Related
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
So, I started learning to code in Python and later Django. The first times it was hard looking at tracebacks and actually figure out what I did wrong and where the syntax error was. Some time has passed now and some way along the way, I guess I got a routine in debugging my Django code. As this was done early in my coding experience, I sat down and wondered if how I was doing this was ineffective and could be done faster. I usually manage to find and correct the bugs in my code, but I wonder if I should be doing it faster?
I usually just use the debug info Django gives when enabled. When things do end up as I thought it would, I break the code flow a lot with a syntax error, and look at the variables at that point in the flow to figure out, where the code does something other than what I wanted.
But can this be improved? Are there some good tools or better ways to debug your Django code?
There are a bunch of ways to do it, but the most straightforward is to simply
use the Python debugger. Just add following line in to a Django view function:
import pdb; pdb.set_trace()
or
breakpoint() #from Python3.7
If you try to load that page in your browser, the browser will hang and you get a prompt to carry on debugging on actual executing code.
However there are other options (I am not recommending them):
* return HttpResponse({variable to inspect})
* print {variable to inspect}
* raise Exception({variable to inspect})
But the Python Debugger (pdb) is highly recommended for all types of Python code. If you are already into pdb, you'd also want to have a look at IPDB that uses ipython for debugging.
Some more useful extension to pdb are
pdb++, suggested by Antash.
pudb, suggested by PatDuJour.
Using the Python debugger in Django, suggested by Seafangs.
I really like Werkzeug's interactive debugger. It's similar to Django's debug page, except that you get an interactive shell on every level of the traceback. If you use the django-extensions, you get a runserver_plus managment command which starts the development server and gives you Werkzeug's debugger on exceptions.
Of course, you should only run this locally, as it gives anyone with a browser the rights to execute arbitrary python code in the context of the server.
A little quickie for template tags:
#register.filter
def pdb(element):
import pdb; pdb.set_trace()
return element
Now, inside a template you can do {{ template_var|pdb }} and enter a pdb session (given you're running the local devel server) where you can inspect element to your heart's content.
It's a very nice way to see what's happened to your object when it arrives at the template.
There are a few tools that cooperate well and can make your debugging task easier.
Most important is the Django debug toolbar.
Then you need good logging using the Python logging facility. You can send logging output to a log file, but an easier option is sending log output to firepython. To use this you need to use the Firefox browser with the firebug extension. Firepython includes a firebug plugin that will display any server-side logging in a Firebug tab.
Firebug itself is also critical for debugging the Javascript side of any app you develop. (Assuming you have some JS code of course).
I also liked django-viewtools for debugging views interactively using pdb, but I don't use it that much.
There are more useful tools like dozer for tracking down memory leaks (there are also other good suggestions given in answers here on SO for memory tracking).
I use PyCharm (same pydev engine as eclipse). Really helps me to visually be able to step through my code and see what is happening.
Almost everything has been mentioned so far, so I'll only add that instead of pdb.set_trace() one can use ipdb.set_trace() which uses iPython and therefore is more powerful (autocomplete and other goodies). This requires ipdb package, so you only need to pip install ipdb
I've pushed django-pdb to PyPI.
It's a simple app that means you don't need to edit your source code every time you want to break into pdb.
Installation is just...
pip install django-pdb
Add 'django_pdb' to your INSTALLED_APPS
You can now run: manage.py runserver --pdb to break into pdb at the start of every view...
bash: manage.py runserver --pdb
Validating models...
0 errors found
Django version 1.3, using settings 'testproject.settings'
Development server is running at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
GET /
function "myview" in testapp/views.py:6
args: ()
kwargs: {}
> /Users/tom/github/django-pdb/testproject/testapp/views.py(7)myview()
-> a = 1
(Pdb)
And run: manage.py test --pdb to break into pdb on test failures/errors...
bash: manage.py test testapp --pdb
Creating test database for alias 'default'...
E
======================================================================
>>> test_error (testapp.tests.SimpleTest)
----------------------------------------------------------------------
Traceback (most recent call last):
File ".../django-pdb/testproject/testapp/tests.py", line 16, in test_error
one_plus_one = four
NameError: global name 'four' is not defined
======================================================================
> /Users/tom/github/django-pdb/testproject/testapp/tests.py(16)test_error()
-> one_plus_one = four
(Pdb)
The project's hosted on GitHub, contributions are welcome of course.
The easiest way to debug python - especially for programmers that are used to Visual Studio - is using PTVS (Python Tools for Visual Studio).
The steps are simple:
Download and install it from https://microsoft.github.io/PTVS/
Set breakpoints and press F5.
Your breakpoint is hit, you can view/change the variables as easy as debugging C#/C++ programs.
That's all :)
If you want to debug Django using PTVS, you need to do the following:
In Project settings - General tab, set "Startup File" to "manage.py", the entry point of the Django program.
In Project settings - Debug tab, set "Script Arguments" to "runserver --noreload". The key point is the "--noreload" here. If you don't set it, your breakpoints won't be hit.
Enjoy it.
I use pyDev with Eclipse really good, set break points, step into code, view values on any objects and variables, try it.
I use PyCharm and stand by it all the way. It cost me a little but I have to say the advantage that I get out of it is priceless. I tried debugging from console and I do give people a lot of credit who can do that, but for me being able to visually debug my application(s) is great.
I have to say though, PyCharm does take a lot of memory. But then again, nothing good is free in life. They just came with their latest version 3. It also plays very well with Django, Flask and Google AppEngine. So, all in all, I'd say it's a great handy tool to have for any developer.
If you are not using it yet, I'd recommend to get the trial version for 30 days to take a look at the power of PyCharm. I'm sure there are other tools also available, such as Aptana. But I guess I just also like the way PyCharm looks. I feel very comfortable debugging my apps there.
From my perspective, we could break down common code debugging tasks into three distinct usage patterns:
Something has raised an exception: runserver_plus' Werkzeug debugger to the rescue. The ability to run custom code at all the trace levels is a killer. And if you're completely stuck, you can create a Gist to share with just a click.
Page is rendered, but the result is wrong: again, Werkzeug rocks. To make a breakpoint in code, just type assert False in the place you want to stop at.
Code works wrong, but the quick look doesn't help. Most probably, an algorithmic problem. Sigh. Then I usually fire up a console debugger PuDB: import pudb; pudb.set_trace(). The main advantage over [i]pdb is that PuDB (while looking as you're in 80's) makes setting custom watch expressions a breeze. And debugging a bunch of nested loops is much simpler with a GUI.
Ah, yes, the templates' woes. The most common (to me and my colleagues) problem is a wrong context: either you don't have a variable, or your variable doesn't have some attribute. If you're using debug toolbar, just inspect the context at the "Templates" section, or, if it's not sufficient, set a break in your views' code just after your context is filled up.
So it goes.
Add import pdb; pdb.set_trace() or breakpoint() (form python3.7) at the corresponding line in the Python code and execute it. The execution will stop with an interactive shell. In the shell you can execute Python code (i.e. print variables) or use commands such as:
c continue execution
n step to the next line within the same function
s step to the next line in this function or a called function
q quit the debugger/execution
Also see: https://poweruser.blog/setting-a-breakpoint-in-python-438e23fe6b28
Sometimes when I wan to explore around in a particular method and summoning pdb is just too cumbersome, I would add:
import IPython; IPython.embed()
IPython.embed() starts an IPython shell which have access to the local variables from the point where you call it.
I just found wdb (http://www.rkblog.rk.edu.pl/w/p/debugging-python-code-browser-wdb-debugger/?goback=%2Egde_25827_member_255996401). It has a pretty nice user interface / GUI with all the bells and whistles. Author says this about wdb -
"There are IDEs like PyCharm that have their own debuggers. They offer similar or equal set of features ... However to use them you have to use those specific IDEs (and some of then are non-free or may not be available for all platforms). Pick the right tool for your needs."
Thought i'd just pass it on.
Also a very helpful article about python debuggers:
https://zapier.com/engineering/debugging-python-boss/
Finally, if you'd like to see a nice graphical printout of your call stack in Django, checkout:
https://github.com/joerick/pyinstrument. Just add pyinstrument.middleware.ProfilerMiddleware to MIDDLEWARE_CLASSES, then add ?profile to the end of the request URL to activate the profiler.
Can also run pyinstrument from command line or by importing as a module.
I highly recommend epdb (Extended Python Debugger).
https://bitbucket.org/dugan/epdb
One thing I love about epdb for debugging Django or other Python webservers is the epdb.serve() command. This sets a trace and serves this on a local port that you can connect to. Typical use case:
I have a view that I want to go through step-by-step. I'll insert the following at the point I want to set the trace.
import epdb; epdb.serve()
Once this code gets executed, I open a Python interpreter and connect to the serving instance. I can analyze all the values and step through the code using the standard pdb commands like n, s, etc.
In [2]: import epdb; epdb.connect()
(Epdb) request
<WSGIRequest
path:/foo,
GET:<QueryDict: {}>,
POST:<QuestDict: {}>,
...
>
(Epdb) request.session.session_key
'i31kq7lljj3up5v7hbw9cff0rga2vlq5'
(Epdb) list
85 raise some_error.CustomError()
86
87 # Example login view
88 def login(request, username, password):
89 import epdb; epdb.serve()
90 -> return my_login_method(username, password)
91
92 # Example view to show session key
93 def get_session_key(request):
94 return request.session.session_key
95
And tons more that you can learn about typing epdb help at any time.
If you want to serve or connect to multiple epdb instances at the same time, you can specify the port to listen on (default is 8080). I.e.
import epdb; epdb.serve(4242)
>> import epdb; epdb.connect(host='192.168.3.2', port=4242)
host defaults to 'localhost' if not specified. I threw it in here to demonstrate how you can use this to debug something other than a local instance, like a development server on your local LAN. Obviously, if you do this be careful that the set trace never makes it onto your production server!
As a quick note, you can still do the same thing as the accepted answer with epdb (import epdb; epdb.set_trace()) but I wanted to highlight the serve functionality since I've found it so useful.
One of your best option to debug Django code is via wdb:
https://github.com/Kozea/wdb
wdb works with python 2 (2.6, 2.7), python 3 (3.2, 3.3, 3.4, 3.5) and pypy. Even better, it is possible to debug a python 2 program with a wdb server running on python 3 and vice-versa or debug a program running on a computer with a debugging server running on another computer inside a web page on a third computer!
Even betterer, it is now possible to pause a currently running python process/thread using code injection from the web interface. (This requires gdb and ptrace enabled)
In other words it's a very enhanced version of pdb directly in your browser with nice features.
Install and run the server, and in your code add:
import wdb
wdb.set_trace()
According to the author, main differences with respect to pdb are:
For those who don’t know the project, wdb is a python debugger like pdb, but with a slick web front-end and a lot of additional features, such as:
Source syntax highlighting
Visual breakpoints
Interactive code completion using jedi
Persistent breakpoints
Deep objects inspection using mouse Multithreading / Multiprocessing support
Remote debugging
Watch expressions
In debugger code edition
Popular web servers integration to break on error
In exception breaking during trace (not post-mortem) in contrary to the werkzeug debugger for instance
Breaking in currently running programs through code injection (on supported systems)
It has a great browser-based user interface. A joy to use! :)
I use PyCharm and different debug tools. Also have a nice articles set about easy set up those things for novices. You may start here. It tells about PDB and GUI debugging in general with Django projects. Hope someone would benefit from them.
If using Aptana for django development, watch this: http://www.youtube.com/watch?v=qQh-UQFltJQ
If not, consider using it.
Most options are alredy mentioned.
To print template context, I've created a simple library for that.
See https://github.com/edoburu/django-debugtools
You can use it to print template context without any {% load %} construct:
{% print var %} prints variable
{% print %} prints all
It uses a customized pprint format to display the variables in a <pre> tag.
I find Visual Studio Code is awesome for debugging Django apps. The standard python launch.json parameters run python manage.py with the debugger attached, so you can set breakpoints and step through your code as you like.
For those that can accidentally add pdb into live commits, I can suggest this extension of #Koobz answer:
#register.filter
def pdb(element):
from django.conf import settings
if settings.DEBUG:
import pdb
pdb.set_trace()
return element
From my own experience , there are two way:
use ipdb,which is a enhanced debugger likes pdb.
import ipdb;ipdb.set_trace() or breakpoint() (from python3.7)
use django shell ,just use the command below. This is very helpfull when you are developing a new view.
python manage.py shell
i highly suggest to use PDB.
import pdb
pdb.set_trace()
You can inspect all the variables values, step in to the function and much more.
https://docs.python.org/2/library/pdb.html
for checking out the all kind of request,response and hits to database.i am using django-debug-toolbar
https://github.com/django-debug-toolbar/django-debug-toolbar
As mentioned in other posts here - setting breakpoints in your code and walking thru the code to see if it behaves as you expected is a great way to learn something like Django until you have a good sense of how it all behaves - and what your code is doing.
To do this I would recommend using WingIde. Just like other mentioned IDEs nice and easy to use, nice layout and also easy to set breakpoints evaluate / modify the stack etc. Perfect for visualizing what your code is doing as you step through it. I'm a big fan of it.
Also I use PyCharm - it has excellent static code analysis and can help sometimes spot problems before you realize they are there.
As mentioned already django-debug-toolbar is essential - https://github.com/django-debug-toolbar/django-debug-toolbar
And while not explicitly a debug or analysis tool - one of my favorites is SQL Printing Middleware available from Django Snippets at https://djangosnippets.org/snippets/290/
This will display the SQL queries that your view has generated. This will give you a good sense of what the ORM is doing and if your queries are efficient or you need to rework your code (or add caching).
I find it invaluable for keeping an eye on query performance while developing and debugging my application.
Just one other tip - I modified it slightly for my own use to only show the summary and not the SQL statement.... So I always use it while developing and testing. I also added that if the len(connection.queries) is greater than a pre-defined threshold it displays an extra warning.
Then if I spot something bad (from a performance or number of queries perspective) is happening I turn back on the full display of the SQL statements to see exactly what is going on. Very handy when you are working on a large Django project with multiple developers.
use pdb or ipdb. Diffrence between these two is ipdb supports auto complete.
for pdb
import pdb
pdb.set_trace()
for ipdb
import ipdb
ipdb.set_trace()
For executing new line hit n key, for continue hit c key.
check more options by using help(pdb)
An additional suggestion.
You can leverage nosetests and pdb together, rather injecting pdb.set_trace() in your views manually. The advantage is that you can observe error conditions when they first start, potentially in 3rd party code.
Here's an error for me today.
TypeError at /db/hcm91dmo/catalog/records/
render_option() argument after * must be a sequence, not int
....
Error during template rendering
In template /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/crispy_forms/templates/bootstrap3/field.html, error at line 28
render_option() argument after * must be a sequence, not int
18
19 {% if field|is_checkboxselectmultiple %}
20 {% include 'bootstrap3/layout/checkboxselectmultiple.html' %}
21 {% endif %}
22
23 {% if field|is_radioselect %}
24 {% include 'bootstrap3/layout/radioselect.html' %}
25 {% endif %}
26
27 {% if not field|is_checkboxselectmultiple and not field|is_radioselect %}
28
{% if field|is_checkbox and form_show_labels %}
Now, I know this means that I goofed the constructor for the form, and I even have good idea of which field is a problem. But, can I use pdb to see what crispy forms is complaining about, within a template?
Yes, I can. Using the --pdb option on nosetests:
tests$ nosetests test_urls_catalog.py --pdb
As soon as I hit any exception (including ones handled gracefully), pdb stops where it happens and I can look around.
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/forms.py", line 537, in __str__
return self.as_widget()
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/forms.py", line 593, in as_widget
return force_text(widget.render(name, self.value(), attrs=attrs))
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/widgets.py", line 513, in render
options = self.render_options(choices, [value])
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/widgets.py", line 543, in render_options
output.append(self.render_option(selected_choices, *option))
TypeError: render_option() argument after * must be a sequence, not int
INFO lib.capture_middleware log write_to_index(http://localhost:8082/db/hcm91dmo/catalog/records.html)
INFO lib.capture_middleware log write_to_index:end
> /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/widgets.py(543)render_options()
-> output.append(self.render_option(selected_choices, *option))
(Pdb) import pprint
(Pdb) pprint.PrettyPrinter(indent=4).pprint(self)
<django.forms.widgets.Select object at 0x115fe7d10>
(Pdb) pprint.PrettyPrinter(indent=4).pprint(vars(self))
{ 'attrs': { 'class': 'select form-control'},
'choices': [[('_', 'any type'), (7, (7, 'type 7', 'RECTYPE_TABLE'))]],
'is_required': False}
(Pdb)
Now, it's clear that my choices argument to the crispy field constructor was as it was a list within a list, rather than a list/tuple of tuples.
'choices': [[('_', 'any type'), (7, (7, 'type 7', 'RECTYPE_TABLE'))]]
The neat thing is that this pdb is taking place within crispy's code, not mine and I didn't need to insert it manually.
During development, adding a quick
assert False, value
can help diagnose problems in views or anywhere else, without the need to use a debugger.
For simple debugging in a complex project is there a reason to use the python logger instead of print? What about other use-cases? Is there an accepted best use-case for each (especially when you're only looking for stdout)?
I've always heard that this is a "best practice" but I haven't been able to figure out why.
The logging package has a lot of useful features:
Easy to see where and when (even what line no.) a logging call is being made from.
You can log to files, sockets, pretty much anything, all at the same time.
You can differentiate your logging based on severity.
Print doesn't have any of these.
Also, if your project is meant to be imported by other python tools, it's bad practice for your package to print things to stdout, since the user likely won't know where the print messages are coming from. With logging, users of your package can choose whether or not they want to propogate logging messages from your tool or not.
One of the biggest advantages of proper logging is that you can categorize messages and turn them on or off depending on what you need. For example, it might be useful to turn on debugging level messages for a certain part of the project, but tone it down for other parts, so as not to be taken over by information overload and to easily concentrate on the task for which you need logging.
Also, logs are configurable. You can easily filter them, send them to files, format them, add timestamps, and any other things you might need on a global basis. Print statements are not easily managed.
Print statements are sort of the worst of both worlds, combining the negative aspects of an online debugger with diagnostic instrumentation. You have to modify the program but you don't get more, useful code from it.
An online debugger allows you to inspect the state of a running program; But the nice thing about a real debugger is that you don't have to modify the source; neither before nor after the debugging session; You just load the program into the debugger, tell the debugger where you want to look, and you're all set.
Instrumenting the application might take some work up front, modifying the source code in some way, but the resulting diagnostic output can have enormous amounts of detail, and can be turned on or off to a very specific degree. The python logging module can show not just the message logged, but also the file and function that called it, a traceback if there was one, the actual time that the message was emitted, and so on. More than that; diagnostic instrumentation need never be removed; It's just as valid and useful when the program is finished and in production as it was the day it was added; but it can have it's output stuck in a log file where it's not likely to annoy anyone, or the log level can be turned down to keep all but the most urgent messages out.
anticipating the need or use for a debugger is really no harder than using ipython while you're testing, and becoming familiar with the commands it uses to control the built in pdb debugger.
When you find yourself thinking that a print statement might be easier than using pdb (as it often is), You'll find that using a logger pulls your program in a much easier to work on state than if you use and later remove print statements.
I have my editor configured to highlight print statements as syntax errors, and logging statements as comments, since that's about how I regard them.
In brief, the advantages of using logging libraries do outweigh print as below reasons:
Control what’s emitted
Define what types of information you want to include in your logs
Configure how it looks when it’s emitted
Most importantly, set the destination for your logs
In detail, segmenting log events by severity level is a good way to sift through which log messages may be most relevant at a given time. A log event’s severity level also gives you an indication of how worried you should be when you see a particular message. For instance, dividing logging type to debug, info, warning, critical, and error. Timing can be everything when you’re trying to understand what went wrong with an application. You want to know the answers to questions like:
“Was this happening before or after my database connection died?”
“Exactly when did that request come in?”
Furthermore, it is easy to see where a log has occurred through line number and filename or method name even in which thread.
Here's a functional logging library for Python named loguru.
If you use logging then the person responsible for deployment can configure the logger to send it to a custom location, with custom information. If you only print, then that's all they get.
Logging essentially creates a searchable plain text database of print outputs with other meta data (timestamp, loglevel, line number, process etc.).
This is pure gold, I can run egrep over the log file after the python script has run.
I can tune my egrep pattern search to pick exactly what I am interested in and ignore the rest. This reduction of cognitive load and freedom to pick my egrep pattern later on by trial and error is the key benefit for me.
tail -f mylogfile.log | egrep "key_word1|key_word2"
Now throw in other cool things that print can't do (sending to socket, setting debug levels, logrotate, adding meta data etc.), you have every reason to prefer logging over plain print statements.
I tend to use print statements because it's lazy and easy, adding logging needs some boiler plate code, hey we have yasnippets (emacs) and ultisnips (vim) and other templating tools, so why give up logging for plain print statements!?
I would add to all other mentionned advantages that the print function in standard configuration is buffered. The flush may occure only at the end of the current block (the one where the print is).
This is true for any program launched in a non interactive shell (codebuild, gitlab-ci for instance) or whose output is redirected.
If for any reason the program is killed (kill -9, hard reset of the computer, …), you may be missing some line of logs if you used print for the same.
However, the logging library will ensure to flush the logs printed to stderr and stdout immediately at any call.
I am a PHPer new to Python (2.7 on Win32) and I would like to know where Python is shoving any errors it finds?
Do I need to turn something on, if so where do I do that?
Or, is the idea that you develop using a shell and watch errors spat out via that?
Please share any other good Python debugging/sanity-saving mechanisms you wish you'd known about earlier - or if you have switched from PHP perhaps you can tell me what the Python equivalents of :
ini_set('error_reporting', 1);
display_errors();
trigger_error();
var_dump();
Try and Exceptions looks fairly similar.
I will probably stumble across these answers myself in time, but in the meantime this issue is bugging me (no pun intended).
Thanks a lot.
Python development is normally done in a shell, and you get a full traceback printed out on any uncaught exception.
If you want to log errors to file, have a look at the logging module. You can either catch exceptions directly, or override the sys.excepthook function which is called for an uncaught error. If you're using a framework for e.g. web development, it may have mechanisms to do this sort of thing already.
What I would like is to be able to step-debug code that runs in a separate process using the multiprocessing package.
I remember looking for a solution about a year ago and not finding one. I was told to just do a lot of logging, but of course it is an inferior method. So maybe someone came up with a solution in the meantime? For example, some mechanism for making the newly-spawned process connect with the debugger?
You could start the process You need to debug manually, without using the Process interface on this process.
You might find WingIDE useful. Its debugger is really nice, and it even supports debugging remote processes with some minimal instrumentation to the code being debugged. It's not free, but well worth the cost, IMHO. (I'm not affiliated with Wingware in any way, just a satisfied customer...)
To enable remote debugging in Wing, you need to copy the file wingdbstub.py to the same directory as the app you want to debug, and import it at the spot in your code you want debugging to begin. (This is covered fairly thoroughly in the WingIDE docs.)
If you take this example and modify the myfunc() method to look as follows:
def myfunc(conn, commands):
import wingdbstub
# ... remainder same as original example
you should be able to launch WingIDE, set a breakpoint immediately after the import line, then launch the example script from a console. It should automatically connect to Wing and stop at your breakpoint.
You might find this post helpful if you have any problems getting the debug connection to work. (The WingIDE docs also do a decent job of covering connection problems.)
Rather than starting your function or class via Process, just call it directly and debug as you normally would.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
So, I started learning to code in Python and later Django. The first times it was hard looking at tracebacks and actually figure out what I did wrong and where the syntax error was. Some time has passed now and some way along the way, I guess I got a routine in debugging my Django code. As this was done early in my coding experience, I sat down and wondered if how I was doing this was ineffective and could be done faster. I usually manage to find and correct the bugs in my code, but I wonder if I should be doing it faster?
I usually just use the debug info Django gives when enabled. When things do end up as I thought it would, I break the code flow a lot with a syntax error, and look at the variables at that point in the flow to figure out, where the code does something other than what I wanted.
But can this be improved? Are there some good tools or better ways to debug your Django code?
There are a bunch of ways to do it, but the most straightforward is to simply
use the Python debugger. Just add following line in to a Django view function:
import pdb; pdb.set_trace()
or
breakpoint() #from Python3.7
If you try to load that page in your browser, the browser will hang and you get a prompt to carry on debugging on actual executing code.
However there are other options (I am not recommending them):
* return HttpResponse({variable to inspect})
* print {variable to inspect}
* raise Exception({variable to inspect})
But the Python Debugger (pdb) is highly recommended for all types of Python code. If you are already into pdb, you'd also want to have a look at IPDB that uses ipython for debugging.
Some more useful extension to pdb are
pdb++, suggested by Antash.
pudb, suggested by PatDuJour.
Using the Python debugger in Django, suggested by Seafangs.
I really like Werkzeug's interactive debugger. It's similar to Django's debug page, except that you get an interactive shell on every level of the traceback. If you use the django-extensions, you get a runserver_plus managment command which starts the development server and gives you Werkzeug's debugger on exceptions.
Of course, you should only run this locally, as it gives anyone with a browser the rights to execute arbitrary python code in the context of the server.
A little quickie for template tags:
#register.filter
def pdb(element):
import pdb; pdb.set_trace()
return element
Now, inside a template you can do {{ template_var|pdb }} and enter a pdb session (given you're running the local devel server) where you can inspect element to your heart's content.
It's a very nice way to see what's happened to your object when it arrives at the template.
There are a few tools that cooperate well and can make your debugging task easier.
Most important is the Django debug toolbar.
Then you need good logging using the Python logging facility. You can send logging output to a log file, but an easier option is sending log output to firepython. To use this you need to use the Firefox browser with the firebug extension. Firepython includes a firebug plugin that will display any server-side logging in a Firebug tab.
Firebug itself is also critical for debugging the Javascript side of any app you develop. (Assuming you have some JS code of course).
I also liked django-viewtools for debugging views interactively using pdb, but I don't use it that much.
There are more useful tools like dozer for tracking down memory leaks (there are also other good suggestions given in answers here on SO for memory tracking).
I use PyCharm (same pydev engine as eclipse). Really helps me to visually be able to step through my code and see what is happening.
Almost everything has been mentioned so far, so I'll only add that instead of pdb.set_trace() one can use ipdb.set_trace() which uses iPython and therefore is more powerful (autocomplete and other goodies). This requires ipdb package, so you only need to pip install ipdb
I've pushed django-pdb to PyPI.
It's a simple app that means you don't need to edit your source code every time you want to break into pdb.
Installation is just...
pip install django-pdb
Add 'django_pdb' to your INSTALLED_APPS
You can now run: manage.py runserver --pdb to break into pdb at the start of every view...
bash: manage.py runserver --pdb
Validating models...
0 errors found
Django version 1.3, using settings 'testproject.settings'
Development server is running at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
GET /
function "myview" in testapp/views.py:6
args: ()
kwargs: {}
> /Users/tom/github/django-pdb/testproject/testapp/views.py(7)myview()
-> a = 1
(Pdb)
And run: manage.py test --pdb to break into pdb on test failures/errors...
bash: manage.py test testapp --pdb
Creating test database for alias 'default'...
E
======================================================================
>>> test_error (testapp.tests.SimpleTest)
----------------------------------------------------------------------
Traceback (most recent call last):
File ".../django-pdb/testproject/testapp/tests.py", line 16, in test_error
one_plus_one = four
NameError: global name 'four' is not defined
======================================================================
> /Users/tom/github/django-pdb/testproject/testapp/tests.py(16)test_error()
-> one_plus_one = four
(Pdb)
The project's hosted on GitHub, contributions are welcome of course.
The easiest way to debug python - especially for programmers that are used to Visual Studio - is using PTVS (Python Tools for Visual Studio).
The steps are simple:
Download and install it from https://microsoft.github.io/PTVS/
Set breakpoints and press F5.
Your breakpoint is hit, you can view/change the variables as easy as debugging C#/C++ programs.
That's all :)
If you want to debug Django using PTVS, you need to do the following:
In Project settings - General tab, set "Startup File" to "manage.py", the entry point of the Django program.
In Project settings - Debug tab, set "Script Arguments" to "runserver --noreload". The key point is the "--noreload" here. If you don't set it, your breakpoints won't be hit.
Enjoy it.
I use pyDev with Eclipse really good, set break points, step into code, view values on any objects and variables, try it.
I use PyCharm and stand by it all the way. It cost me a little but I have to say the advantage that I get out of it is priceless. I tried debugging from console and I do give people a lot of credit who can do that, but for me being able to visually debug my application(s) is great.
I have to say though, PyCharm does take a lot of memory. But then again, nothing good is free in life. They just came with their latest version 3. It also plays very well with Django, Flask and Google AppEngine. So, all in all, I'd say it's a great handy tool to have for any developer.
If you are not using it yet, I'd recommend to get the trial version for 30 days to take a look at the power of PyCharm. I'm sure there are other tools also available, such as Aptana. But I guess I just also like the way PyCharm looks. I feel very comfortable debugging my apps there.
From my perspective, we could break down common code debugging tasks into three distinct usage patterns:
Something has raised an exception: runserver_plus' Werkzeug debugger to the rescue. The ability to run custom code at all the trace levels is a killer. And if you're completely stuck, you can create a Gist to share with just a click.
Page is rendered, but the result is wrong: again, Werkzeug rocks. To make a breakpoint in code, just type assert False in the place you want to stop at.
Code works wrong, but the quick look doesn't help. Most probably, an algorithmic problem. Sigh. Then I usually fire up a console debugger PuDB: import pudb; pudb.set_trace(). The main advantage over [i]pdb is that PuDB (while looking as you're in 80's) makes setting custom watch expressions a breeze. And debugging a bunch of nested loops is much simpler with a GUI.
Ah, yes, the templates' woes. The most common (to me and my colleagues) problem is a wrong context: either you don't have a variable, or your variable doesn't have some attribute. If you're using debug toolbar, just inspect the context at the "Templates" section, or, if it's not sufficient, set a break in your views' code just after your context is filled up.
So it goes.
Add import pdb; pdb.set_trace() or breakpoint() (form python3.7) at the corresponding line in the Python code and execute it. The execution will stop with an interactive shell. In the shell you can execute Python code (i.e. print variables) or use commands such as:
c continue execution
n step to the next line within the same function
s step to the next line in this function or a called function
q quit the debugger/execution
Also see: https://poweruser.blog/setting-a-breakpoint-in-python-438e23fe6b28
Sometimes when I wan to explore around in a particular method and summoning pdb is just too cumbersome, I would add:
import IPython; IPython.embed()
IPython.embed() starts an IPython shell which have access to the local variables from the point where you call it.
I just found wdb (http://www.rkblog.rk.edu.pl/w/p/debugging-python-code-browser-wdb-debugger/?goback=%2Egde_25827_member_255996401). It has a pretty nice user interface / GUI with all the bells and whistles. Author says this about wdb -
"There are IDEs like PyCharm that have their own debuggers. They offer similar or equal set of features ... However to use them you have to use those specific IDEs (and some of then are non-free or may not be available for all platforms). Pick the right tool for your needs."
Thought i'd just pass it on.
Also a very helpful article about python debuggers:
https://zapier.com/engineering/debugging-python-boss/
Finally, if you'd like to see a nice graphical printout of your call stack in Django, checkout:
https://github.com/joerick/pyinstrument. Just add pyinstrument.middleware.ProfilerMiddleware to MIDDLEWARE_CLASSES, then add ?profile to the end of the request URL to activate the profiler.
Can also run pyinstrument from command line or by importing as a module.
I highly recommend epdb (Extended Python Debugger).
https://bitbucket.org/dugan/epdb
One thing I love about epdb for debugging Django or other Python webservers is the epdb.serve() command. This sets a trace and serves this on a local port that you can connect to. Typical use case:
I have a view that I want to go through step-by-step. I'll insert the following at the point I want to set the trace.
import epdb; epdb.serve()
Once this code gets executed, I open a Python interpreter and connect to the serving instance. I can analyze all the values and step through the code using the standard pdb commands like n, s, etc.
In [2]: import epdb; epdb.connect()
(Epdb) request
<WSGIRequest
path:/foo,
GET:<QueryDict: {}>,
POST:<QuestDict: {}>,
...
>
(Epdb) request.session.session_key
'i31kq7lljj3up5v7hbw9cff0rga2vlq5'
(Epdb) list
85 raise some_error.CustomError()
86
87 # Example login view
88 def login(request, username, password):
89 import epdb; epdb.serve()
90 -> return my_login_method(username, password)
91
92 # Example view to show session key
93 def get_session_key(request):
94 return request.session.session_key
95
And tons more that you can learn about typing epdb help at any time.
If you want to serve or connect to multiple epdb instances at the same time, you can specify the port to listen on (default is 8080). I.e.
import epdb; epdb.serve(4242)
>> import epdb; epdb.connect(host='192.168.3.2', port=4242)
host defaults to 'localhost' if not specified. I threw it in here to demonstrate how you can use this to debug something other than a local instance, like a development server on your local LAN. Obviously, if you do this be careful that the set trace never makes it onto your production server!
As a quick note, you can still do the same thing as the accepted answer with epdb (import epdb; epdb.set_trace()) but I wanted to highlight the serve functionality since I've found it so useful.
One of your best option to debug Django code is via wdb:
https://github.com/Kozea/wdb
wdb works with python 2 (2.6, 2.7), python 3 (3.2, 3.3, 3.4, 3.5) and pypy. Even better, it is possible to debug a python 2 program with a wdb server running on python 3 and vice-versa or debug a program running on a computer with a debugging server running on another computer inside a web page on a third computer!
Even betterer, it is now possible to pause a currently running python process/thread using code injection from the web interface. (This requires gdb and ptrace enabled)
In other words it's a very enhanced version of pdb directly in your browser with nice features.
Install and run the server, and in your code add:
import wdb
wdb.set_trace()
According to the author, main differences with respect to pdb are:
For those who don’t know the project, wdb is a python debugger like pdb, but with a slick web front-end and a lot of additional features, such as:
Source syntax highlighting
Visual breakpoints
Interactive code completion using jedi
Persistent breakpoints
Deep objects inspection using mouse Multithreading / Multiprocessing support
Remote debugging
Watch expressions
In debugger code edition
Popular web servers integration to break on error
In exception breaking during trace (not post-mortem) in contrary to the werkzeug debugger for instance
Breaking in currently running programs through code injection (on supported systems)
It has a great browser-based user interface. A joy to use! :)
I use PyCharm and different debug tools. Also have a nice articles set about easy set up those things for novices. You may start here. It tells about PDB and GUI debugging in general with Django projects. Hope someone would benefit from them.
If using Aptana for django development, watch this: http://www.youtube.com/watch?v=qQh-UQFltJQ
If not, consider using it.
Most options are alredy mentioned.
To print template context, I've created a simple library for that.
See https://github.com/edoburu/django-debugtools
You can use it to print template context without any {% load %} construct:
{% print var %} prints variable
{% print %} prints all
It uses a customized pprint format to display the variables in a <pre> tag.
I find Visual Studio Code is awesome for debugging Django apps. The standard python launch.json parameters run python manage.py with the debugger attached, so you can set breakpoints and step through your code as you like.
For those that can accidentally add pdb into live commits, I can suggest this extension of #Koobz answer:
#register.filter
def pdb(element):
from django.conf import settings
if settings.DEBUG:
import pdb
pdb.set_trace()
return element
From my own experience , there are two way:
use ipdb,which is a enhanced debugger likes pdb.
import ipdb;ipdb.set_trace() or breakpoint() (from python3.7)
use django shell ,just use the command below. This is very helpfull when you are developing a new view.
python manage.py shell
i highly suggest to use PDB.
import pdb
pdb.set_trace()
You can inspect all the variables values, step in to the function and much more.
https://docs.python.org/2/library/pdb.html
for checking out the all kind of request,response and hits to database.i am using django-debug-toolbar
https://github.com/django-debug-toolbar/django-debug-toolbar
As mentioned in other posts here - setting breakpoints in your code and walking thru the code to see if it behaves as you expected is a great way to learn something like Django until you have a good sense of how it all behaves - and what your code is doing.
To do this I would recommend using WingIde. Just like other mentioned IDEs nice and easy to use, nice layout and also easy to set breakpoints evaluate / modify the stack etc. Perfect for visualizing what your code is doing as you step through it. I'm a big fan of it.
Also I use PyCharm - it has excellent static code analysis and can help sometimes spot problems before you realize they are there.
As mentioned already django-debug-toolbar is essential - https://github.com/django-debug-toolbar/django-debug-toolbar
And while not explicitly a debug or analysis tool - one of my favorites is SQL Printing Middleware available from Django Snippets at https://djangosnippets.org/snippets/290/
This will display the SQL queries that your view has generated. This will give you a good sense of what the ORM is doing and if your queries are efficient or you need to rework your code (or add caching).
I find it invaluable for keeping an eye on query performance while developing and debugging my application.
Just one other tip - I modified it slightly for my own use to only show the summary and not the SQL statement.... So I always use it while developing and testing. I also added that if the len(connection.queries) is greater than a pre-defined threshold it displays an extra warning.
Then if I spot something bad (from a performance or number of queries perspective) is happening I turn back on the full display of the SQL statements to see exactly what is going on. Very handy when you are working on a large Django project with multiple developers.
use pdb or ipdb. Diffrence between these two is ipdb supports auto complete.
for pdb
import pdb
pdb.set_trace()
for ipdb
import ipdb
ipdb.set_trace()
For executing new line hit n key, for continue hit c key.
check more options by using help(pdb)
An additional suggestion.
You can leverage nosetests and pdb together, rather injecting pdb.set_trace() in your views manually. The advantage is that you can observe error conditions when they first start, potentially in 3rd party code.
Here's an error for me today.
TypeError at /db/hcm91dmo/catalog/records/
render_option() argument after * must be a sequence, not int
....
Error during template rendering
In template /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/crispy_forms/templates/bootstrap3/field.html, error at line 28
render_option() argument after * must be a sequence, not int
18
19 {% if field|is_checkboxselectmultiple %}
20 {% include 'bootstrap3/layout/checkboxselectmultiple.html' %}
21 {% endif %}
22
23 {% if field|is_radioselect %}
24 {% include 'bootstrap3/layout/radioselect.html' %}
25 {% endif %}
26
27 {% if not field|is_checkboxselectmultiple and not field|is_radioselect %}
28
{% if field|is_checkbox and form_show_labels %}
Now, I know this means that I goofed the constructor for the form, and I even have good idea of which field is a problem. But, can I use pdb to see what crispy forms is complaining about, within a template?
Yes, I can. Using the --pdb option on nosetests:
tests$ nosetests test_urls_catalog.py --pdb
As soon as I hit any exception (including ones handled gracefully), pdb stops where it happens and I can look around.
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/forms.py", line 537, in __str__
return self.as_widget()
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/forms.py", line 593, in as_widget
return force_text(widget.render(name, self.value(), attrs=attrs))
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/widgets.py", line 513, in render
options = self.render_options(choices, [value])
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/widgets.py", line 543, in render_options
output.append(self.render_option(selected_choices, *option))
TypeError: render_option() argument after * must be a sequence, not int
INFO lib.capture_middleware log write_to_index(http://localhost:8082/db/hcm91dmo/catalog/records.html)
INFO lib.capture_middleware log write_to_index:end
> /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/django/forms/widgets.py(543)render_options()
-> output.append(self.render_option(selected_choices, *option))
(Pdb) import pprint
(Pdb) pprint.PrettyPrinter(indent=4).pprint(self)
<django.forms.widgets.Select object at 0x115fe7d10>
(Pdb) pprint.PrettyPrinter(indent=4).pprint(vars(self))
{ 'attrs': { 'class': 'select form-control'},
'choices': [[('_', 'any type'), (7, (7, 'type 7', 'RECTYPE_TABLE'))]],
'is_required': False}
(Pdb)
Now, it's clear that my choices argument to the crispy field constructor was as it was a list within a list, rather than a list/tuple of tuples.
'choices': [[('_', 'any type'), (7, (7, 'type 7', 'RECTYPE_TABLE'))]]
The neat thing is that this pdb is taking place within crispy's code, not mine and I didn't need to insert it manually.
During development, adding a quick
assert False, value
can help diagnose problems in views or anywhere else, without the need to use a debugger.