A PyCharm 2018.3.3 Release Candidate is now available. It comes with several Python-related improvements and platform enhancements, which we want to double-check with you before finally releasing the update. Get it now from our Confluence page.
Apart from releasing this PyCharm 2018.3.3 RC, we’ve also rolled out an update for PyCharm 2018.2. It delivers a fix for lost Django support settings when adding an environment variable; for info on how to update it, skip to the end of this post.
Improved in PyCharm 2018.3.3 RC
The debugger’s on-demand variable loading policy for NumPy arrays and Pandas DataFrames
When the on-demand variable loading policy is enabled in the debug mode, PyCharm doesn’t load NumPy arrays and Pandas DataFrames values by default. Previously, loading all the values caused severe slowdowns in case of large datasets. You can load the values manually from the debugger’s variables view when needed by simply clicking “Show value”.
File encoding specification inspection for Python 3
In PyCharm 2018.3.3, we’ve added the option to enable the “No encoding specified for a file” inspection for Python 3. This option is useful for those who work on Python 2/3 compatible code.
- A fix for an extra __init__.py file created when moving a class into another module
- Several Docker and Docker Compose support fixes
- Fixes for embedded terminal
- Many fixes coming from WebStorm and IntelliJ IDEA – see the release notes for details
The release candidate (RC) is not an early access program (EAP) build and does not bundle an EAP license. To use PyCharm Professional Edition RC, you will need a currently active PyCharm subscription. If none is available, a free 30-day trial will start.
We’ve fixed a bug with lost Django settings when an environment variable is added for PyCharm 2018.2 (PY-32767). If you’re using PyCharm 2018.2, please update to the new version.