Implement step 7.3.1: Parse text box content

Author: davidnmbond

PanoramicData.Render.Test/DrawingShapeRunElementTests.cs

@@ -136,6 +136,36 @@ public void Parse_AnchorDrawingWithDiamondShape_ReturnsDrawingShapeRunElement()
 		shape.PresetKind.ToString().Should().Be("Diamond");
 	}
 
+	[Fact]
+	public void Parse_InlineDrawingWithDrawingMlTextFrame_ExtractsText()
+	{
+		var drawing = CreateInlineShapeWithTextBody("rect", 914400L, 457200L, ["First line", "Second line"]);
+		var run = new Run(drawing);
+
+		var elements = RunElementParser.Parse(run);
+
+		var shape = elements.Should().ContainSingle()
+			.Which.Should().BeOfType<DrawingShapeRunElement>()
+			.Subject;
+		shape.TextFrame.HasTextFrame.Should().BeTrue();
+		shape.TextFrame.Text.Should().Be("First line\nSecond line");
+	}
+
+	[Fact]
+	public void Parse_InlineDrawingWithWordTextBoxContent_ExtractsText()
+	{
+		var drawing = CreateInlineShapeWithTextBoxContent("rect", 914400L, 457200L, ["Box line one", "Box line two"]);
+		var run = new Run(drawing);
+
+		var elements = RunElementParser.Parse(run);
+
+		var shape = elements.Should().ContainSingle()
+			.Which.Should().BeOfType<DrawingShapeRunElement>()
+			.Subject;
+		shape.TextFrame.HasTextFrame.Should().BeTrue();
+		shape.TextFrame.Text.Should().Be("Box line one\nBox line two");
+	}
+
 	// -------------------------------------------------------------------------
 	// Helpers
 	// -------------------------------------------------------------------------
@@ -204,4 +234,59 @@ private static Drawing CreateAnchorShape(string presetName, long widthEmu, long
 		};
 		return new Drawing(anchor);
 	}
+
+	private static Drawing CreateInlineShapeWithTextBody(string presetName, long widthEmu, long heightEmu, IReadOnlyList<string> lines)
+	{
+		var presetGeom = new A.PresetGeometry();
+		presetGeom.SetAttribute(new OpenXmlAttribute("prst", string.Empty, presetName));
+		var textBody = new OpenXmlUnknownElement("txBody")
+		{
+			InnerXml = "<bodyPr/>" + string.Concat(lines.Select(line => $"<p><r><t>{System.Security.SecurityElement.Escape(line)}</t></r></p>"))
+		};
+		var spPr = new A.ShapeProperties(presetGeom);
+		var graphicData = new A.GraphicData(spPr, textBody)
+		{
+			Uri = "http://schemas.openxmlformats.org/drawingml/2006/main"
+		};
+		var graphic = new A.Graphic(graphicData);
+		var inline = new DW.Inline(
+			new DW.Extent { Cx = widthEmu, Cy = heightEmu },
+			graphic)
+		{
+			DistanceFromTop = 0,
+			DistanceFromBottom = 0,
+			DistanceFromLeft = 0,
+			DistanceFromRight = 0
+		};
+		return new Drawing(inline);
+	}
+
+	private static Drawing CreateInlineShapeWithTextBoxContent(string presetName, long widthEmu, long heightEmu, IReadOnlyList<string> lines)
+	{
+		var presetGeom = new A.PresetGeometry();
+		presetGeom.SetAttribute(new OpenXmlAttribute("prst", string.Empty, presetName));
+		var textBox = new OpenXmlUnknownElement("wps:txbx")
+		{
+			InnerXml = "<w:txbxContent xmlns:w=\"http://schemas.openxmlformats.org/wordprocessingml/2006/main\">"
+				+ string.Concat(lines.Select(line => $"<w:p><w:r><w:t>{System.Security.SecurityElement.Escape(line)}</w:t></w:r></w:p>"))
+				+ "</w:txbxContent>"
+		};
+		var spPr = new A.ShapeProperties(presetGeom);
+		var graphicData = new A.GraphicData(spPr, textBox)
+		{
+			Uri = "http://schemas.openxmlformats.org/drawingml/2006/main"
+		};
+		var graphic = new A.Graphic(graphicData);
+		var inline = new DW.Inline(
+			new DW.Extent { Cx = widthEmu, Cy = heightEmu },
+			graphic)
+		{
+			DistanceFromTop = 0,
+			DistanceFromBottom = 0,
+			DistanceFromLeft = 0,
+			DistanceFromRight = 0
+		};
+		return new Drawing(inline);
+	}
+
 }

PanoramicData.Render/ShapeTextFrameParser.cs

@@ -16,6 +16,22 @@ public static ShapeTextFrameInfo Parse(OpenXmlElement drawingRoot)
 	{
 		ArgumentNullException.ThrowIfNull(drawingRoot);
 
+		var textBoxContent = drawingRoot.Descendants().FirstOrDefault(e => e.LocalName == "txbxContent");
+		if (textBoxContent is not null)
+		{
+			var paragraphLines = textBoxContent.ChildElements
+				.Where(e => e.LocalName == "p")
+				.Select(ExtractText)
+				.Where(text => text.Length > 0)
+				.ToList();
+
+			return new ShapeTextFrameInfo
+			{
+				HasTextFrame = true,
+				Text = string.Join("\n", paragraphLines)
+			};
+		}
+
 		var txBody = drawingRoot.Descendants().FirstOrDefault(e => e.LocalName == "txBody");
 		if (txBody is null)
 		{
@@ -24,15 +40,7 @@ public static ShapeTextFrameInfo Parse(OpenXmlElement drawingRoot)
 
 		var bodyPr = txBody.ChildElements.FirstOrDefault(e => e.LocalName == "bodyPr");
 		var paragraphs = txBody.ChildElements.Where(e => e.LocalName == "p").ToList();
-		var lines = new List<string>();
-		for (var i = 0; i < paragraphs.Count; i++)
-		{
-			var text = string.Concat(paragraphs[i].Descendants().Where(d => d.LocalName == "t").Select(t => t.InnerText));
-			if (text.Length > 0)
-			{
-				lines.Add(text);
-			}
-		}
+		var lines = paragraphs.Select(ExtractText).Where(text => text.Length > 0).ToList();
 
 		var autoFitMode = ShapeTextAutoFitMode.None;
 		if (bodyPr is not null)
@@ -63,6 +71,11 @@ public static ShapeTextFrameInfo Parse(OpenXmlElement drawingRoot)
 		};
 	}
 
+	private static string ExtractText(OpenXmlElement paragraphLikeElement)
+	{
+		return string.Concat(paragraphLikeElement.Descendants().Where(d => d.LocalName == "t").Select(t => t.InnerText));
+	}
+
 	private static long ParseLongAttribute(OpenXmlElement? element, string localName)
 	{
 		if (element is null)

docs/plans/current-status.md

@@ -10,26 +10,22 @@ Phase 7: Advanced Features — **IN PROGRESS**
 
 ## Current Step
 
-Steps 7.2.1–7.2.8 — multi-level lists — **COMPLETE**
+Step 7.3.1 — text box parsing — **COMPLETE**
 
-Completed multi-level list support milestone:
-- Added restart metadata (`RestartAfterLevel`) to resolved numbering level styles
-- Extended `NumberingStyleResolver` to map `w:lvlRestart`
-- Implemented `ListNumberingFormatter` for decimal/alpha/roman/bullet label text and `%n` pattern expansion
-- Implemented `ListNumberingState` for per-instance counters and restart-aware sequence tracking
-- Integrated list label emission into `RenderCommandEmitter` with hanging-indent style positioning
-- Added support for bullet label fonts via resolved/configured numbering style font family
-- Added unit tests for numbering sequences, restart behavior, label positioning, and bullet font rendering
+Completed initial text box parsing support:
+- Extended `ShapeTextFrameParser` to parse Wordprocessing text boxes via `w:txbxContent`
+- Preserved existing DrawingML `txBody` parsing while preferring explicit text box content when present
+- Added parser coverage for inline shapes containing DrawingML text bodies and Wordprocessing text box content
 
-1748 tests passing (1743 → 1748, +5 list integration tests).
+1750 tests passing (1748 → 1750, +2 text box parsing tests).
 
 ## Next Step
 
-Steps 7.3.1–7.3.7 — text boxes parsing, layout, and positioning
+Steps 7.3.2–7.3.7 — text box layout, margins, auto-size, and positioning
 
 ## Last Commit
 
-Implement steps 7.2.5-7.2.8: List render integration and positioning (commit 9a4d948)
+Implement step 7.3.1: Parse text box content (commit pending)
 
 ## Implementation Notes
 

docs/plans/phase-7-advanced-features.md

@@ -35,7 +35,7 @@ Implement the remaining document features needed for high-fidelity rendering of
 
 ### 7.3 Text Boxes
 
-- [ ] 7.3.1 — Parse text box elements (`w:txbxContent` inside `wsp:txbx` or VML `v:textbox`)
+- [x] 7.3.1 — Parse text box elements (`w:txbxContent` inside `wsp:txbx` or VML `v:textbox`)
 - [ ] 7.3.2 — Lay out text box content using the text layout engine (text boxes can contain paragraphs, tables, images)
 - [ ] 7.3.3 — Position text box as a floating object with anchor and wrapping
 - [ ] 7.3.4 — Handle text box internal margins