Teasing Streamlit’s Next Biggest Changes

The 5 latest releases seem to prepare the bigger Streamlit milestones. In this video you will see how the new features are smaller steps to big changes like multipage v2, dialog window and partial rerun/experimental fragment
______
☕ Want to support me? www.buymeacoffee.com/andfanilo
🐦Follow my daily updates on Twitter: / andfanilo
🗣️ Find me on: andfanilo.com
🔗 Links
- Popover Demo: popover-container-demo.stream...
- Subtitle Demo: doc-video-subtitle-inputs.str...
- Multiple Chat Input Demo: inline-chat-input-demo.stream...
- Partial Rerun / Experimental Fragment Demo: st-experimental-fragment.stre...
👉 My tools (Affiliate links to support me!)
- 📷 Main Camera - amzn.to/3QvTosB
- 📷 Main Lens - amzn.to/3svYCwq
- 🎙️ Microphone - amzn.to/47l5ewA
- 🎵 Music & Sound Effects - www.epidemicsound.com/referra...
- 📕 Streamlit Official Book - amzn.to/3QuRPwl
00:00 Intro
00:23 Chat Inputs
03:12 Popover to Dialog+Rerun
06:12 Multipage almost v2
08:42 Everything Else
______
🖋 I could've edited this last weekend, but I found a great Wuxia I binge-watched instead XD
⚠️ Disclaimer: This video is not sponsored, I receive no compensation by any brand quoted in this video. Views are my own and do not represent my employer's.
Links included in this description might be affiliate links. If you purchase a product or service with the links that I provide I may receive a small commission. Thank you for supporting my channel so I can continue providing you with free content!
#streamlit #python #datascience #dataapps

[@rockNbrain]

Great video Fanilo

[@andfanilo]

Thanks for watching and for the support 🙂 looking forward to those features finally releasing

[@nicapotato7745]

Thanks for sharing the deeper details, I'm a data coder getting my mind blown here haha

[@andfanilo]

Eheh thanks for watching! Any feature you’re looking forward to trying ??

[@comfixit]

A great job presenting the new features. You are the best of the yourtube coder bros. A+. Would have been an A++ if you had included a drinking tea reference. I miss those. Excellent work.

[@andfanilo]

Taking note for the next one 😂 thanks for the support, see you on the next one!

[@Sonntagssoziologe]

I like your chanel so much.

[@andfanilo]

Happy to hear that! Stay tuned for the next one :)

[@michael-jones]

Thank you for your content. I saw a click “to accept cookies” button. Do you have a video on cookies and persistent sessions?

[@andfanilo]

Hey! I don't know why Chrome dislpayed that "add cookies" because I wasn't passing any cookies in the demo. Just Google reacting the random text_inputs... I don't have a video on cookies/persistent sessions, but there are multiple Streamlit components to manage cookies though, the moset popular I think being github.com/Mohamed-512/Extra-Streamlit-Components Otherwise it would be better to configure that at the load balancer level, NGINX or other; as you don't have access to Streamlit's underlying Tornado server to configure this.

[@rodrigobogado653]

oh very good tips,

[@andfanilo]

Glad you like them! Hope to see you on the next one :)

[@salvatorepascarella8368]

Hi I really enjoy your videos and the way you explain it. I was wondering if you could bring a video about any way to filter a data_editor in Streamlit or if there’s a workaround about it. Thanks and keep going!

[@andfanilo]

Hello, thanks for the kind words :) I'm not totally sure I understood the filter part, do you mean widgets external to data_editor that filter it? Or filters integrated in the column for example like in Excel? Or a feature that Mitosheet/Aggrid provide that data_editor doesn't?

[@salvatorepascarella8368]

@@andfanilo Sorry for not being clear, I was talking about some filters Excel-like, where you can filter the columns you want.

[@RyanDoozer]

hype!

[@andfanilo]

👏 soon !

[@apiladhikari1557]

bad 'setin' index #(should be between [#,#]) I've been facing this issue in my code for a long time now, even the github issue for it is marked as closed without properly solving it, how do you think i can tackle this?

[@CarlosEMijaresR]

My problem is that I already have my app deployed in Streamlit, and the terminal command to hide the menu works perfectly for me, but how do I make it work the same with my app deployed on the platform directly?

[@andfanilo]

docs.streamlit.io/develop/concepts/configuration/options You can add the configuration option by pushing a `.streamlit/config.toml` file at the root of your project, it should take the following form: ``` [client] showSidebarNavigation = false ``` or pass those as environment variables `STREAMLIT_CLIENT_SHOW_SIDEBAR_NAVIGATION=false` from the configuration menu of your deployment

[@pedrotiagobh]

Hi, I had this error below while I was trying to create the .exe file from my streamlit code: Error: The entrypoint file "app.py" is not included in the bundled files. Can someone help me?

[@andfanilo]

I haven't tried stlite in a while but from github.com/whitphx/stlite/tree/main/packages/desktop there are some new configurations options to add to point to the entrypoiny.. But the honest best way would be to head to github.com/whitphx/stlite/issues if it still doesn't work, the author is very helpful there!

[@thamerrro4234]

Too late, I'm halfway done switching my app to react...

[@thamerrro4234]

still, I'll mostly likely use streamlit again, for upcoming projects.

[@andfanilo]

Ahah can't blame you if you found a good JS workflow 😊 any React framework to recommend for people who want to switch?

[@pmshadow]

@@andfanilo I am bulding an internal application at my hedge fund, Nextjs for the "Software part" and Streamlit is our data playground. What do you think of this combination?

[@andfanilo]

Sounds nice, I like it 😊 if you have a good Next.js theme to recommend and a JS interactive charting library, please share!

[@thamerrro4234]

​ @andfanilo I'm no react expert, I just started react I'll share what I use Next.js + React as my "frontend" And I use FastAPI as a python backend. I have some (AI components) I used" vercel/ai js package" for two days then I decided not to use It and switched to fetch for streaming texts(from my python backend). it works very well. I also use Mantine as a component library it is very nice, easy to use and has no "Pro" components that I know of. For state management I use createContext and useContext they work very well, almost like st.session I have a folder called /context, in it there are two files (contexts.tsx + types.ts) I then I pass the context to layout.tsx so I could use the contexts from anywhere. just like st.session. Once I got everything going, took like a week or two (note I'm starting fresh in react) I think I did quite good (thanks to copilot)