When comparing PyCharm vs Jupyter, the Slant community recommends PyCharm for most people. In the question“What are the best Python IDEs or editors?” PyCharm is ranked 1st while Jupyter is ranked 19th. The most important reason people chose PyCharm is:
PyCharm has CVS, Git, Subversion and Mercurial integration.
Ranked in these QuestionsQuestion Ranking
Pro Version control integration
PyCharm has CVS, Git, Subversion and Mercurial integration.
Pro Catches run-time information when running the code
PyCharm can leverage run-time information when running your application with the built-in debugger to figure out what types can possibly be passed to which functions, etc.
Pro One of the best autocompletion engines around
PyCharm has two types of autocompletion: structural completion and word expansion.
Both types of autocompletion work extremely well, have little to no problems and are quite fast even when loading suggestions on the go.
Pro Free version available
There's a community edition (with limited features) that's free to use. You can also get a 30-day trial of the Professional edition.
Pro Great pip support
PyCharm offers great pip integration. When opening a project it automatically checks for a
requirements.txt file in the root of the project. If it's found, it checks if all the libraries are available in the interpreter. If one or more libraries are missing, it issues a warning and asks whether you want to install any missing libraries.
Pro Excellent refactoring support
There are many refactoring options including renaming and changing signature across entire projects. It also includes the an ability to preview changes before committing and exclude anything unwanted.
Pro Automatically figures out what test to run based on the method the cursor rests at a given time
PyCharm, based on what method or class the cursor rests, can figure out what tests to run and perform them with a keyboard shortcut or two, without breaking up the flow and need to switch to a command line interface.
Pro Great for navigating large codebases
PyCharm has amazing code navigation implementations. It supports both goto symbol and goto declaration. The former finds classes, variables, functions, etc by name. While the latter is used by moving the cursor on top of a symbol and by using the mouse or a keyboard combination it finds the declaration of that symbol and takes you there.
Both of these features are extremely helpful when consulting large code-bases and when trying to understand an API written by someone else.
Pro Excellent integration with debugging tools
All the debugging can be done inside the IDE. Breakpoints in the code can be added using keyboard shortcuts or the mouse. When the code is executed through the debugger a toolbar pops up with all the relevant context needed for the debugging process.
The whole process is smooth and painless and you don't even have to switch windows to do the debugging.
Pro Supports installing third party libraries
No need to go to the command line to download a new package, PyCharm has an easy system to browse, download, and update third party packages.
Pro Sophisticated static analysis tools
Pro Remote debugging over ssh coupled with automatic deployment creates a streamlined workflow
The professional version allows remote debugging over ssh, which together with automatic deployment creates a streamlined workflow.
Pro Built-in Django support
Pycharm has excellent django support, from templating to management commands, it has it all.
Pro Free student access to Professional Edition
With a valid .edu address students can register to use the Professional edition and enjoy all the perks of the full paid version for free.
Though it should be mentioned that the with the free student acess you cannot use PyCharm for any commercial purposes, even accepting donations for an open source project.
Pro Vim mode for people used to Vim commands
IdeaVim supports motion keys, insert mode commands, marks, registers, visual mode commands, vim regexps, key mapping, macros, digraphs, some ex and :set commands. You can find a full comparison in the IdeaVim reference manual.
Pro Easy to optimize code with built-in profiling tools
If you have a yappi profiler installed on your interpreter, PyCharm starts the profiling session with it by default, otherwise it uses the standard cProfile profiler.
Pro Has a lot of plugins
PyCharm offers a high variety of plugins like Pylin, Mypy etc. covering all the above mentioned. Plus it has a built-in support to detect wrong formatted/named things (inspection).
Pro Web-based development allows for usage literally anywhere
Because the editor is a web app (the Jupyter Notebook program is a web server that you run on the host machine), it is possible to use this on quite literally any machine. Morever, you can have Jupyter Notebook run on one machine (like a VM that you have provisioned in the cloud) and access the web page / do your editing from a different machine (like a Chromebook).
Pro Supports multiple different programming languages
Jupyter Notebook, formerly known as ipython, used to be specific to Python; however, in recent iterations, it has become capable of general purpose usage for any programming language. Thus it is possible to use this and have a consistent developer workflow, regardless of language.
Most IDEs require you to separately run Python to see the output of a particular piece of code. By contrast, Jupyter Notebook can evaluate Python statements inline, giving you the immediate feedback of interactive use of the interpreter while keeping your changes saved.
Pro Open source
Because it is open source, you can review the source code and also propose extensions and fixes to it. It is also possible to fork the repository and make changes to it to customize it for your specific use case.
Pro Graphing , charting, and other math/numeric capabilities
The interactive editor is able to display complex equations, charts, graphs, etc. making this particular editor very well-regarded among data scientists.
Con Very high memory usage
Con Vim mode is limited
Con Not suitable to edit project's files written in other languages
There is a high variety of support for a lot of languages like markdown etc. Not for Java and so on, but it is a Python IDE.
Con Rendering is awful
Con Some relatively basic functionality requires paid license
Con Assinine licensing scheme
JetBrains licensing, especially if you have multiple products, is a blocker. You just can't a fixed line-item price (for departmental budgeting) for their licenses.
Con It cannot reindex on the fly packages installed from git source
If you've installed a package with the command:
pip install -e git+https://email@example.comfirstname.lastname@example.org#egg=package
you have two options available to make PyCharm update/see it:
- restart PyCharm
- invalidate caches
Con Sometimes all autocomplete stuff dies with over 9k Java exceptions with no visible reason
Con Odd Autosave "feature", can't be disabled fully
PyCharm automatically saves your files for you, always, without telling you. You can't disable this. There's a way to indicate if a file has been modified via an indicator in the tab (not enabled by default - why?).
If you exit it won't ask you if you want to save the modified file. Totally unintuitive and contrary to all other established workflows. It's ok to try something new, but give users the option to have the "normal" behaviour of any other IDE/editor out there. Can be a deal breaker for those that need to know/have control over when they save their files. (PyCharm offers a history to undo the automatic save, but why force a user to undo something with extra steps that shouldn't have happened in the first place?)
Con Not possible to run scripts in a single console
Con Interactive usage takes some getting used to
While the interactiveness is extremely, extremely powerful and useful, it does take a little bit of work getting to a point where it is "normal".
Con First time setup is more difficult than for other IDEs
Since Jupyter Notebook really requires two programs (the server and your browser) getting things setup in a way that works for you is a little more complex than for an ordinary IDE. For example, if you run the server and edit on the same machine, creating a little wrapper script that starts the server and then launches the browser pointing to it and gives an icon to this script is a small amount of setup but is more involved than a simple installer for other IDEs. Likewise, if you do remote development, creating a URL that will lazily spawn the Jupyter Notebook server and then turn it down when it is no longer in use is also a little bit of work to setup.
Con Non-trivial security configuration for remote access
By default, the editor is only accessible from localhost; however, if you want to run Jupyter on a VM in the cloud and do your editing through a web browser on a different computer (e.g. a Chromebook), there is some non-trivial security work to ensure that it is set up in a secure manner.