Call Map is distributed as a Python package. It requires Python 3.5
,
pip>=8.1.2
, and jedi==0.10.0
[*]. Other requirements can be installed
automatically once you have the correct versions of pip
and jedi
.
(If you don't have Python 3.5
, see Getting Python 3.5.)
Before installing for the first time, consider Using Virtualenv.
To install call_map
, from the top level directory of call_map
, run:
pip3 install --upgrade pip # make sure pip is >=8.1.2 pip3 install --no-cache-dir -e . # caching sometimes breaks
Note that some Python 3 distributions call the package manager pip
, and some
call it pip3
.
I have included instructions using conda
, apt
, and brew
if you prefer
obtaining as many depencencies as possible using those package managers. I also
include instructions for obtaining Python 3.5
with apt
and brew
.
More details about PyQt5
specifically can be found at the Riverbank download
page (Riverbank is the developer of PyQt5
).
[*] | A small part of call_map relies on internal structures of jedi , which
may change in future releases. I plan to make call_map rely less on
jedi internals in future call_map releases. |
Setting up virtual environment is optional but recommended. Virtualenv makes it easy to start over if things break, by deleting the virtualenv directory. To set it up, run:
python -m venv --system-site-packages $HOME/py3env export PATH=$HOME/py3env/bin:$PATH # update command hash table in bash/zsh hash pip3 hash python
You can obtain the official release or use a package manager. Specific instructions for Mac OS and Ubuntu are included below.
You can use brew
:
brew install python3
You can also use brew
to compile PyQt5:
brew install --with-python3 pyqt5
You can use apt
to install Python 3.5
and PyQt5
:
sudo apt install python3-pyqt5 virtualenv
You can use conda
to install pyqt
and qtconsole
, then pip
to install
jedi
and call_map
:
pip install conda hash conda conda install pygments conda install pyqt conda install qtconsole conda install toolz
Now follow the pip
installation instructions in Standard Installation with
Pip. pip
will not reinstall the packages you installed with conda.