Merge branch 'codegen2' into type-annotations

Author: gvwilson

.circleci/config.yml

@@ -2,10 +2,17 @@
 All notable changes to this project will be documented in this file.
 This project adheres to [Semantic Versioning](http://semver.org/).
 
-## [6.0.1] - 2025-02-16
+## [6.0.1] - 2025-03-14
+
+### Updated
+- Updated Plotly.js from version 3.0.0 to version 3.0.1. See the [plotly.js CHANGELOG](https://github.com/plotly/plotly.js/blob/master/CHANGELOG.md#301----2025-02-18) for more information.
+
 
 ### Fixed
-- Fix `TypeError` when using `orjson` to serialize `pandas.NA`.
+- Fix `TypeError` when using `orjson` to serialize `pandas.NA` [[#5040](https://github.com/plotly/plotly.py/pull/5040)].
+- Fix issue where using `category_orders` on `px.pie` raised `ColumnNotFoundError` [[#5000](https://github.com/plotly/plotly.py/pull/5000)].
+- Fix incorrect `DeprecationWarning` shown when creating a `px` chart [[#5080](https://github.com/plotly/plotly.py/pull/5080), [#5086](https://github.com/plotly/plotly.py/pull/5086)]
+
 
 ## [6.0.0] - 2025-01-28
 

CHANGELOG.md

@@ -70,7 +70,6 @@ Publishing to PyPI:
 ### Update documentation site
 
 1. Search for the previous version string in the docs and replace it with the new version string, including but not necessarily limited to the following files:
-    - `doc/python/getting-started.md`
     - `doc/apidoc/conf.py`
     - `doc/requirements.txt`
 2. `doc-prod` should already have been merged on a regular basis into `main`, but

RELEASE.md

@@ -103,8 +103,11 @@ def build_datatype_py(node):
     )
     buffer.write("import copy as _copy\n")
 
-    if node.name_property in deprecated_mapbox_traces:
-        buffer.write(f"from warnings import warn\n")
+    if (
+        node.name_property in deprecated_mapbox_traces
+        or node.name_property == "template"
+    ):
+        buffer.write(f"import warnings\n")
 
     # Write class definition
     buffer.write(
@@ -355,10 +358,22 @@ def __init__(self"""
     buffer.write("\n\n")
     for subtype_node in subtype_nodes:
         name_prop = subtype_node.name_property
-        buffer.write(
-            f"""
+        if datatype_class == "Template" and name_prop == "data":
+            buffer.write(
+                f"""
+        # Template.data contains a 'scattermapbox' key, which causes a
+        # go.Scattermapbox trace object to be created during validation.
+        # In order to prevent false deprecation warnings from surfacing,
+        # we suppress deprecation warnings for this line only.
+        with warnings.catch_warnings():
+            warnings.filterwarnings("ignore", category=DeprecationWarning)
+            self._init_provided('{name_prop}', arg, {name_prop})"""
+            )
+        else:
+            buffer.write(
+                f"""
         self._init_provided('{name_prop}', arg, {name_prop})"""
-        )
+            )
 
     ### Literals
     if literal_nodes:
@@ -382,7 +397,7 @@ def __init__(self"""
     if node.name_property in deprecated_mapbox_traces:
         buffer.write(
             f"""
-        warn(
+        warnings.warn(
             "*{node.name_property}* is deprecated!"
             + " Use *{node.name_property.replace("mapbox", "map")}* instead."
             + " Learn more at: https://plotly.com/python/mapbox-to-maplibre/",

codegen/datatypes.py

@@ -26,7 +26,7 @@
 # The short X.Y version
 version = ""
 # The full version, including alpha/beta/rc tags
-release = "6.0.0"
+release = "6.0.1"
 
 
 # -- General configuration ---------------------------------------------------

doc/apidoc/conf.py

@@ -136,18 +136,6 @@ fig.update_traces(diagonal_visible=False)
 fig.show()
 ```
 
-## PCA analysis in Dash
-
-[Dash](https://plotly.com/dash/) is the best way to build analytical apps in Python using Plotly figures. To run the app below, run `pip install dash`, click "Download" to get the code and run `python app.py`.
-
-Get started  with [the official Dash docs](https://dash.plotly.com/installation) and **learn how to effortlessly [style](https://plotly.com/dash/design-kit/) & [deploy](https://plotly.com/dash/app-manager/) apps like this with <a class="plotly-red" href="https://plotly.com/dash/">Dash Enterprise</a>.**
-
-
-```python hide_code=true
-from IPython.display import IFrame
-snippet_url = 'https://python-docs-dash-snippets.herokuapp.com/python-docs-dash-snippets/'
-IFrame(snippet_url + 'pca-visualization', 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>

doc/python/ml-pca.md

@@ -782,6 +782,66 @@ fig.add_annotation(
 
 fig.show()
 ```
+### Specifying Source Lines or Figure Notes on the Bottom of a Figure
+
+This example shows how to add a note about the data source or interpretation at the bottom of the figure. This example aligns the note in the bottom right corner using the title element and container coordinates and then uses an annotation to add a figure title. A near zero container coordinate is an easy and robust way to put text -- such as a source line or figure note -- at the bottom of a figure. It is easier to specify the bottom of the figure in container coordinates than using paper coordinates, since uncertainty about the size of legends and x-axis labels make the paper coordinate of the bottom of the figure uncertain. Making the y container coordinate very slightly positive avoids cutting off the descending strokes of letters like y, p, and q.  Only the title command supports container coordinates, so this example re-purposes the title element to insert the note and re-purposes an annotation element for the title. The top of the figure is typically less cluttered and more predictable than the bottom of the figure, so an annotation with its bottom at a paper y-coordinate slightly greater than 1 is a reasonable title ___location on many graphs.
+
+```python
+import plotly.express as px
+df = px.data.iris()
+fig.update_layout(
+        title=dict(text="Note: this is the Plotly title element.",
+                 # keeping this title string short avoids getting the text cut off in small windows
+                 # if you need longer text, consider 1) embedding your graphic on a web page and
+                 # putting the note in the HTML to use the browser's automated word wrap,
+                 # 2) using this approach and also specifying a graph width that shows the whole title,
+                 # or 3) using <BR> tags to wrap the text onto multiple lines
+                yref="container",
+                y=0.005,
+                 # The "paper" x-coordinates lets us align this with either the right or left
+                 # edge of the plot region. 
+                 # The code to align this flush with the right edge of the plot area is 
+                 # predictable and simple.  
+                 # Putting the title in the lower left corner, aligned with the left edge of the axis labeling would
+                 # require graph specific coordinate adjustments.
+                xref="paper",
+                xanchor="right",
+                x=1, 
+                font=dict(size=12)),
+                plot_bgcolor="white",
+
+  # We move the legend out of the right margin so the right-aligned note is 
+  # flush with the right most element of the graph.
+  # Here we put the legend in a corner of the graph region
+  # because it has consistent coordinates at all screen resolutions.
+  legend=dict(
+                yanchor="top",
+                y=1,
+                xanchor="right",
+                x=1,
+                borderwidth=1)
+                )
+
+# Insert a title by repurposing an annotation 
+fig.add_annotation(
+    yref="paper",
+    yanchor="bottom",
+    y=1.025,  # y = 1 is the top of the plot area; the top is typically uncluttered, so placing 
+              # the bottom of the title slightly above the graph region works on a wide variety of graphs
+            text="This title is a Plotly annotation",
+
+    # Center the title horizontally over the plot area
+    xref="paper",
+    xanchor="center",
+    x=0.5, 
+
+    showarrow=False,
+    font=dict(size=18)
+    )
+
+fig.show()
+```
+
 
 ### Customize Displayed Text with a Text Template
 

doc/python/text-and-annotations.md

@@ -1,4 +1,4 @@
-plotly==6.0.0
+plotly==6.0.1
 jupytext==1.16.4
 jupyter-client<7
 jupyter

doc/requirements.txt

@@ -6,7 +6,7 @@
   },
   "dependencies": {
     "lodash-es": "^4.17.21",
-    "plotly.js": "3.0.0"
+    "plotly.js": "3.0.1"
   },
   "devDependencies": {
     "@types/plotly.js": "^2.33.4",