While in a Jupyter Notebook I am trying to save the result of the IPython %%prun magic function to a specific file which lives in a subfolder of the current directory, say filename = r'subfoler\profile_result.txt'. I would like to be able to link this file name via
%%prun -T filename
...
However, when I do this, it just prints the profiler output to a file called filename in the current directory. This question has a similar problem, for which the solution is to pass in the variable via $filename. This does not work, and instead just saves the output to a file called $filename. Is this a bug, is it not possible to do what I want, or am I incorrectly passing in the python variable?
%prun docs has this note:
.. versionchanged:: 7.3
User variables are no longer expanded,
the magic line is always left unmodified.
I'm not sure where the expansion is normally done, but prun now uses
opts, arg_str = self.parse_options(parameter_s, 'D:l:rs:T:q',
list_all=True, posix=False)
to handle the arguments.
%history on the other hand uses
#magic_arguments
#argument...
args = parse_argstring(self.history, parameter_s)
Without getting into details, that's a different parsing route.
I have version IPython 7.16.1
%debug, %time and %timeit have the same note.
The relevant pull-request: https://github.com/ipython/ipython/pull/11516/
The reasoning: "Most useful for magics that take Python code on the line, e.g. %timeit, etc."
Is it possible to save an IPython workspace (defined functions, different kinds of variables, etc) so that it can be loaded later?
This would be a similar function to save.image() in MATLAB or R. Similar questions has been asked before, such as:
Save session in IPython like in MATLAB?
However, since a few years passed, I am wondering if there is a good solution now.
You can use the dill python package:
import dill
filepath = 'session.pkl'
dill.dump_session(filepath) # Save the session
dill.load_session(filepath) # Load the session
To install it:
pip install dill
EDIT: this answer (and gist) has been modified to work for IPython 6
I added a somewhat ad-hoc solution that automates the process of storing/restoring user space variables using the underlying code from IPython's %store magic which is from what I understand what you wanted. See the gist here. Note that this only works for objects that can be pickled.
I can't guarantee its robustness, especially if any of the autorestore mechanisms in IPython change in the future, but it has been working for me with IPython 2.1.0. Hopefully this will at least point you in the right direction.
To reiterate the solution here:
Add the save_user_variables.py script below to your ipython folder (by default $HOME/.ipython). This script takes care of saving user variables on exit.
Add this line to your profile's ipython startup script (e.g., $HOME/.ipython/profile_default/startup/startup.py):
get_ipython().ex("import save_user_variables;del save_user_variables")
In your ipython profile config file (by default $HOME/.ipython/profile_default/ipython_config.py) find the following line:
# c.StoreMagics.autorestore = False
Uncomment it and set it to true. This automatically reloads stored variables on startup. Alternatively you can reload the last session manually using %store -r.
save_user_variables.py
def get_response(quest,default=None,opts=('y','n'),please=None,fmt=None):
try:
raw_input = input
except NameError:
pass
quest += " ("
quest += "/".join(['['+o+']' if o==default else o for o in opts])
quest += "): "
if default is not None: opts = list(opts)+['']
if please is None: please = quest
if fmt is None: fmt = lambda x: x
rin = input(quest)
while fmt(rin) not in opts: rin = input(please)
return default if default is not None and rin == '' else fmt(rin)
def get_user_vars():
"""
Get variables in user namespace (ripped directly from ipython namespace
magic code)
"""
import IPython
ip = IPython.get_ipython()
user_ns = ip.user_ns
user_ns_hidden = ip.user_ns_hidden
nonmatching = object()
var_hist = [ i for i in user_ns
if not i.startswith('_') \
and (user_ns[i] is not user_ns_hidden.get(i, nonmatching)) ]
return var_hist
def shutdown_logger():
"""
Prompts for saving the current session during shutdown
"""
import IPython, pickle
var_hist = get_user_vars()
ip = IPython.get_ipython()
db = ip.db
# collect any variables that need to be deleted from db
keys = map(lambda x: x.split('/')[1], db.keys('autorestore/*'))
todel = set(keys).difference(ip.user_ns)
changed = [db[k] != ip.user_ns[k.split('/')[1]]
for k in db.keys('autorestore/*') if k.split('/')[1] in ip.user_ns]
try:
if len(var_hist) == 0 and len(todel) == 0 and not any(changed): return
if get_response("Save session?", 'n', fmt=str.lower) == 'n': return
except KeyboardInterrupt:
return
# Save interactive variables (ignore unsaveable ones)
for name in var_hist:
obj = ip.user_ns[name]
try:
db[ 'autorestore/' + name ] = obj
except pickle.PicklingError:
print("Could not store variable '%s'. Skipping..." % name)
del db[ 'autorestore/' + name ]
# Remove any previously stored variables that were deleted in this session
for k in todel:
del db['autorestore/'+k]
import atexit
atexit.register(shutdown_logger)
del atexit
Although not so convenient as save.image(), you can use one of the checkpoint/restore applications. If you're using Linux, you might try http://criu.org. I'm using it from time to time to dump my ipython state and restore it later.
In order to dump a shell app with CRIU, you need to find its PID (e.g. pstree -p) and then use something like this (you'll need a second terminal for this; CRIU can't dump stopped jobs):
sudo criu dump -t PID --images-dir ~/tmp/imgs --log-file dump.log -v4 --shell-job
this will write all necessary images to ~/tmp/imgs (remember the --shell-job option). In order to restore the state later to the current terminal (don't forget to hit enter to get the next ipython prompt):
sudo criu restore --images-dir ~/tmp/imgs/ --log-file restore.log -v4 --shell-job
Check out the logs in case of any problems.
Obviously CRIU will work with any app (with some limits, of course). It's just an idea so you can use it for ipython.
You can try
%save name lines
Like if you have input 67 commands and you want to save all of them:
%save myhistory 1-67
you can certainly do this in the ipython notebook.
when the notebook is saved--either manually or by default config--the notebook is persisted as an .ipynb file, which is just a json file (an example in a github gist).
Next time you start the ipython server in the directory where that file resides, the server will detect it.
when you open that notebook in the browser, all of the code & config is there but unexecuted; you can execute the code in every cell by selecting execute all cells from the cells menu.
in addition, you can manually persist snapshots of your notebook, as ipynb_checkpoints, which are stored in a directory of that name preceded by a dot.
and finally, from the file menu option, you can persist your notebook as a pure python source file (.py)
Is there a way in Python to detect, within a process, where that process is being executed? I have some code that includes the getpass.getpass() function, which is broken in Spyder, and it's annoying to go back and forth between the command line and the IDE all the time. It would be useful if I could add code like:
if not being run from Spyder:
use getpass
else:
use alternative
Here is the solution I ended up using. After reading Markus's answer, I noticed that Spyder adds half a dozen or so environment variables to os.environ with names like SPYDER_ENCODING, SPYDER_SHELL_ID, etc. Detecting the presence of any of these seems relatively unambiguous, compared to detecting the absence of a variable with as generic a name as 'PYTHONSTARTUP'. The code is simple, and works independently of Spyder's startup script (as far as I can tell):
if any('SPYDER' in name for name in os.environ)
# use alternative
else:
# use getpass
Since the string is at the beginning of each environment variable name, you could also use str.startswith, but it's less flexible, and a little bit slower (I was curious):
>>> import timeit
>>> s = timeit.Timer("[name.startswith('SPYDER') for name in os.environ]", "import os")
>>> i = timeit.Timer("['SPYDER' in name for name in os.environ]", "import os")
>>> s.timeit()
16.18333065883474
>>> i.timeit()
6.156869294143846
The sys.executable method may or may not be useful depending on your installation. I have a couple WinPython installations and a separate Python 2.7 installation, so I was able to check the condition sys.executable.find('WinPy') == -1 to detect a folder name in the path of the executable Spyder uses. Since the warning that shows in IDLE when you try to use getpass is less "loud" than it could be, in my opinion, I ended up also checking the condition sys.executable.find('pythonw.exe') == -1 to make it slightly louder. Using sys.executable only, that method looks like:
if sys.executable.find('pythonw.exe') == sys.executable.find('WinPy') == -1:
# use getpass
else:
# use alternative
But since I want this to work on other machines, and it's much more likely that another user would modify their WinPython installation folder name than that they would rename their IDLE executable, my final code uses sys.executable to detect IDLE and os.environ to detect Spyder, providing a "louder" warning in either case and keeping the code from breaking in the latter.
if any('SPYDER' in name for name in os.environ) \
or 'pythonw.exe' in sys.executable:
password = raw_input('WARNING: PASSWORD WILL BE SHOWN ON SCREEN\n\n' * 3
+ 'Please enter your password: ')
else:
password = getpass.getpass("Please enter your password: ")
By default, Spyder uses a startup scrip, see Preferences -> Console -> Adanced setting. This option is usually set to the scientific_startup.py file that loads pylab et al.
The easiest solution is to just add a global variable to the file and then use that in your if statement, e.g. add this line at the end of scientific_startup.py:
SPYDER_IDE_ACTIVE = True
In your script:
if not 'SPYDER_IDE_ACTIVE' in globals():
use getpass
else:
use alternative
This will work without throwing an error. You can also use exceptions if you like that more.
A second solution would be (if you cannot modify that file for some reason) to just check if the environment variable PYTHONSTARTUP is set. On my machine (using the Anaconda Python stack), it is not set for a regular Python shell. You could do
import os
if not 'PYTHONSTARTUP' in os.environ:
use getpass
else:
use alternative
Spyder provides the option of executing the current editor script in a native system terminal. This would produce identical behavior as if you were running from the command line. To set this up, open the Run Settings dialog by hitting F6. Then select the radio button "Execute in an external System terminal". Now run the script as usual by hitting F5. You should be able to use getpass in the normal fashion with this approach.
You could add env variable when running in Spyder and check it in code.
I have a .pythonrc in my path, which gets loaded when I run python:
python
Loading pythonrc
>>>
The problem is that my .pythonrc is not loaded when I execute files:
python -i script.py
>>>
It would be very handy to have tab completion (and a few other things) when I load things interactively.
From the Python documentation for -i:
When a script is passed as first argument or the -c option is used, enter interactive mode after executing the script or the command, even when sys.stdin does not appear to be a terminal. The PYTHONSTARTUP file is not read.
I believe this is done so that scripts run predictably for all users, and do not depend on anything in a user's particular PYTHONSTARTUP file.
As Greg has noted, there is a very good reason why -i behaves the way it does. However, I do find it pretty useful to be able to have my PYTHONSTARTUP loaded when I want an interactive session. So, here's the code I use when I want to be able to have PYTHONSTARTUP active in a script run with -i.
if __name__ == '__main__':
#do normal stuff
#and at the end of the file:
import sys
if sys.flags.interactive==1:
import os
myPythonPath = os.environ['PYTHONSTARTUP'].split(os.sep)
sys.path.append(os.sep.join(myPythonPath[:-1]))
pythonrcName = ''.join(myPythonPath[-1].split('.')[:-1]) #the filename minus the trailing extension, if the extension exists
pythonrc = __import__(pythonrcName)
for attr in dir(pythonrc):
__builtins__.__dict__[attr] = getattr(pythonrc, attr)
sys.path.remove(os.sep.join(myPythonPath[:-1]))
del sys, os, pythonrc
Note that this is fairly hacky and I never do this without ensuring that my pythonrc isn't accidentally clobbering variables and builtins.
Apparently the user module provides this, but has been removed in Python 3.0. It is a bit of a security hole, depending what's in your pythonrc...
In addition to Chinmay Kanchi and Greg Hewgill's answers, I'd like to add that IPython and BPython work fine in this case. Perhaps it's time for you to switch? :)
I find myself frequently using Python's interpreter to work with databases, files, etc -- basically a lot of manual formatting of semi-structured data. I don't properly save and clean up the useful bits as often as I would like. Is there a way to save my input into the shell (db connections, variable assignments, little for loops and bits of logic) -- some history of the interactive session? If I use something like script I get too much stdout noise. I don't really need to pickle all the objects -- though if there is a solution that does that, it would be OK. Ideally I would just be left with a script that ran as the one I created interactively, and I could just delete the bits I didn't need. Is there a package that does this, or a DIY approach?
IPython is extremely useful if you like using interactive sessions. For example for your use-case there is the %save magic command, you just input %save my_useful_session 10-20 23 to save input lines 10 to 20 and 23 to my_useful_session.py (to help with this, every line is prefixed by its number).
Furthermore, the documentation states:
This function uses the same syntax as %history for input ranges, then saves the lines to the filename you specify.
This allows for example, to reference older sessions, such as
%save current_session ~0/
%save previous_session ~1/
Look at the videos on the presentation page to get a quick overview of the features.
From Andrew Jones's website (archived):
import readline
readline.write_history_file('/home/ahj/history')
There is a way to do it. Store the file in ~/.pystartup...
# Add auto-completion and a stored history file of commands to your Python
# interactive interpreter. Requires Python 2.0+, readline. Autocomplete is
# bound to the Esc key by default (you can change it - see readline docs).
#
# Store the file in ~/.pystartup, and set an environment variable to point
# to it: "export PYTHONSTARTUP=/home/user/.pystartup" in bash.
#
# Note that PYTHONSTARTUP does *not* expand "~", so you have to put in the
# full path to your home directory.
import atexit
import os
import readline
import rlcompleter
historyPath = os.path.expanduser("~/.pyhistory")
def save_history(historyPath=historyPath):
import readline
readline.write_history_file(historyPath)
if os.path.exists(historyPath):
readline.read_history_file(historyPath)
atexit.register(save_history)
del os, atexit, readline, rlcompleter, save_history, historyPath
and then set the environment variable PYTHONSTARTUP in your shell (e.g. in ~/.bashrc):
export PYTHONSTARTUP=$HOME/.pystartup
You can also add this to get autocomplete for free:
readline.parse_and_bind('tab: complete')
Please note that this will only work on *nix systems. As readline is only available in Unix platform.
If you are using IPython you can save to a file all your previous commands using the magic function %history with the -f parameter, p.e:
%history -f /tmp/history.py
After installing Ipython, and opening an Ipython session by running the command:
ipython
from your command line, just run the following Ipython 'magic' command to automatically log your entire Ipython session:
%logstart
This will create a uniquely named .py file and store your session for later use as an interactive Ipython session or for use in the script(s) of your choosing.
Also, reinteract gives you a notebook-like interface to a Python session.
In addition to IPython, a similar utility bpython has a "save the code you've entered to a file" feature
I had to struggle to find an answer, I was very new to iPython environment.
This will work
If your iPython session looks like this
In [1] : import numpy as np
....
In [135]: counter=collections.Counter(mapusercluster[3])
In [136]: counter
Out[136]: Counter({2: 700, 0: 351, 1: 233})
You want to save lines from 1 till 135 then on the same ipython session use this command
In [137]: %save test.py 1-135
This will save all your python statements in test.py file in your current directory ( where you initiated the ipython).
There is %history magic for printing and saving the input history (and optionally the output).
To store your current session to a file named my_history.py:
>>> %hist -f my_history.py
History IPython stores both the commands you enter, and the results it produces. You can easily go through previous commands with the up- and down-arrow keys, or access your history in more sophisticated ways.
You can use the %history magic function to examine past input and output. Input history from previous sessions is saved in a database, and IPython can be configured to save output history.
Several other magic functions can use your input history, including %edit, %rerun, %recall, %macro, %save and %pastebin. You can use a standard format to refer to lines:
%pastebin 3 18-20 ~1/1-5
This will take line 3 and lines 18 to 20 from the current session, and lines 1-5 from the previous session.
See %history? for the Docstring and more examples.
Also, be sure to explore the capabilities of %store magic for lightweight persistence of variables in IPython.
Stores variables, aliases and macros in IPython’s database.
d = {'a': 1, 'b': 2}
%store d # stores the variable
del d
%store -r d # Refresh the variable from IPython's database.
>>> d
{'a': 1, 'b': 2}
To autorestore stored variables on startup, specifyc.StoreMagic.autorestore = True in ipython_config.py.
The %history command is awesome, but unfortunately it won't let you save things that were %paste 'd into the sesh. To do that I think you have to do %logstart at the beginning (although I haven't confirmed this works).
What I like to do is
%history -o -n -p -f filename.txt
which will save the output, line numbers, and '>>>' before each input (o, n, and p options). See the docs for %history here.
Just putting another suggesting in the bowl:
Spyder
It has History log and Variable explorer. If you have worked with MatLab, then you'll see the similarities.
As far as Linux goes, one can use script command to record the whole session. It is part of util-linux package so should be on most Linux systems . You can create and alias or function that will call script -c python and that will be saved to a typescript file. For instance, here's a reprint of one such file.
$ cat typescript
Script started on Sat 14 May 2016 08:30:08 AM MDT
Python 2.7.6 (default, Jun 22 2015, 17:58:13)
[GCC 4.8.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> print 'Hello Pythonic World'
Hello Pythonic World
>>>
Script done on Sat 14 May 2016 08:30:42 AM MDT
Small disadvantage here is that the script records everything , even line-feeds, whenever you hit backspaces , etc. So you may want to use col to clean up the output (see this post on Unix&Linux Stackexchange) .
there is another option --- pyslice.
in the "wxpython 2.8 docs demos and tools", there is a open source program named "pyslices".
you can use it like a editor, and it also support using like a console ---- executing each line like a interactive interpreter with immediate echo.
of course, all the blocks of codes and results of each block will be recorded into a txt file automatically.
the results are logged just behind the corresponding block of code. very convenient.
In IPython, I first use
In [2]: %hist
to view my past code. I select the chunk I want to save and then paste it into file my_file.py using the %%file magic (short for %%writefile)
In [3]: %%file my_file.py
...: # paste code here
...:
...:
hitting return two times in the end.
To append to file use the option -a: %%file -a my_file.py.
If needed, I can list, edit, etc. the file in the underlying command line using the exclamation mark
In [5]: !ls -l my_file.py
In [6]: !vi my_file.py
Some comments were asking how to save all of the IPython inputs at once. For %save magic in IPython, you can save all of the commands programmatically as shown below, to avoid the prompt message and also to avoid specifying the input numbers.
currentLine = len(In)-1
%save -f my_session 1-$currentLine
The -f option is used for forcing file replacement and the len(IN)-1 shows the current input prompt in IPython, allowing you to save the whole session programmatically.
For those using spacemacs, and ipython that comes with python-layer, save magic creates a lot of unwanted output, because of the constant auto-completion command working in the backround such as:
len(all_suffixes)
';'.join(__PYTHON_EL_get_completions('''len'''))
';'.join(__PYTHON_EL_get_completions('''all_substa'''))
len(all_substantives_w_suffixes)
';'.join(__PYTHON_EL_get_completions('''len'''))
';'.join(__PYTHON_EL_get_completions('''all'''))
';'.join(__PYTHON_EL_get_completions('''all_'''))
';'.join(__PYTHON_EL_get_completions('''all_w'''))
';'.join(__PYTHON_EL_get_completions('''all_wo'''))
';'.join(__PYTHON_EL_get_completions('''all_wor'''))
';'.join(__PYTHON_EL_get_completions('''all_word'''))
';'.join(__PYTHON_EL_get_completions('''all_words'''))
len(all_words_w_logograms)
len(all_verbs)
To avoid this just save the ipython buffer like you normally save any other: spc f s
If you use bpython, all your command history is by default saved to ~/.pythonhist.
To save the commands for later reusage you can copy them to a python script file:
$ cp ~/.pythonhist mycommands.py
Then edit that file to clean it up and put it under Python path (global or virtual environment's site-packages, current directory, mentioning in *.pth, or some other way).
To include the commands into your shell, just import them from the saved file:
>>> from mycommands import *
I'd like to suggest another way to maintain python session through tmux on linux. you run tmux, attach your self to the session you opened (if not attached after opening it directly). execute python and do whatever you are doing on it. then detach from session. detaching from a tmux session does not close the session. the session remains open.
pros of this method:
you can attach to this session from any other device (in case you can ssh your pc)
cons of this method:
this method does not relinquish the resources used by the opened python session until you actually exist the python interpreter.
To save input and output on XUbuntu:
In XWindows, run iPython from the Xfce terminal app
click Terminal in the top menu bar and look for save contents in the dropdown
I find this saves the input and output, going all the way back to when I opened the terminal. This is not ipython specific, and would work with ssh sessions or other tasks run from the terminal window.
You can use built-in function open: I use it in all of my
programs in which I need to store some history (Including Calculator, etc.)
for example:
#gk-test.py or anything else would do
try: # use the try loop only if you haven't created the history file outside program
username = open("history.txt").readline().strip("\n")
user_age = open("history.txt").readlines()[1].strip("\n")
except FileNotFoundError:
username = input("Enter Username: ")
user_age = input("Enter User's Age: ")
open("history.txt", "w").write(f"{username}\n{user_age}")
#Rest of the code is secret! try it your own!
I would thank to all of them who liked my comments! Thank you for reading this!