UPDATE

using: pyinstaller v. 6.6.0

pyi-makespec --icon=icon.ico --add-data main.py:. --add-data cert:. --add-data cert:cert --add-data static:static --add-data .streamlit:.streamlit --hidden-import streamlit --collect-all streamlit --copy-metadata streamlit run_main.py

Streamlit to Executable

tutorial

Create a virtual environment

pyenv virtualenv <version><env-name>#or python -m venv .<env-name>

Activate the virtual environment

pyenv activate <env-name>#or .\op a #.venv\Scripts\activate.bat

Verify that the virtual environment

python --version

To deactivate the virtual environment

pyenv deactivate #or .\op d #.venv\Scripts\deactivate.bat

Install Streamlit and [pyinstaller OR auto-py-to-exe] and any other lib you'll use

#I'm using streamlit==1.8.1 But in the #Tutorial I explain what could change for us.#So You could use the latest pip install streamlit pyinstaller

Now we add the main file called here app.py

echo> app.py

And we need to create an entry point for our executable

echo> run_app.py

At this point you should be able to add content

app.py:

importstreamlitasstif__name__=='__main__': st.header("Hello world")

run_app.py

fromstreamlit.webimportcli#this uri depends based on version of your streamlitif__name__=='__main__': cli._main_run_clExplicit('app.py', 'streamlit run') #we will CREATE this function inside our streamlit framework

Now we need to navigate into the path inside streamlit veins

in the version we are using here it is

.envir\Lib\site-packages\streamlit\web\cli.py

It's because we are using our virtual environment

now add our magic function

#... def main(log_level="info"):# [...]# you can you the name you prefer ofcourse# as long as you use underscore at the beginningdef_main_run_clExplicit(file, command_line, args=[],flag_options={}): main._is_running_with_streamlit=Truebootstrap.run(file, command_line, args,flag_options) # ...if __name__ == "__main__":#... main()

Now we need a hook to get streamlit metadata.

From the root of the project add a new folder and

create a file to be more organized because

in this folder will be created pycache

infos that we need to keep safe

.\hooks\hook-streamlit.py

the dash could indicate it will be read for pyinstaller as any other hook

( I actually don't know exacly why it is as it is)

fromPyInstaller.utils.hooksimportcopy_metadatadatas=copy_metadata('streamlit')

OK now we have all we need to compile the app

just run this line to create our first run_app.spec file

(if you are using the auto-py-to-exe we can't edit spec files from here we should edit from the interface in the advance options)

pyinstaller --onefile --additional-hooks-dir=./hooks run_app.py --clean #the onfile indicate we are create a output file join# everthing in it's binary#Some use case you actually need --ondir instead#the --clean delete cache and remove temporary files before building.#--additional-hooks-dir An additional path to search for hooks. This option can be used multiple times.

Now we just have to create an folder and file with streamlit configs

It can be add in your root project and the output folder

or just the output folder.

.streamlit\config.toml

[global] developmentMode = false [server] port = 8502

Assuming your have create this folder in the root project

as an expert streamlit you are

you can simply copy it to the output folder

xcopy /s /e .streamlit output/.streamlit #select D = directory

And perform the same for your app.py that contains streamlit logic

copy app.py output/app.py

Everything in place you now have only one thing to add

the DATAS to the new hook we created

in the run_app.spec that was generated add the following...

... a=Analysis( //... datas=[ (".env/Lib/site-packages/altair/vegalite/v4/schema/vega-lite-schema.json", "./altair/vegalite/v4/schema/"), (".env/Lib/site-packages/streamlit/static", "./streamlit/static"), (".env/Lib/site-packages/streamlit/runtime", "./streamlit/runtime"), //...) ...

## this path pair should be in that way# but I believe it is because we add the tuple as this templete# (absolut_path, parent_path)# so for files that is in the root of `Lib/site-packages` # We can add only the dot as parent # i.e: (".envir/Lib/site-packages/wmi.py",".")# for folders the behaviour is the same

All the modifications in the datas should be loaded using the command

pyinstaller run_app.spec --clean

🎈 It's done! run your run_app.exe file and see the magic 🪄

Huge Thanks To: hmasdevI'm organizing the solution from  hmasdev in the Streamlit Forum

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
.streamlit		.streamlit
build/run_app		build/run_app
dist		dist
hooks		hooks
.gitignore		.gitignore
README.md		README.md
app.py		app.py
op.bat		op.bat
run_app.py		run_app.py
run_app.spec		run_app.spec

sefgit/PyInstallerStreamlit

Folders and files

Latest commit

History

Repository files navigation

UPDATE

Streamlit to Executable

tutorial

Create a virtual environment

Activate the virtual environment

Verify that the virtual environment

To deactivate the virtual environment

Install Streamlit and [pyinstaller OR auto-py-to-exe] and any other lib you'll use

Now we add the main file called here app.py

And we need to create an entry point for our executable

At this point you should be able to add content

app.py:

run_app.py

Now we need to navigate into the path inside streamlit veins

in the version we are using here it is

.envir\Lib\site-packages\streamlit\web\cli.py

It's because we are using our virtual environment

now add our magic function

Now we need a hook to get streamlit metadata.

From the root of the project add a new folder and

create a file to be more organized because

in this folder will be created pycache

infos that we need to keep safe

.\hooks\hook-streamlit.py

the dash could indicate it will be read for pyinstaller as any other hook

( I actually don't know exacly why it is as it is)

OK now we have all we need to compile the app

just run this line to create our first run_app.spec file

(if you are using the auto-py-to-exe we can't edit spec files from here we should edit from the interface in the advance options)

Now we just have to create an folder and file with streamlit configs

It can be add in your root project and the output folder

or just the output folder.

.streamlit\config.toml

Assuming your have create this folder in the root project

as an expert streamlit you are

you can simply copy it to the output folder

And perform the same for your app.py that contains streamlit logic

Everything in place you now have only one thing to add

the DATAS to the new hook we created

in the run_app.spec that was generated add the following...

All the modifications in the datas should be loaded using the command

🎈 It's done! run your run_app.exe file and see the magic 🪄

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Languages

Packages