the only case that'll be a problem is when the installed version on the computer is different than the one you're using.
However, I believe you can fix that by copying all the libs on PYTHONPATH that are placed on your local installation to the virtualenv folder.
I don't know about 2 or 3, but you could also try to manually compile python to your USB disk and pip your libraries.
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
Although this question is very popular here in StackOverflow, after spending some time here and in the Google, I still haven't find a concrete answer on what is the most appropriate way to do SOAP consuming in Python 3.
I took a look at Does a Python 3 SOAP client module exist?, and I hope it is outdated and today some solution to this may have appeared.
I was thinking about some ideas:
Use 2to3 script to port some existing libraries to Python 3 (SOAPy, suds, etc).
Load an external module, by mixing technologies (Py3k + Jython, Py3k + Python 2.6, etc.)
Write in hardcode Python classes that corresponds to definitions of WSDL files (which implies in tight-coupling/high maintenance).
Write the software in Python 3.0, call the "python2.6-only" module functions through the execnet package. Which requires the Python 2.6 to be installed on the machine and the software written in Python3.0 to be a frozen binary.
Any ideas?
Thanks in advance
I would probably start by trying your suggested 2to3 port. For many things, it works pretty well. It would still be a day or two worth of work to convert something like suds, I imagine.
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 9 years ago.
We think about whether we should convert a quite large python web application to Python 3 in the near future.
All experiences, possible challenges or guidelines are highly appreciated.
My suggestion is that you stick with Python 2.6+, but simply add the -3 flag to warn you about incompatibilities with Python 3.0. Then you can make sure your Python 2.6 can be easily upgraded to Python 3.0 via 2to3, without actually making that jump quite yet. I would suggest you hold back at the moment, because you may at some point want to use a library and find out that it is only available for 2.6 and not 3.0; if you make sure to cleanup things flagged by -3, then you will be easily able to make the jump, but you will also be able to take advantage of the code that is only available for 2.6+ and which is not yet ready for 3.0.
For each third-party library that you use, make sure it has Python 3 support. A lot of the major Python libraries are migrated to 3 now. Check the docs and mailing lists for the libraries.
When all the libraries you depend on are supported, I suggest you go for it.
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
Could someone tell me which is better of the two for bundling Python applications — cx_Freeze or PyInstaller? I'm looking for a comparison based on factors such as:
Popularity (i.e. larger user base)
Footprint of the built binary
Cross platform compatibility
Ease of use
I tried both for a current project and decided to use cx_freeze. I found it easier to get started. It has an option to bundle dependencies in a zip archive, which makes it easy to check that everything was properly included.
I had trouble getting PyInstaller to include certain egg dependencies. It couldn't handle conditional imports as well as I needed and looking through the bundled archive was difficult. On Windows, it requires pywin32 (so it can't be used with virtualenv) and version 1.4 doesn't work with Python 2.6. There's no information on whether Python 2.7 is supported.
Why not use something like GUI2EXE?
GUI2Exe is a Graphical User Interface
frontend to all the "executable
builders" available for the Python
programming language. It can be used
to build standalone Windows
executables, Linux applications and
Mac OS application bundles and plugins
starting from Python scripts.
For my experience, I found that for some programs py2exe doesn't work right, but cx_freeze does. haven't tried pyinstaller.