I try to write a programm in python that notifies me, when a shell like cmd gets opened.
Until now I did the following in python.
Check for new starting processes, get the name of the process and check if its name is cmd.exe.
This works if I start a cmd process manually myself.
But it Turns out if i open a shell with subprocess.getoutput(command) from the subprocess library in python there is no shell listed in the prosesses and I also cant see it in taskmanager.
So I assumed its a childprocess of the pythonscripts process running?
My next Idea was to list all the modules a process is using and check for cmd.exe in the modules.
It turns out the pythonscript with subprocess.getoutput(command) does not use cmd.exe in the modules. Strange.
So right now I am not sure how I could detect the shell or if I am even on the right way.
Maybe I need to find the childprocesses of a the pythonprocess? Or is it possible to get a shell without calling cmd.exe I honestly dont know enough about it.
Maybe its better to check for chertain dlls in the used methods by a process?
I also tried to look in the subprocess.py library but it is difficult for me to understand and it seems to atleast pass over cmd as a parameter for subprocess.getoutput() method.
Can somebody help?
Thank you.
UPDATE:
I use this code to detect the process:
import wmi
c = wmi.WMI()
process_watcher = c.Win32_Process.watch_for("creation")
while True:
new_process = process_watcher()
print(new_process.Caption, new_process.ProcessId)
if new_process.Caption =="cmd.exe":
pid = new_process.ProcessID
break
But if I run this code
import subprocess
output = subprocess.getoutput("ipconfig")
print(output)
The only process detected is pythonw.exe
But if I run
import subprocess
while True:
output = subprocess.getoutput("ipconfig")
print(output)
At some point it find cmd.exe.
So I assume that wmi takes to long to detect the process. So cmd is already closed and does not get found.
Any Ideas how to do this a better way?
I didnt know practic version of solution.But you can use pyautogui for it if you want.You can write a program with pyautogui that notifies you when it find cmd logo at task bar.Example:
import pyautogui
cmdlogo = pyautogui.locateOnScreen('get screenshot of cmd logo and write file name here example:'cmd.png'')
While True:
if cmdlogo:
print('write here what yo want to say when it finds cmd')
else:
pyautogui.sleep(5)
I have a number of different codes which all take a text file with data as input and write to a different one as output. Three of these codes are written in Python 2.7, but one is written in IDL. My goal is to create one "master" python program which can run all of these codes, by typing "python master.py". However, because of the limitations of my system, I am unable to use the 'pyIDL' or 'pyIDLy' modules referred to in this question. Not sure if it matters, but this is using a linux command prompt.
Currently my 'master.py' code looks like this:
import os
os.system("python pycode_1.py")
os.system("idl")
os.system(".com idlcode.pro")
os.system(".r idlcode,"imputfile.dat"")
os.system("exit")
os.system("python pycode_2.py")
os.system("python pycode_3.py")
This code runs the first python code and enters IDL fine. However, it does not enter the later comands into IDL. This means that the IDL command prompt comes up, but I cannot run the IDL code that follows.
I would be very appreciative about any advice to solve this issue.
Thanks in advance!
If you have IDL 8.5 or later, it ships with the IDL-Python bridge built in. Then your code would look something like:
from idlpy import *
IDL.idlcode()
Hope this helps.
So I have worked out a solution that seems to work well for this problem. The issue above was the use of the os.system function to do things it couldn't. My new codde is:
import os
import subprocess
os.system("python python_code1.py")
p=subprocess.Popen("idl", stdin=subprocess.PIPE, stdout=subprocess.PIPE)
p.stdin.write(".com idlcode.pro\n")
p.stdin.write("idlcode\n")
p.stdin.write("exit")
p.wait()
os.system("python python_code2.py")
os.system("python python_code3.py")
I've been looking at multiple examples on here and elsewhere, but nothing seems for work for me. I know nothing about python. All I am trying to do is run a perl script simply located at
sdb1/media/process.pl
The example code that I've found runs all over the place, and mostly seems like it has extra stuff that I don't need. What I'm trying right now is
#! /usr/bin/python
pipe = subprocess.Popen(["perl", "/sdb1/media/process.pl"], stdout=subprocess.PIPE)
But that just gives me the error
NameError: name 'subprocess' is not defined
If I've missed anything important, let me know. Otherwise, thanks for your time.
you need to import the subprocess library to run subprocess
#! /usr/bin/python
import subprocess
pipe = subprocess.Popen(["perl", "/sdb1/media/process.pl"], stdout=subprocess.PIPE)
Alternatively, if you are just using that same function a lot of times, you can do
from subprocess import Popen
then you can just call
pipe = Popen(["perl", "/sdb1/media/process.pl"], stdout=subprocess.PIPE)
I would have commented but I need 50 rep.
I've got a bash file that I normally execute using Cygwin.
I need to run this file from my Python code.
I tried this:
for bashfile in files:
p = Popen(bashfile, cwd=dname) #dname is the current directory of the script
stdout, stderr = p.communicate()
I've also seen a similar question here, but when trying to run it that way it says that it can't find the directory of my bash file...
Any ideas? Thanks! :-)
Edit: bashfile has a full path.
Do you need its output to get it directly to Python? If not this may be very fast and easy solution:
os.system("""here some code you use to execute in Terminal""")
You can also try this, though it does (and will no matter what you try) matter where the directory is. This, as far as the output goes, may be a little bit cleaner than the os method.
import commands
cmd="bash ./script.sh"
commands.getoutput(cmd)
If the case is that you need to change the directory:
cmd = "/path/to/your/script/script.sh"
The added benefit of using this method, versus say, os is that you can assign the output to a variable...
fun_times = commands.getoutput("bash ./script.sh")
whereas...
not_fun_times = os.system("./script.sh")
will throw an error.
etc, etc.
I just started working on Python, and I have been trying to run an outside executable from Python.
I have an executable for a program written in Fortran. Let’s say the name for the executable is flow.exe. And my executable is located in C:\Documents and Settings\flow_model. I tried both os.system and popen commands, but so far I couldn't make it work. The following code seems like it opens the command window, but it wouldn't execute the model.
# Import system modules
import sys, string, os, arcgisscripting
os.system("C:/Documents and Settings/flow_model/flow.exe")
How can I fix this?
If using Python 2.7 or higher (especially prior to Python 3.5) you can use the following:
import subprocess
subprocess.call(args, *, stdin=None, stdout=None, stderr=None, shell=False)
Runs the command described by args. Waits for command to complete, then returns the returncode attribute.
subprocess.check_call(args, *, stdin=None, stdout=None, stderr=None, shell=False)
Runs command with arguments. Waits for command to complete. If the return code was zero then returns, otherwise raises CalledProcessError. The CalledProcessError object will have the return code in the returncode attribute
Example: subprocess.check_call([r"C:\pathToYourProgram\yourProgram.exe", "your", "arguments", "comma", "separated"])
In regular Python strings, the \U character combination signals a
extended Unicode code point escape.
Here is the link to the documentation: http://docs.python.org/3.2/library/subprocess.html
For Python 3.5+ you can now use run() in many cases: https://docs.python.org/3.5/library/subprocess.html#subprocess.run
Those whitespaces can really be a bother. Try os.chdir('C:/Documents\ and\ Settings/') followed by relative paths for os.system, subprocess methods, or whatever...
If best-effort attempts to bypass the whitespaces-in-path hurdle keep failing, then my next best suggestion is to avoid having blanks in your crucial paths. Couldn't you make a blanks-less directory, copy the crucial .exe file there, and try that? Are those havoc-wrecking space absolutely essential to your well-being...?
The simplest way is:
import os
os.startfile("C:\Documents and Settings\flow_model\flow.exe")
It works; I tried it.
I'd try inserting an 'r' in front of your path if I were you, to indicate that it's a raw string - and then you won't have to use forward slashes. For example:
os.system(r"C:\Documents and Settings\flow_model\flow.exe")
Your usage is correct. I bet that your external program, flow.exe, needs to be executed in its directory, because it accesses some external files stored there.
So you might try:
import sys, string, os, arcgisscripting
os.chdir('c:\\documents and settings\\flow_model')
os.system('"C:\\Documents and Settings\\flow_model\\flow.exe"')
(Beware of the double quotes inside the single quotes...)
Use subprocess, it is a smaller module so it runs the .exe quicker.
import subprocess
subprocess.Popen([r"U:\Year 8\kerbal space program\KSP.exe"])
By using os.system:
import os
os.system(r'"C:/Documents and Settings/flow_model/flow.exe"')
Try
import subprocess
subprocess.call(["C:/Documents and Settings/flow_model/flow.exe"])
If it were me, I'd put the EXE file in the root directory (C:) and see if it works like that. If so, it's probably the (already mentioned) spaces in the directory name. If not, it may be some environment variables.
Also, try to check you stderr (using an earlier answer by int3):
import subprocess
process = subprocess.Popen(["C:/Documents and Settings/flow_model/flow.exe"], \
stderr = subprocess.PIPE)
if process.stderr:
print process.stderr.readlines()
The code might not be entirely correct as I usually don't use Popen or Windows, but should give the idea. It might well be that the error message is on the error stream.
in python 2.6 use string enclosed inside quotation " and apostrophe ' marks. Also a change single / to double //.
Your working example will look like this:
import os
os.system("'C://Documents and Settings//flow_model//flow.exe'")
Also You can use any parameters if Your program ingest them.
os.system('C://"Program Files (x86)"//Maxima-gcl-5.37.3//gnuplot//bin//gnuplot -e "plot [-10:10] sin(x),atan(x),cos(atan(x)); pause mouse"')
finally You can use string variable, as an example is plotting using gnuplot directly from python:
this_program='C://"Program Files (x86)"//Maxima-gcl-5.37.3//gnuplot//bin//gnuplot'
this_par='-e "set polar; plot [-2*pi:2*pi] [-3:3] [-3:3] t*sin(t); pause -1"'
os.system(this_program+" "+this_par)
import os
path = "C:/Documents and Settings/flow_model/"
os.chdir(path)
os.system("flow.exe")
Note added by barlop
A commenter asked why this works. Here is why.
The OP's problem is os.system("...") doesn't work properly when there is a space in the path. (Note os.system can work with ('"...."') but anyhow)
Had the OP tried their program from a cmd prompt they'd have seen the error clearly.
C:\carp>type blah.py
import os
os.system(R"C:\Program Files (x86)\Google\Chrome\Application\chrome.exe")
C:\carp>python blah.py
'C:\Program' is not recognized as an internal or external command,
operable program or batch file.
C:\carp>
So it's fine for os.system("calc.exe") (there calc.exe is in the path environment variable). Or for os.system(R"c:\windows\system32\calc.exe"). There's no space in that path.
C:\>md "aa bb cc"
C:\>copy c:\windows\system32\calc.exe "c:\aa bb cc\cccalc.exe"
1 file(s) copied.
This works (Given file "c:\aa bb cc\cccalc.exe" )
import os
os.chdir(R"c:\aa bb cc")
os.system("cccalc.exe")
Other options are subprocess.run and subprocess.popen.
Is that trying to execute C:\Documents with arguments of "and", "Settings/flow_model/flow.exe"?
Also, you might consider subprocess.call().
There are loads of different solutions, and the results will strongly depend on:
the OS you are using: Windows, Cygwin, Linux, MacOS
the python version you are using: Python2 or Python3x
As I have discovered some things that are claimed to work only in Windows, doesn't, probably because I happen to use Cygwin which is outsmarting the OS way to deal with Windows paths. Other things only work in pure *nix based OS's or in Python2 or 3.
Here are my findings:
Generally speaking, os.system() is the most forgiving method.
os.startfile() is the least forgiving. (Windows only && if you're lucky)
subprocess.Popen([...]) not recommended
subprocess.run(winView, shell=True) the recommended way!
Remembering that using subprocess for anything may pose a security risk.
Try these:
import os, subprocess
...
winView = '/cygdrive/c/Windows/explorer.exe %s' % somefile
...
# chose one of these:
os.system(winView)
subprocess.Popen(['/cygdrive/c/Windows/explorer.exe', 'somefile.png'])
subprocess.run(winView, shell=True)
Q: Why would you want to use explorer in Windows?
A: Because if you just want to look at the results of some new file, explorer will automatically open the file with whatever default windows program you have set for that file type. So no need to re-specify the default program to use.
That's the correct usage, but perhaps the spaces in the path name are messing things up for some reason.
You may want to run the program under cmd.exe as well so you can see any output from flow.exe that might be indicating an error.
for the above question this solution works.
just change the path to where your executable file is located.
import sys, string, os
os.chdir('C:\\Downloads\\xpdf-tools-win-4.00\\xpdf-tools-win-4.00\\bin64')
os.system("C:\\Downloads\\xpdf-tools-win-4.00\\xpdf-tools-win-4.00\bin64\\flowwork.exe")
'''import sys, string, os
os.chdir('C:\\Downloads\\xpdf-tools-win-4.00\\xpdf-tools-win-4.00\\bin64')
os.system(r"C:\\Downloads\\xpdf-tools-win-4.00\\xpdf-tools-win-4.00\bin64\\pdftopng.exe test1.pdf rootimage")'''
Here test1.pdf rootimage is for my code .