Merge pull request #9425 from quarto-dev/bugfix/9422

jupyter - don't merge streams across other output cells

Author: cscheid

news/changelog-1.5.md

@@ -70,6 +70,7 @@ All changes included in 1.5:
 - ([#8998](https://github.com/quarto-dev/quarto-cli/issues/8998)): Interpret slide separation markers `---` correctly when creating the `.ipynb` intermediate notebook from a `.qmd` file.
 - ([#9133](https://github.com/quarto-dev/quarto-cli/issues/9133)): Fix issue with Jupyter engine when using paths containing special characters.
 - ([#9255](https://github.com/quarto-dev/quarto-cli/issues/9255)): Support cell source fields of type `string`.
+- ([#9422](https://github.com/quarto-dev/quarto-cli/issues/9422)): Improve the stream merging algorithm in output cells to avoid merging outputs that should not be merged.
 
 ## Website Listings
 

src/core/jupyter/jupyter-fixups.ts

@@ -23,32 +23,26 @@ export function fixupStreams(nb: JupyterNotebook): JupyterNotebook {
     if (cell.cell_type !== "code" || cell.outputs === undefined) {
       continue;
     }
-    let i = 0;
     if (cell.outputs.length === 0) {
       continue;
     }
+    const newOutputs: JupyterOutput[] = [cell.outputs[0]];
+    let i = 1;
     while (i < cell.outputs.length) {
+      const prevOutput: JupyterOutput = newOutputs[newOutputs.length - 1];
       const thisOutput: JupyterOutput = cell.outputs[i];
-      if (thisOutput.output_type === "stream") {
-        // collect all the stream outputs with the same name
-        const streams = cell.outputs.filter((output) =>
-          output.output_type === "stream" && output.name === thisOutput.name
-        );
-        // join them together
-        const joinedText = streams.map((output) => output.text ?? []).flat();
-        const newOutput: JupyterOutput = {
-          output_type: "stream",
-          name: thisOutput.name,
-          text: joinedText,
-        };
-        cell.outputs[i] = newOutput;
-        cell.outputs = cell.outputs.filter((output, j) =>
-          i === j ||
-          (output.output_type !== "stream" || output.name !== thisOutput.name)
-        );
+      if (
+        thisOutput.output_type === "stream" &&
+        prevOutput.output_type === "stream" &&
+        thisOutput.name === prevOutput.name
+      ) {
+        prevOutput.text = [...prevOutput.text ?? [], ...thisOutput.text ?? []];
+      } else {
+        newOutputs.push(thisOutput);
       }
       i++;
     }
+    cell.outputs = newOutputs;
   }
   return nb;
 }

src/core/jupyter/jupyter.ts

@@ -1610,7 +1610,13 @@ async function mdFromCodeCell(
       if (output.output_type === "stream") {
         const stream = output as JupyterOutputStream;
         if (asis && stream.name === "stdout") {
-          md.push(stream.text.join(""));
+          let text: string[] = [];
+          if (typeof stream.text === "string") {
+            text = [stream.text];
+          } else {
+            text = stream.text;
+          }
+          md.push(text.join(""));
         } else {
           md.push(mdOutputStream(stream));
         }
@@ -1750,21 +1756,28 @@ function isMarkdown(output: JupyterOutput, options: JupyterToMarkdownOptions) {
 }
 
 function mdOutputStream(output: JupyterOutputStream) {
+  let text: string[] = [];
+  if (typeof output.text === "string") {
+    text = [output.text];
+  } else {
+    text = output.text;
+  }
+
   // trim off warning source line for notebook
   if (output.name === "stderr") {
-    if (output.text[0]) {
-      const firstLine = output.text[0].replace(
+    if (text[0]) {
+      const firstLine = text[0].replace(
         /<ipython-input.*?>:\d+:\s+/,
         "",
       );
       return mdCodeOutput(
-        [firstLine, ...output.text.slice(1)].map(colors.stripColor),
+        [firstLine, ...text.slice(1)].map(colors.stripColor),
       );
     }
   }
 
   // normal default handling
-  return mdCodeOutput(output.text.map(colors.stripColor));
+  return mdCodeOutput(text.map(colors.stripColor));
 }
 
 async function mdOutputError(

src/core/jupyter/types.ts

@@ -166,12 +166,12 @@ export interface JupyterOutput {
   metadata?: Record<string, unknown>;
   data?: Record<string, unknown>;
   name?: string;
-  text?: string[];
+  text?: string[] | string;
 }
 
 export interface JupyterOutputStream extends JupyterOutput {
   name: "stdout" | "stderr";
-  text: string[];
+  text: string[] | string;
 }
 
 export interface JupyterOutputDisplayData extends JupyterOutput {

tests/docs/smoke-all/2024/04/19/9422.qmd

@@ -0,0 +1,47 @@
+---
+title: "Generate Plots within Sub-Sections"
+format:
+  html:
+    echo: false
+#     embed-resources: true
+engine: jupyter
+_quarto:
+  tests:
+    html:
+      ensureHtmlElements:
+        - ["section#scatter img"]
+---
+
+```{python}
+#| label: setup
+import matplotlib.pyplot as plt
+```
+
+```{python}
+#| label: define-data
+
+data = {'apple': 10, 'orange': 15, 'lemon': 5, 'lime': 20}
+names = list(data.keys())
+values = list(data.values())
+```
+
+## Plots and Subsections
+
+```{python}
+#| output: asis
+
+print("\n### Bar\n")
+
+plt.bar(names, values)
+plt.show(block=False)
+
+print("\n### Scatter\n")
+
+plt.scatter(names, values)
+plt.show(block=False)
+
+print("\n### Line\n")
+
+plt.plot(names, values)
+plt.show(block=False)
+```