Merge pull request #4823 from plotly/anywidget

Switch to use anywidget

Author: marthacryan

.circleci/config.yml

@@ -46,6 +46,9 @@ commands:
             . venv/bin/activate
             pip install --upgrade pip wheel
             pip install -r ./test_requirements/requirements_<<parameters.py>>_optional.txt
+            cd js
+            npm ci
+            npm run build
       - run:
           name: Install plotly-geo
           command: |
@@ -347,9 +350,9 @@ jobs:
             . venv/bin/activate
             pip install --upgrade pip wheel
             pip install -r ./test_requirements/requirements_38_core.txt black inflect
-            pip install jupyterlab~=3.0
+            pip install jupyterlab
       - run:
-          name: Update jupyterlab-plotly version
+          name: Update jupyter widget plotly.js version
           command: |
             cd packages/python/plotly
             . venv/bin/activate
@@ -403,7 +406,7 @@ jobs:
             conda config --remove channels defaults
             conda config --add channels conda-forge
             conda create -n env --yes python=3.9 conda-build=3.28.4 conda-verify
-            conda install -n env -c conda-forge jupyterlab=3 nodejs=16
+            conda install -n env -c conda-forge jupyterlab nodejs=16
             conda init bash
             mkdir output
 
@@ -412,9 +415,9 @@ jobs:
           command: |
             eval "$(conda shell.bash hook)"
             conda activate env
-            cd packages/javascript/jupyterlab-plotly
+            cd packages/python/plotly/js
             npm ci
-            npm run build:prod
+            npm run build
             git status
 
       - run:
@@ -438,13 +441,13 @@ jobs:
             git status
 
       - run:
-          name: NPM Pack
+          name: Build Widget javascript bundle
           command: |
             eval "$(conda shell.bash hook)"
             conda activate env
-            cd packages/javascript/jupyterlab-plotly
-            npm pack
-            mv jupyterlab-plotly*.tgz ../../../output
+            cd packages/python/plotly/js
+            npm ci
+            npm run build
 
       - run:
           name: Zip output
@@ -491,7 +494,12 @@ jobs:
             pip install -r requirements.txt
             if [ "${CIRCLE_BRANCH}" != "doc-prod" ]; then
               pip uninstall -y plotly
-              pip install -e ../packages/python/plotly
+              cd ../packages/python/plotly
+              pip install -e .
+              cd js
+              npm ci
+              npm run build
+              cd ../../../../doc
             fi
             cd ..
       - run:

.gitignore

@@ -59,6 +59,7 @@ doc/check-or-enforce-order.py
 packages/javascript/jupyterlab-plotly/lib/
 packages/python/plotly/jupyterlab_plotly/labextension/
 packages/python/plotly/jupyterlab_plotly/nbextension/index.js*
+packages/python/plotly/plotly/package_data/widgetbundle.js
 
 test/percy/*.html
 test/percy/pandas2/*.html

README.md

@@ -37,13 +37,10 @@
   </a>
 </div>
 
-
 ## Quickstart
 
 `pip install plotly==5.24.1`
 
-Inside [Jupyter](https://jupyter.org/install) (installable with `pip install "jupyterlab>=3" "ipywidgets>=7.6"`):
-
 ```python
 import plotly.express as px
 fig = px.bar(x=["a", "b", "c"], y=[1, 3, 2])
@@ -52,7 +49,6 @@ fig.show()
 
 See the [Python documentation](https://plotly.com/python/) for more examples.
 
-
 ## Overview
 
 [plotly.py](https://plotly.com/python/) is an interactive, open-source, and browser-based graphing library for Python :sparkles:
@@ -82,7 +78,7 @@ Built on top of [plotly.js](https://github.com/plotly/plotly.js), `plotly.py` is
 
 ## Installation
 
-plotly.py may be installed using pip...
+plotly.py may be installed using pip
 
 ```
 pip install plotly==5.24.1
@@ -94,43 +90,19 @@ or conda.
 conda install -c plotly plotly=5.24.1
 ```
 
-### JupyterLab Support
+### Jupyter Widget Support
 
-For use in [JupyterLab](https://jupyterlab.readthedocs.io/en/stable/), install the `jupyterlab` and `ipywidgets`
+For use as a Jupyter widget, install `jupyter` and `anywidget`
 packages using `pip`:
 
 ```
-pip install "jupyterlab>=3" "ipywidgets>=7.6"
+pip install jupyter anywidget
 ```
 
 or `conda`:
 
 ```
-conda install "jupyterlab>=3" "ipywidgets>=7.6"
-```
-
-The instructions above apply to JupyterLab 3.x. **For JupyterLab 2 or earlier**, run the following commands to install the required JupyterLab extensions (note that this will require [`node`](https://nodejs.org/) to be installed):
-
-```
-# JupyterLab 2.x renderer support
-jupyter labextension install [email protected] @jupyter-widgets/jupyterlab-manager
-```
-
-Please check out our [Troubleshooting guide](https://plotly.com/python/troubleshooting/) if you run into any problems with JupyterLab.
-
-### Jupyter Notebook Support
-
-For use in the Jupyter Notebook, install the `notebook` and `ipywidgets`
-packages using `pip`:
-
-```
-pip install "notebook>=5.3" "ipywidgets>=7.5"
-```
-
-or `conda`:
-
-```
-conda install "notebook>=5.3" "ipywidgets>=7.5"
+conda install jupyter anywidget
 ```
 
 ### Static Image Export
@@ -143,41 +115,18 @@ command line utility (legacy as of `plotly` version 4.9).
 #### Kaleido
 
 The [`kaleido`](https://github.com/plotly/Kaleido) package has no dependencies and can be installed
-using pip...
+using pip
 
 ```
 pip install -U kaleido
 ```
 
-or conda.
+or conda
 
 ```
 conda install -c conda-forge python-kaleido
 ```
 
-#### Orca
-
-While Kaleido is now the recommended image export approach because it is easier to install
-and more widely compatible, [static image export](https://plotly.com/python/static-image-export/)
-can also be supported
-by the legacy [orca](https://github.com/plotly/orca) command line utility and the
- [`psutil`](https://github.com/giampaolo/psutil) Python package.
-
-These dependencies can both be installed using conda:
-
-```
-conda install -c plotly plotly-orca==1.3.1 psutil
-```
-
-Or, `psutil` can be installed using pip...
-
-```
-pip install psutil
-```
-
-and orca can be installed according to the instructions in the [orca README](https://github.com/plotly/orca).
-
-
 ### Extended Geo Support
 
 Some plotly.py features rely on fairly large geographic shape files. The county

contributing.md

@@ -157,7 +157,14 @@ complete installation and avoid gdal-config errors.
 (plotly_dev) $ pip install -e packages/python/chart-studio/
 (plotly_dev) $ pip install -e packages/python/plotly-geo/
 ```
-This will ensure that the installed packages links to your local development
+
+**Note**: To test `go.FigureWidget` locally, you'll need to generate the javascript bundle as follows:
+```
+cd packages/python/plotly/js
+npm install && npm run build
+```
+
+Running `pip install -e` will ensure that the installed packages links to your local development
 directory, meaning that all changes you make reflect directly in your
 environment (don't forget to restart the Jupyter kernel though!). For more
 information see the
@@ -211,7 +218,7 @@ make that pull request!
 
 
 ## Update to a new version of Plotly.js
-First update the version of the `plotly.js` dependency in `packages/javascript/jupyterlab-plotly/package.json`.
+First update the version of the `plotly.js` dependency in `packages/python/plotly/js/package.json`.
 
 Then run the `updateplotlyjs` command with:
 

doc/python/getting-started.md

@@ -96,25 +96,16 @@ IFrame(snippet_url + 'getting-started', width='100%', height=1200)
 
 #### JupyterLab Support
 
-To use `plotly` in [JupyterLab](https://jupyterlab.readthedocs.io/en/stable/), install the `jupyterlab` and `ipywidgets` packages in the same environment as you installed `plotly`, using `pip`:
+To use `plotly` in [JupyterLab](https://jupyterlab.readthedocs.io/en/stable/), install the `jupyterlab` and `anywidget` packages in the same environment as you installed `plotly`, using `pip`:
 
 ```
-$ pip install "jupyterlab>=3" "ipywidgets>=7.6"
+$ pip install jupyterlab anywidget
 ```
 
 or `conda`:
 
 ```
-$ conda install "jupyterlab>=3" "ipywidgets>=7.6"
-```
-
-The `plotly` jupyterlab extension is included when you install `plotly` using `pip` or `conda`. When you run Jupyter Lab, ensure you are running it in the same environment that you installed `plotly` in so it has access to the `plotly` jupyterlab extension.
-
-**JupyterLab 2 or earlier** does not support the prebuilt extensions installed using `pip` and `conda`. If you are using JupyterLab 2, run the following command to install the required JupyterLab extensions (note that this will require [`node`](https://nodejs.org/) to be installed):
-
-```
-# JupyterLab 2.x renderer support
-jupyter labextension install [email protected] @jupyter-widgets/jupyterlab-manager
+$ conda install jupyterlab anywidget
 ```
 
 Launch JupyterLab with:

doc/python/troubleshooting.md

@@ -68,87 +68,6 @@ IFrame(snippet_url + 'renderers', width='100%', height=1200)
 <div style="font-size: 0.9em;"><div style="width: calc(100% - 30px); box-shadow: none; border: thin solid rgb(229, 229, 229);"><div style="padding: 5px;"><div><p><strong>Sign up for Dash Club</strong> → Free cheat sheets plus updates from Chris Parmer and Adam Schroeder delivered to your inbox every two months. Includes tips and tricks, community apps, and deep dives into the Dash architecture.
 <u><a href="https://go.plotly.com/dash-club?utm_source=Dash+Club+2022&utm_medium=graphing_libraries&utm_content=inline">Join now</a></u>.</p></div></div></div></div>
 
-<!-- #region -->
-### JupyterLab Problems
-
-In order to use `plotly` in JupyterLab, you *must have the `jupyterlab-plotly` extension installed* as detailed in the [Getting Started guide](/python/getting-started). When you install `plotly`, this extension is automatically made available to any JupyterLab 3.x installation in the same Python environment. 
-
-To list your current extensions, run the following command in a terminal shell **from the same environment as JupyterLab is launched**:
-
-```bash
-# Check that jupyterlab-plotly is installed
-$ jupyter labextension list
-```
-
-Please note that the *extension version matters*: the extension versions in the [Getting Started](/python/getting-started) guide match the version of `plotly` at the top of the guide and so they should be installed together. Note also that these extensions are meant to work with JupyterLab 1 or above but not 0.x.
-
-If automatic installation of the extension is not working in your environment, or if you are using JupyterLab 1.x or 2.0, you may install it manually using the following command, which requires `node` to be installed.
-
-```bash
-# Manually reinstall the extension
-$ jupyter labextension install @jupyter-widgets/jupyterlab-manager jupyterlab-plotly
-```
-
-If you have [installed additional python environments](https://ipython.readthedocs.io/en/stable/install/kernel_install.html) (or kernels) to use with JupyterLab, or if you are using a centrally hosted JupyterLab installation, you need to make sure that the extensions are installed in the python environment used to launch JupyterLab (the "server" environment). If you accidentally installed the extensions (and run the command above) in one of the additional python environments ("processing" environments), then it is possible for the command above to list the correct extensions but for them to not be available in the JupyterLab front-end you have loaded in your browser. To check if this is the problem, you can [look at the active extension list through your browser via the JupyterLab Extension Manager](https://jupyterlab.readthedocs.io/en/stable/user/extensions.html#using-the-extension-manager), which will always list the extensions in the "server" environment. To summarize: if you use JupyterLab with multiple python environments, the extensions must be installed in the "server" environment, and the plotly python library must be installed in each "processing" environment that you intend to use.
-
-> Note that version 4.14.3 of `plotly` or earlier needed two extensions (`jupyterlab-plotly` and `plotlywidget`) to be installed manually running, and that `plotlywidget` requires `@jupyter-widgets/jupyterlab-manager` to be installed:
-
-```bash
-# Instructions for `plotly` 4.x
-$ jupyter labextension install jupyterlab-plotly plotlywidget @jupyter-widgets/jupyterlab-manager 
-```
-
-If you have the correct version(s) of the extension(s) installed and active in your active JupyterLab sessions and are still seeing problems, the issue may clear up if you rebuild JupyterLab. This shouldn't be required in principle but some users have resolved their issues this way. To rebuild JupyterLab, shut down JupyterLab and run the following command in a terminal shell **from the same environment as JupyterLab was launched**:
-
-```bash
-# rebuilding JupyterLab
-$ jupyter lab build
-```
-
-To uninstall your Plotly extensions prior to reinstalling them, run the following commands in a terminal shell before reinstalling them by following the instructions in the [Getting Started guide](/python/getting-started):
-
-```bash
-# uninstalling extensions to reinstall
-$ jupyter labextension uninstall jupyterlab-plotly
-$ jupyter labextension uninstall plotlywidget
-```
-
-If you run into "out of memory" problems while installing the extensions or building JupyterLab, try running these commands before running `jupyter labextension install`...
-
-```bash
-# Avoid "JavaScript heap out of memory" errors during extension installation
-# (OS X/Linux)
-export NODE_OPTIONS=--max-old-space-size=4096
-# (Windows)
-set NODE_OPTIONS=--max-old-space-size=4096
-```
-
-...and these commands afterwards.
-
-```bash
-# Unset NODE_OPTIONS environment variable
-# (OS X/Linux)
-unset NODE_OPTIONS
-# (Windows)
-set NODE_OPTIONS=
-```
-
-### Jupyter Classic Notebook Problems
-
-The classic Jupyter Notebook (i.e. launched with `jupyter notebook`) sometimes suffers from a problem whereby if you close the window and reopen it, your plots render as blank spaces.
-
-The easiest solution is to force the `notebook` renderer to reload by calling `fig.show("notebook")` instead of just `fig.show()`.
-
-If this problem is recurrent, you may safely run the following code in a Notebook (not in JupyterLab!) at any time and it should restore your figures (for example, you may put it at the top of your notebook for easy access):
-
-```python
-import plotly.io as pio
-pio.renderers.default='notebook'
-```
-
-As a last resort, you can "Restart & Clear Output" from the Kernel menu and rerun your notebook.
-
-<!-- #endregion -->
 
 ### VSCode Notebook, Nteract and Streamlit Problems
 

doc/requirements.txt

@@ -1,6 +1,5 @@
 plotly==5.24.1
 jupytext==1.16.4
-ipywidgets==7.7.2
 jupyter-client<7
 jupyter
 notebook
@@ -46,3 +45,4 @@ dask==2022.2.0
 polars
 geoparse<=2.0.3
 xarray==2022.9.0
+anywidget