Re: Is it time to develope using Python 3K?
We'll see if/when that starts happening. If Python 2 is what companies use in production, their developers will build also new packages to prioritize python 2. On Fri, Jan 16, 2015 at 9:03 PM, Kent Tenney kten...@gmail.com wrote: Anyway inevitably there will be packages not available for python3, (legacy) and newer packages will not be available for python2 (because they leverage py3 features) On Fri, Jan 16, 2015 at 9:28 AM, 'Terry Brown' via leo-editor leo-editor@googlegroups.com wrote: This thread has inspired me to get things working in Python3, i.e. Leo plus all the plugins I use. I have PyQt5 installed in Python3, so this includes fixing Qt4/5 stuff. Notes so far: paramiko is a package for Ubuntu 14.04 for 2.7 but not 3, no big deal, pip3 can install it. I've been loading the deceased plugin 'qtframecommands' from the .pyc file for who knows how long. :-) The livecode plugin uses https://pypi.python.org/pypi/meta, but this happens leo-editor:0 sudo pip3 install meta Downloading/unpacking meta Downloading meta-0.4.1.tar.gz (45kB): 45kB downloaded Running setup.py (path:/tmp/pip_build_root/meta/setup.py) egg_info for package meta Installing collected packages: meta Running setup.py install for meta Installing depyc script to /usr/local/bin File /usr/local/lib/python3.4/dist-packages/meta/bytecodetools/print_code.py, line 12 print instr ^ SyntaxError: invalid syntax odd that pip thinks it's available for py3. I think Leo's livecode plugin would be much better without the meta dependency, it's being used for ast object to source code conversion, maybe livecode could adequately guess at the code from the input code. I.e. given input of d[2][3:7] = 6*7 meta handles the recreation of the d[2][3:7] part on the output side, I think. Anyway inevitably there will be packages not available for python3, not a surprise. Cheers -Terry -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
On Wednesday, January 14, 2015 at 9:31:32 AM UTC-6, Edward K. Ream wrote: On Wed, Jan 14, 2015 at 8:05 AM, Ville M. Vainio vivai...@gmail.com wrote: FWIW, I'm pretty pessimistic about Python 3 at this point. Python 2 seems to be good enough for most people. I am not real optimistic about Python 3 myself. As Kent says, more and more packages are being ported. But Guido has recently promised to support Python 2K until at least 2020 (a change from 2015). This indicates that all is not going well. My guess is that there are some big Python 2K shops that still have no real notion about how they are going to transition to 3K. It doesn't matter how many people *have* made the transition as long as there are important players who haven't or can't. This is a complex topic. Here are some further thoughts: Guido himself clearly believed (and probably still believes) that Python 2 is *not* good enough. Otoh, Guido's remarks in recent PyCon keynote speeches indicate that he will never again attempt such a radical break with existing code. He seems somewhat unhappy with the transition to Python 3. That may be an understatement. Or not. He has reasons to put a brave face on things. Kent's remark that more and more packages are being ported to Python 3 is more important than I originally acknowledged. The available packages, not the problems of big shops, are what most people care about, or should care about. In some ways, the big shops don't matter all that much. They can take care of themselves and they can always stick with Python 2.7. When (not if) Python 2.7 is no longer supported, big shops can start paying the price that the core Python developers are now paying in supporting the Python 2 code base. Finally, there are a growing list of reasons why Python 3 is simply better than Python 2. I actually would *not* say that Python 3's support for unicode is one of those reasons, but that's debatable. What is not debatable is that Python 3 has cool new features and modules that Python 2.8 will *never* have: - My favorite is pip install, the killer feature of Python 3.4. Yesterday I did pip install ipython[all] and everything Just Worked(tm). This is the first time I have *ever* managed to install tornado on Windows, and thus the first time that ipython notebook has ever worked on windows. - The important asyncio module https://docs.python.org/3/library/asyncio.html is Python 3 only. Yes, there is a backport to Python 2: https://pypi.python.org/pypi/trollius - Function annotations, https://www.python.org/dev/peps/pep-3107/, will never be part of Python 2 unless they are supported in Python 2.8. - Similarly, the yield from syntax, pep 380, https://www.python.org/dev/peps/pep-0380/, will never be part of Python 2. You can quibble about how important these features are (except, pip :-), but there is no doubt that Python 2 is a dead end. Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
This thread has inspired me to get things working in Python3, i.e. Leo plus all the plugins I use. I have PyQt5 installed in Python3, so this includes fixing Qt4/5 stuff. Notes so far: paramiko is a package for Ubuntu 14.04 for 2.7 but not 3, no big deal, pip3 can install it. I've been loading the deceased plugin 'qtframecommands' from the .pyc file for who knows how long. :-) The livecode plugin uses https://pypi.python.org/pypi/meta, but this happens leo-editor:0 sudo pip3 install meta Downloading/unpacking meta Downloading meta-0.4.1.tar.gz (45kB): 45kB downloaded Running setup.py (path:/tmp/pip_build_root/meta/setup.py) egg_info for package meta Installing collected packages: meta Running setup.py install for meta Installing depyc script to /usr/local/bin File /usr/local/lib/python3.4/dist-packages/meta/bytecodetools/print_code.py, line 12 print instr ^ SyntaxError: invalid syntax odd that pip thinks it's available for py3. I think Leo's livecode plugin would be much better without the meta dependency, it's being used for ast object to source code conversion, maybe livecode could adequately guess at the code from the input code. I.e. given input of d[2][3:7] = 6*7 meta handles the recreation of the d[2][3:7] part on the output side, I think. Anyway inevitably there will be packages not available for python3, not a surprise. Cheers -Terry -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
PIP is routinely used with python 2 - in fact a lot of modules in PyPI is Python 2 only. You can get it here: https://pip.pypa.io/en/latest/installing.html On Fri, Jan 16, 2015 at 1:20 PM, Edward K. Ream edream...@gmail.com wrote: On Wednesday, January 14, 2015 at 9:31:32 AM UTC-6, Edward K. Ream wrote: On Wed, Jan 14, 2015 at 8:05 AM, Ville M. Vainio vivai...@gmail.com wrote: FWIW, I'm pretty pessimistic about Python 3 at this point. Python 2 seems to be good enough for most people. I am not real optimistic about Python 3 myself. As Kent says, more and more packages are being ported. But Guido has recently promised to support Python 2K until at least 2020 (a change from 2015). This indicates that all is not going well. My guess is that there are some big Python 2K shops that still have no real notion about how they are going to transition to 3K. It doesn't matter how many people *have* made the transition as long as there are important players who haven't or can't. This is a complex topic. Here are some further thoughts: Guido himself clearly believed (and probably still believes) that Python 2 is *not* good enough. Otoh, Guido's remarks in recent PyCon keynote speeches indicate that he will never again attempt such a radical break with existing code. He seems somewhat unhappy with the transition to Python 3. That may be an understatement. Or not. He has reasons to put a brave face on things. Kent's remark that more and more packages are being ported to Python 3 is more important than I originally acknowledged. The available packages, not the problems of big shops, are what most people care about, or should care about. In some ways, the big shops don't matter all that much. They can take care of themselves and they can always stick with Python 2.7. When (not if) Python 2.7 is no longer supported, big shops can start paying the price that the core Python developers are now paying in supporting the Python 2 code base. Finally, there are a growing list of reasons why Python 3 is simply better than Python 2. I actually would *not* say that Python 3's support for unicode is one of those reasons, but that's debatable. What is not debatable is that Python 3 has cool new features and modules that Python 2.8 will *never* have: - My favorite is pip install, the killer feature of Python 3.4. Yesterday I did pip install ipython[all] and everything Just Worked(tm). This is the first time I have *ever* managed to install tornado on Windows, and thus the first time that ipython notebook has ever worked on windows. - The important asyncio module https://docs.python.org/3/library/asyncio.html is Python 3 only. Yes, there is a backport to Python 2: https://pypi.python.org/pypi/trollius - Function annotations, https://www.python.org/dev/peps/pep-3107/, will never be part of Python 2 unless they are supported in Python 2.8. - Similarly, the yield from syntax, pep 380, https://www.python.org/dev/peps/pep-0380/, will never be part of Python 2. You can quibble about how important these features are (except, pip :-), but there is no doubt that Python 2 is a dead end. Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
Further thought - it's still unclear which one is the dead end, python 2 or python 3. There is very little motivation still to prioritize python 3 when python 2 has all the users and developers - everyone ensures their stuff works on python 2, and python 3 is voluntary extra. On Fri, Jan 16, 2015 at 1:20 PM, Edward K. Ream edream...@gmail.com wrote: On Wednesday, January 14, 2015 at 9:31:32 AM UTC-6, Edward K. Ream wrote: On Wed, Jan 14, 2015 at 8:05 AM, Ville M. Vainio vivai...@gmail.com wrote: FWIW, I'm pretty pessimistic about Python 3 at this point. Python 2 seems to be good enough for most people. I am not real optimistic about Python 3 myself. As Kent says, more and more packages are being ported. But Guido has recently promised to support Python 2K until at least 2020 (a change from 2015). This indicates that all is not going well. My guess is that there are some big Python 2K shops that still have no real notion about how they are going to transition to 3K. It doesn't matter how many people *have* made the transition as long as there are important players who haven't or can't. This is a complex topic. Here are some further thoughts: Guido himself clearly believed (and probably still believes) that Python 2 is *not* good enough. Otoh, Guido's remarks in recent PyCon keynote speeches indicate that he will never again attempt such a radical break with existing code. He seems somewhat unhappy with the transition to Python 3. That may be an understatement. Or not. He has reasons to put a brave face on things. Kent's remark that more and more packages are being ported to Python 3 is more important than I originally acknowledged. The available packages, not the problems of big shops, are what most people care about, or should care about. In some ways, the big shops don't matter all that much. They can take care of themselves and they can always stick with Python 2.7. When (not if) Python 2.7 is no longer supported, big shops can start paying the price that the core Python developers are now paying in supporting the Python 2 code base. Finally, there are a growing list of reasons why Python 3 is simply better than Python 2. I actually would *not* say that Python 3's support for unicode is one of those reasons, but that's debatable. What is not debatable is that Python 3 has cool new features and modules that Python 2.8 will *never* have: - My favorite is pip install, the killer feature of Python 3.4. Yesterday I did pip install ipython[all] and everything Just Worked(tm). This is the first time I have *ever* managed to install tornado on Windows, and thus the first time that ipython notebook has ever worked on windows. - The important asyncio module https://docs.python.org/3/library/asyncio.html is Python 3 only. Yes, there is a backport to Python 2: https://pypi.python.org/pypi/trollius - Function annotations, https://www.python.org/dev/peps/pep-3107/, will never be part of Python 2 unless they are supported in Python 2.8. - Similarly, the yield from syntax, pep 380, https://www.python.org/dev/peps/pep-0380/, will never be part of Python 2. You can quibble about how important these features are (except, pip :-), but there is no doubt that Python 2 is a dead end. Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
On Fri, Jan 16, 2015 at 12:13 PM, Ville M. Vainio vivai...@gmail.com wrote: Further thought - it's still unclear which one is the dead end, python 2 or python 3. There is very little motivation still to prioritize python 3 when python 2 has all the users and developers - everyone ensures their stuff works on python 2, and python 3 is voluntary extra. Tell that to the core Python developers. Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
Anyway inevitably there will be packages not available for python3, (legacy) and newer packages will not be available for python2 (because they leverage py3 features) On Fri, Jan 16, 2015 at 9:28 AM, 'Terry Brown' via leo-editor leo-editor@googlegroups.com wrote: This thread has inspired me to get things working in Python3, i.e. Leo plus all the plugins I use. I have PyQt5 installed in Python3, so this includes fixing Qt4/5 stuff. Notes so far: paramiko is a package for Ubuntu 14.04 for 2.7 but not 3, no big deal, pip3 can install it. I've been loading the deceased plugin 'qtframecommands' from the .pyc file for who knows how long. :-) The livecode plugin uses https://pypi.python.org/pypi/meta, but this happens leo-editor:0 sudo pip3 install meta Downloading/unpacking meta Downloading meta-0.4.1.tar.gz (45kB): 45kB downloaded Running setup.py (path:/tmp/pip_build_root/meta/setup.py) egg_info for package meta Installing collected packages: meta Running setup.py install for meta Installing depyc script to /usr/local/bin File /usr/local/lib/python3.4/dist-packages/meta/bytecodetools/print_code.py, line 12 print instr ^ SyntaxError: invalid syntax odd that pip thinks it's available for py3. I think Leo's livecode plugin would be much better without the meta dependency, it's being used for ast object to source code conversion, maybe livecode could adequately guess at the code from the input code. I.e. given input of d[2][3:7] = 6*7 meta handles the recreation of the d[2][3:7] part on the output side, I think. Anyway inevitably there will be packages not available for python3, not a surprise. Cheers -Terry -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
Oh, I didn't understand that the question was whether or not to abandon 2, that wouldn't be good. I'd welcome help from someone running Leo trunk in a virtualenv with Py3 and PyQt5 On Wed, Jan 14, 2015 at 9:26 AM, Edward K. Ream edream...@gmail.com wrote: On Tue, Jan 13, 2015 at 10:06 AM, Kent Tenney kten...@gmail.com wrote: I recently tried to switch to v3, as I remember, I was thwarted by problems having to do with PyQT in a virtualenv. That may be, but I am talking about Leo's core developers, who presumably can create a stable Python3K environment. Otoh, this whole thread is probably a minor, or even trivial question: Leo must support Python 2K for the foreseeable future. Edward Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
On Wed, Jan 14, 2015 at 9:40 AM, Kent Tenney kten...@gmail.com wrote: Oh, I didn't understand that the question was whether or not to abandon 2, that wouldn't be good. The question is merely whether to encourage Leo's core developers to use Python 3 when writing (and checking!) their code. Leo will continue to support Python 2 forever (as long as Python 2 is supported). Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
Re: Is it time to develope using Python 3K?
FWIW, I'm pretty pessimistic about Python 3 at this point. Python 2 seems to be good enough for most people. On Tue, Jan 13, 2015 at 6:06 PM, Kent Tenney kten...@gmail.com wrote: I recently tried to switch to v3, as I remember, I was thwarted by problems having to do with PyQT in a virtualenv. I currently run Leo in a virtualenv, I think I linked to system files to make PyQT work, but I don't remember how I did it. So, I'd like a virtualenv with 3.x Python to run Leo from, but since 'pip install PyQT' doesn't work, I gave up. On Tue, Jan 13, 2015 at 9:54 AM, Edward K. Ream edream...@gmail.com wrote: I would like to have most (all?) Leo developers develop and test using 3.x. Would that be a problem for anyone? Guido has stated that there will be no version 2.8 of Python: http://legacy.python.org/dev/peps/pep-0404/ (Hahaha: Pep 404) although some disagree: http://blog.startifact.com/posts/the-call-of-python-28.html Considering that Guido is the BDFL, I rate the odds of Python 2.8 at less that 1%. I expect to see people using Python 2.x for a long time, but Python 3.x should be available for anyone who wants it, and not just on PythonAnywhere. Afaik, it is now possible to develop Leo using Python 3K exclusively. In particular, Python 3.4 supports pip, a big step forward. For example, the following work:: pip install docutils pip install pylint pip install sphinx But pip does not appear able to install pyqt (I may be mistaken). Anyway, docutils, pylint and sphinx work fine (on Windows) with Python 3K. Your comments, please. Edward -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups leo-editor group. To unsubscribe from this group and stop receiving emails from it, send an email to leo-editor+unsubscr...@googlegroups.com. To post to this group, send email to leo-editor@googlegroups.com. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
