[FIX] xlsx: import new lines in sharedStrings

When implementing the multiline text feature, we forgot to change the
xlsx import, that explicitly removed new lines from the sharedStrings
because we didn't support it at the time.

There was also a bug where the whitespace was not preserved in the
sharedStrings, because we extracted the text from the `<si>` element
instead of the `<t>` element that had the attribute xml:space="preserve".

And for multi-line cells to be considered as such by Excel, they need
to have the style `wrapText="1"`.

Task: 4044862
Part-of: #5408
Signed-off-by: Rémi Rahir (rar) <[email protected]>
Signed-off-by: Adrien Minne (adrm) <[email protected]>

Author: hokolomopo

src/xlsx/conversion/sheet_conversion.ts

@@ -2,6 +2,7 @@ import {
   buildSheetLink,
   largeMax,
   markdownLink,
+  replaceNewLines,
   splitReference,
   toCartesian,
   toXC,
@@ -120,9 +121,8 @@ function convertRows(
   return rows;
 }
 
-/** Remove newlines (\n) in shared strings, We do not support them */
 function convertSharedStrings(xlsxSharedStrings: string[]): string[] {
-  return xlsxSharedStrings.map((str) => str.replace(/\n/g, ""));
+  return xlsxSharedStrings.map(replaceNewLines);
 }
 
 function convertCells(

src/xlsx/extraction/misc_extractor.ts

@@ -39,15 +39,10 @@ export class XlsxMiscExtractor extends XlsxBaseExtractor {
       { parent: this.rootFile.file.xml, query: "si" },
       (ssElement): string => {
         // Shared string can either be a simple text, or a rich text (text with formatting, possibly in multiple parts)
-        if (ssElement.children[0].tagName === "t") {
-          return this.extractTextContent(ssElement) || "";
-        }
         // We don't support rich text formatting, we'll only extract the text
-        else {
-          return this.mapOnElements({ parent: ssElement, query: "t" }, (textElement): string => {
-            return this.extractTextContent(textElement) || "";
-          }).join("");
-        }
+        return this.mapOnElements({ parent: ssElement, query: "t" }, (textElement): string => {
+          return this.extractTextContent(textElement) || "";
+        }).join("");
       }
     );
   }

src/xlsx/functions/worksheet.ts

@@ -97,7 +97,10 @@ export function addRows(
         const attributes: XMLAttributes = [["r", xc]];
 
         // style
-        const id = normalizeStyle(construct, extractStyle(data, styleId, formatId, borderId));
+        const id = normalizeStyle(
+          construct,
+          extractStyle(data, content, styleId, formatId, borderId)
+        );
         // don't add style if default
         if (id) {
           attributes.push(["s", id]);

src/xlsx/helpers/content_helpers.ts

@@ -1,4 +1,4 @@
-import { DEFAULT_FONT_SIZE } from "../../constants";
+import { DEFAULT_FONT_SIZE, NEWLINE } from "../../constants";
 import { deepEquals, splitReference, toUnboundedZone } from "../../helpers";
 import {
   ConditionalFormattingOperatorValues,
@@ -91,6 +91,7 @@ export function convertWidthFromExcel(width: number | undefined): number | undef
 
 export function extractStyle(
   data: WorkbookData,
+  content: string | undefined,
   styleId: number | undefined,
   formatId: number | undefined,
   borderId: number | undefined
@@ -116,7 +117,7 @@ export function extractStyle(
       vertical: style.verticalAlign
         ? V_ALIGNMENT_EXPORT_CONVERSION_MAP[style.verticalAlign]
         : undefined,
-      wrapText: style.wrapping === "wrap" || undefined,
+      wrapText: style.wrapping === "wrap" || content?.includes(NEWLINE) ? true : undefined,
     },
   };
 

tests/__xlsx__/xlsx_demo_data.xlsx

@@ -832,6 +832,26 @@ describe("Test XLSX export", () => {
     });
   });
 
+  test("Multi-line cells are exported with text wrap", async () => {
+    const model = new Model();
+    setCellContent(model, "A1", "This is a\nmultiline cell");
+
+    const exportedXlsx = await exportPrettifiedXlsx(model);
+    const styleSheet = parseXML(
+      exportedXlsx.files.find((f) => f["contentType"] === "styles")!["content"]
+    );
+    const workSheet = parseXML(
+      exportedXlsx.files.find((f) => f["contentType"] === "sheet")!["content"]
+    );
+
+    const A1 = workSheet.querySelector("c[r='A1']")!;
+    expect(A1.getAttribute("s")).toBe("1");
+
+    const styles = styleSheet.querySelectorAll("xf");
+    expect(styles[1].getAttribute("applyAlignment")).toBe("1");
+    expect(styles[1].querySelector("alignment")!.getAttribute("wrapText")).toBe("1");
+  });
+
   describe("formulas", () => {
     beforeAll(() => {
       functionRegistry.add("NOW", {

tests/xlsx/xlsx_export.test.ts

@@ -865,10 +865,16 @@ describe("Import xlsx data", () => {
   });
 
   describe("Misc tests", () => {
-    test("Newlines characters in strings are removed", () => {
+    test("Newlines characters in strings are preserved", () => {
       const testSheet = getWorkbookSheet("jestMiscTest", convertedData)!;
       const textWithNewLineInXLSX = testSheet.cells["A1"];
-      expect(textWithNewLineInXLSX).toEqual("This text have a newLine"); // newline should have been removed at import
+      expect(textWithNewLineInXLSX).toEqual("This text have\n a newLine");
+    });
+
+    test("White spaces are preserved", () => {
+      const testSheet = getWorkbookSheet("jestMiscTest", convertedData)!;
+      const textWithNewLineInXLSX = testSheet.cells["B1"];
+      expect(textWithNewLineInXLSX).toEqual("      with whitespace before");
     });
 
     test("Can hide gridLines", () => {