1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
|
|project| & PyInstaller
#######################
`PyInstaller <https://www.pyinstaller.org/>`_ lets you freeze your python application into a
stand-alone executable. This installer supports Linux, macOS, Windows, and more; and is also
compatible with 3rd-party Python modules, such as |pymodname|.
For more details, see the `official documentation <https://www.pyinstaller.org/documentation.html>`_.
Preparation
===========
Install the `PyInstaller` via **pip** with the following command::
pip install pyinstaller
If you're using a virtual environment, remember to activate it before installing `PyInstaller`.
After installation, the `pyinstaller` binary is located in your virtual environment's `bin/`
directory, or where your Python executable is located. If that directory isn't in your `PATH`,
include the whole path when you run `pyinstaller`.
.. warning:: If you already have a PySide2 or Shiboken2 version installed in your
system path, PyInstaller uses them instead of your virtual environment version.
Freeze an application
=======================
`PyInstaller` has many options that you can use. To list them all, run `pyinstaller -h`.
There are two main features:
* the option to package the whole project (including shared libraries) into one executable file
(`--onefile`)
* the option to place it in a directory containing the libraries
Additionally, on Windows when the command is running, you can open a console with the `-c` option
(or `--console` or `--nowindowed` equivalent).
Otherwise, you can specify to not open such a console window on macOS and Windows with the `-w`
option (or `--windowed` or `--noconsole` equivalent).
Create an example
-----------------
Now, consider the following script, named `hello.py`::
import sys
import random
from PySide2.QtWidgets import (QApplication, QLabel, QPushButton,
QVBoxLayout, QWidget)
from PySide2.QtCore import Slot, Qt
class MyWidget(QWidget):
def __init__(self):
QWidget.__init__(self)
self.hello = ["Hallo Welt", "你好，世界", "Hei maailma",
"Hola Mundo", "Привет мир"]
self.button = QPushButton("Click me!")
self.text = QLabel("Hello World")
self.text.setAlignment(Qt.AlignCenter)
self.layout = QVBoxLayout()
self.layout.addWidget(self.text)
self.layout.addWidget(self.button)
self.setLayout(self.layout)
# Connecting the signal
self.button.clicked.connect(self.magic)
@Slot()
def magic(self):
self.text.setText(random.choice(self.hello))
if __name__ == "__main__":
app = QApplication(sys.argv)
widget = MyWidget()
widget.resize(800, 600)
widget.show()
sys.exit(app.exec_())
Since it has a UI, you use the `--windowed` option.
The command line to proceed looks like this::
pyinstaller --name="MyApplication" --windowed hello.py
This process creates two directories: `dist/` and `build/`. The application executable and the
required shared libraries are placed in `dist/MyApplication`.
To run the application, go to `dist/MyApplication` and run the program::
cd dist/MyApplication/
./MyApplication
.. note:: The directory inside `dist/` and the executable have the same name.
Use the `--onefile` option if you prefer to have everything bundled into one executable, without
the shared libraries next to it::
pyinstaller --name="MyApplication" --windowed --onefile hello.py
This process takes a bit longer, but in the end you have one executable in the `dist/` directory::
cd dist/
./MyApplication
Some Caveats
============
PyInstaller Issue
-----------------
As mentioned before, if available, `PyInstaller` picks a system installation of PySide2 or
Shiboken2 instead of your `virtualenv` version without notice. This is negligible if those
two versions are the same.
If you're working with different versions, this can result in frustrating debugging sessions
when you think you are testing the latest version, but `PyInstaller` is working with an older
version.
Issue with numpy in Python 2.7.16
---------------------------------
A recent issue with PyInstaller is the appearance of Python 2.7.16. This Python version creates
an issue that is known from Python 3 as a `Tcl/Tk` problem. This rarely shows up in Python 3
because `Tcl/Tk` is seldom used with `PyInstaller`.
On Python 2.7.16, this problem is common, as many developers use numpy. For some reason,
installing `numpy` creates a dependency to `Tcl/Tk`, which can be circumvented only by explicitly
excluding `Tcl/Tk` by adding this line to spec-file's analysis section::
excludes=['FixTk', 'tcl', 'tk', '_tkinter', 'tkinter', 'Tkinter'],
Safety Instructions
-------------------
- When using `PyInstaller` with `virtualenv`, make sure that there is no system
installation of PySide2 or shiboken2.
- Before compiling, use `pip -uninstall pyside2 shiboken2 -y` multiple times, until
none of the programs are found anymore.
- Pip is usually a good tool. But to be 100 % sure, you should directly remove
the PySide2 and shiboken2 folders from site-packages.
- Be sure to use the right version of pip. The safest way to really run the right
pip, is to use the Python that you mean: Instead of the pip command, better use::
<path/to/your/>python -m pip
|
