I'm learning about Python Packages from Learn Python the Hard Way and one of the exercises it says:
Put a script in the bin directory that you can run
To me, it seems kind of vague. I'm not exactly sure what kind of scripts would go into the bin folder. The Hitchhiker's Guide to Packaging says
put into bin any scripts you’ve written that use your package and which you think would be useful for your users. If you don’t have any, then remove the bin directory.
But I'm still left wondering what kind of script would go in there. So, I know its may sound like a dumb question, but can someone give me an example of when, and why one would put "a script" in their package's bin folder?
I just recently got through Ex46 in LPTHW myself. Like you, I was confused by the scripts. In case the other answer was too advanced for you, I ended up just putting in a simple "hello world" script:
#!/usr/bin/env python
from test3 import printstring
printstring.printstring("test script working")
print "test over"
I named that file testscript3.py (*Note, I learned later that it would be more convenient to leave off the .py filename extension if it were a real script that I wanted to seem like a system command)
My file test3.py was like so:
def printstring(s='you did not provide string'):
print s
Here's some newbie things that I learned while trying to get this process to work:
The #! symbol is sometimes pronounced shebang and the simple explanation is that the command on that line tells the shell to use python to run the script. If you leave off the ".py" filename extension, then the user of the script doesn't need to care what interpreter is needed to run the script. See wikipedia shebang article.
I ran the following command to package the distribution:
python setup.py sdist
After doing that, I was able to install the package and script by running
sudo pip install test3-0.1.tar.gz
One thing I worried about was permissions on the script file. However, I noticed that distutils took care of this when packaging (changed mode to 755 or whatever).
You can find my whole project for this example on github.
For example Django's project creating, Scrapy's project creating, django-admin.py and scrapy are both scripts in bin folder.
You could get even more examples by checking almost python-based tools.
Related
I have been using PyCharm since I began learning Python because of its amazing UI that helped me learn a great deal about the language. As I progress into more advanced projects, I am beginning to prefer using a text editor / command line combo so that I can build my own venv's and have better access to source control. My question is, how does PyCharm manage custom local packages that I created so that I can import them wherever I want in the directory? For instance a project that I have built exclusively in PyCharm that runs will raise numerous import errors when trying to run that same project in VS code, or even a command line shell (yes I did have the PyCharm created venv activated before running on both attempts). For further examples, here is the project structure I am confused about:
RootDirectory
package_1_folder
__init__.py
pckg_1_class.py
program_using_pckg_1_folder
class_using_pckg1class.py
venv
The above structure has no issues being imported and used in PyCharm, however VS code / Sublime when used with command prompt / gitbash will raise either an ImportError or a ModuleNotFound error. I have even gone as far as adding the desired packages to my laptops windows PATH, using sys.path.append (I know this is not good practice I was only trying to get it to work), and even modified the .pylintrc file with the project path with no success. Any help explaining why these errors are happening would be greatly appreciated :)
NOTE:
I have been able to use the packages in VS code as long as the program importing the module is located at the root directory level, but not in its own folder in the root directory. Again, this statement WILL work in PyCharm, I just want to know how PyCharm is able to achieve this.
After numerous attempts to locate how the system was keeping tack of module within the IDE, I found that my answer was not visible from the IDE. I found a .idea folder in my root directory that contains a few .xml documents that manage the directory including where to read modules from.
I am doing an application in Excel and I'd like to use python language. I've seen a pretty cool library called xlwings, but to run it a user need to have python installed.
Is there any possibility to prepare this kind of application that will be launch from a PC without Python?
Any suggestion are welcome!
A small workaround could be to package your application with cx_freeze or pyinstaller. Then it can run on a machine without installing python. The downside is of course that the program tend to be a bit bulky in size.
no. you must need install a python and for interpreting the python function etc.
It is possible using xlloop. This is a customized client-server approach, where the client is an excel .xll which must be installed on client's machine.
The server can be written in many languages, including python, and of course it must be launched on a server that has python installed. Currently the .xll is available only for 32 bits.
This might not be exactly what you're looking for (i.e.- "without python"), but you could ship it as a virtualenv .zip folder with all the necessary packages and respective python installation included in the folder... and also include a run.command bash executable inside the folder that the user can click on to run your program. You'll have to tell the user (inside a README.txt or something) how to enable the executable... For a particular application mine looked like...
Please complete the following steps after installation of MyApplication.zip:
Step 1: Move the MyApplication folder to your Desktop.
Step 2: Open your Terminal in Go then Applications then Utilities.
Step 3: In the Terminal, paste cd ~/Desktop/MyApplication
then press Enter.
Step 4: Then, paste chmod +x run.command
then press Enter.
From now on, clicking the “run” file in MyApplication will run the application.
If clicking the “run” file still doesn’t work, then do…
Click Ok on the popup then Open System Preferences then Security & Privacy
then Click on Open Anyway button then Click Open button on the popup.
NOTE: Please do not move or delete the MyApplication folder, or any of its files.
Then my run.command executable looked like the following:
cd ~/Desktop/MyApplication
PATH=$PATH:$HOME/Desktop/MyApplication/lib/python2.7/site-packages:$HOME/Desktop/MyApplication/lib/python2.7/site-packages/aeosa:$HOME/Desktop/MyApplication/lib/python2.7/lib-dynload:$HOME/Desktop/MyApplication/lib/python2.7/lib-old:$HOME/Desktop/MyApplication/lib/python2.7/lib-tk:$HOME/Desktop/MyApplication/Extras/lib/python:$HOME/Desktop/MyApplication/lib/python2.7/plat-mac/lib-scriptpackages:$HOME/Desktop/MyApplication/lib/python2.7/plat-mac:$HOME/Desktop/MyApplication/lib/python2.7/plat-darwin:$HOME/Desktop/MyApplication/lib/python2.7:$HOME/Desktop/MyApplication/lib/python27.zip
source bin/activate
python main.py
Three things are important here:
1) In run.command, I first change directory into the virtualenv I created for this project. I'm not trying to be dynamic here, since this is all just a quick hack of a solution, so as part of the README.txt, I simply tell the user to put the folder they downloaded onto their desktop. Otherwise, the cd command wouldn't work.
2) I set the client's PATH variable, separating entries with colons. You can see a list of the paths you'd need using a quick check on your computer...
import sys
for path in sys.path:
print path
3) I then call source bin/activate to activate the virtualenv.
4) I then execute my main.py file, which is in the top level of my virtualenv, along with run.command.
5) I only wrote this bash file for Mac OS X, so I don't know how it would translate to windows PCs. I think the only thing that'd change is that the path literals would need to use backslashes instead of forward slashes, but again, I haven't done this for windows so I'm uncertain.
Read more about setting up a virtualenv here: https://virtualenv.readthedocs.org/en/latest/index.html
Finally, note that you will not be able to send your final, working .zip folder over most email services, since they tend to prevent people from sending executables back and forth, so you'll have to use a service like Dropbox or something to send it to people.
Here is the problem I am trying to solve. I don't have a specific question in the title because I don't even know what I need.
We have an ancient Hadoop computing cluster with a very old version of Python installed. What we have done is installed a new version (2.7.9) to a local directory (that we have perms on) visible to the entire cluster, and have a virtualenv with the packages we need. Let's call this path /n/2.7.9/venv/
We are using Hadoopy to distribute Python jobs on the cluster. Hadoopy distributes the python code (the mappers and reducers) to the cluster, which are assumed to be executable and come with a shebang, but it doesn't do anything like activate a virtualenv.
If I hardcode the shebang in the .py files to /n/2.7.9/venv/, everything works. But I want to put the .py files in a library; these files should have some generic shebang like #!/usr/bin/env python. But I tried this and it does not work, because at runtime the virtualenv is not "activated" by the script and therefore it bombs with import errors.
So if anyone has any ideas on how to solve this problem I would be grateful. Essentially I want #!/usr/bin/env python to resolve to /n/2.7.9/venv/ without /n/2.7.9/venv/ being active, or some other solution where I cannot hardcode the shebang.
Currently I am solving this problem by having a run function in the library, and putting a wrapper around this function in the main code (that calls the library) with the hardcoded shebang in it. This is less offensive because the hardcoded shebang makes sense in the main code, but it is still messy because I have to have an executable wrapper file around every function I want to run from the library.
I would change the environment variable PYTHONPATH and also the environment variable PATH. Point PYTHONPATH to your virtual environment and PATH to the directory that contains your new python executable, and make sure the path to your python executable comes first.
I accepted John Schmitt's answer because it led me to the solution. However, I am posting what I actually did, because it might be useful for other Hadoopy users.
What I actually did was :
args['cmdenvs'] = ['export VIRTUAL_ENV=/n/2.7.9/ourvenv','export PYTHONPATH=/n/2.7.9/ourvenv', 'export PATH=/n/2.7.9/ourvenv/bin:$PATH']
and passed args into Hadoopy's launch function. In the executable .py files, I put the generic #!/usr/bin/env python shebang.
Not a major issue but just an annoyance I've come upon while doing class work. I have my Notepad++ set up to run Python code straight from Notepad++ but I've noticed when trying to access files I have to use the full path to the file even given the source text file is in the same folder as the Python program being run.
However, when running my Python program through cmd I can just type in the specific file name sans the entire path.
Does anyone have a short answer as to why this might be or maybe how to reconfigure Notepad++?
Thanks in advance.
The problem is that your code is assuming that the current working directory is the same as the script directory. This is not true in general. Of course it is true if you're in a cmd window, and you cd to the script directory before running it.
If you don't want to rely on that (e.g., because you want to be able to run scripts from Notepad++, or directly from Explorer), what you want to do is use the script directory explicitly. For example:
import os
import sys
scriptdir = os.path.abspath(os.path.dirname(sys.argv[0]))
with open(os.path.join(scriptdir, 'myfile.txt')) as f:
# etc.
If you have a ton of files that your scripts reference in a ton of places, it might be better to explicitly set the working directory. Just add one line:
os.chdir(scriptdir)
For anything beyond quick&dirty scripts, it's usually better to build an installable package and use pkg_resources to access the data files. Read the Tutorial on Packaging and Distributing Projects for more details. But as long as you're only hacking up scripts to help you maintain your specific system, the scriptdir solution is workable.
In the properties of the shortcut that you use to start Notepad++, you can change its working directory, to whichever directory you're more accustomed to starting from in Python. You can also begin your python program with the appropriate os.chdir() command.
I have been putting my code on github, but I've run into an implementation snag. I run the same code on many computers (including a computer that I do not have root access on).
One piece of code (a bash script) calls some python code like:
python somecode.py
The shell will run the correct version of python, but it won't find somecode.py.
What I've tried:
Fail #1: I tried to add both the directory which contains somecode.py and the full path to the file to the PATH; to no avail. [Errno 2] No such file or directory
Fail #2: I can make it work for one computer ONLY if I add the full path to the correct version of python in the top line:
#!/usr/local/cool/python/version/location
However this breaks it running on any other computer.
Fail #3: I can also make it work if I make the bash script say:
python /full/path/to/github/place/somecode.py
but again, this only works for ONE computer because the paths are different for different computers.
What I really want to do: I want to be able to use the same code (both bash script and somecode.py) on multiple computers.
Any suggestions about how to do this properly is welcome. Thanks!
Solution
Added:
#!/usr/bin/env python
To the top of my somecode.py code;
mv somecode.py somecode
chmod +x somecode
Make sure PATH has /full/path/to/directory/with/somecode.
Bash script now says only:
somecode
and it works.
For problem #2 try
#!/usr/bin/env python
though it may find different versions of Python on different machines, but as long as that's not a problem this should fix that particular problem
See this SO question Python deployment and /usr/bin/env portability too. And this post by Alex Martelli re use of this.
If you say python somefile.py then it will take the location of somefile.py as the current directory, not from $PATH. It will take the location of python from $PATH.
If you say somefile.py then it will take the location of somefile.py from $PATH, and the location of python from the #! line of your python script, which can use the PATH if you follow #Levon's suggestion.