@nuasite/cms 0.12.0 → 0.12.4

package/package.json

@@ -14,7 +14,7 @@
     "directory": "packages/astro-cms"
   },
   "license": "Apache-2.0",
-  "version": "0.12.0",
+  "version": "0.12.4",
   "module": "src/index.ts",
   "types": "src/index.ts",
   "type": "module",
@@ -26,25 +26,25 @@
     }
   },
   "dependencies": {
-    "@astrojs/compiler": "^2.13.0",
-    "@babel/parser": "^7.24.0",
-    "node-html-parser": "^6.1.13",
+    "@astrojs/compiler": "^2.13.1",
+    "@babel/parser": "^7.29.0",
+    "node-html-parser": "^7.1.0",
     "yaml": "^2.8.2"
   },
   "devDependencies": {
-    "@milkdown/core": "^7.8.0",
-    "@milkdown/ctx": "^7.8.0",
-    "@milkdown/plugin-listener": "^7.8.0",
-    "@milkdown/preset-commonmark": "^7.8.0",
-    "@milkdown/preset-gfm": "^7.8.0",
-    "@milkdown/prose": "^7.8.0",
-    "@milkdown/utils": "^7.8.0",
-    "@preact/signals": "^2.6.1",
-    "@tailwindcss/vite": "^4.1.11",
-    "@types/bun": "latest",
+    "@milkdown/core": "^7.19.0",
+    "@milkdown/ctx": "^7.19.0",
+    "@milkdown/plugin-listener": "^7.19.0",
+    "@milkdown/preset-commonmark": "^7.19.0",
+    "@milkdown/preset-gfm": "^7.19.0",
+    "@milkdown/prose": "^7.19.0",
+    "@milkdown/utils": "^7.19.0",
+    "@preact/signals": "^2.8.1",
+    "@tailwindcss/vite": "^4.2.1",
+    "@types/bun": "1.3.10",
     "clsx": "^2.1.1",
-    "marked": "^17.0.1",
-    "preact": "^10.28.0",
+    "marked": "^17.0.3",
+    "preact": "^10.28.4",
     "prosemirror-commands": "^1.7.1",
     "prosemirror-inputrules": "^1.5.1",
     "prosemirror-keymap": "^1.2.3",
@@ -54,12 +54,12 @@
     "prosemirror-state": "^1.4.4",
     "prosemirror-transform": "^1.10.5",
     "prosemirror-view": "^1.41.5",
-    "tailwind-merge": "^3.4.0"
+    "tailwind-merge": "^3.5.0"
   },
   "peerDependencies": {
-    "astro": "^5.16.6",
+    "astro": "^5.17",
     "typescript": "^5",
-    "vite": "^6",
+    "vite": "^7.3.1",
     "@aws-sdk/client-s3": "^3.0.0"
   },
   "peerDependenciesMeta": {

package/src/editor/components/create-page-modal.tsx

@@ -20,7 +20,7 @@ export function CreatePageModal() {
 				openMarkdownEditorForNewPage(col?.name, col)
 			}
 		}
-	}, [visible, collections])
+	}, [collections])
 
 	const handleClose = () => {
 		resetCreatePageState()

package/src/handlers/source-writer.ts

@@ -1,5 +1,6 @@
 import fs from 'node:fs/promises'
 import path from 'node:path'
+import { NodeType, parse as parseHtml } from 'node-html-parser'
 import { getProjectRoot } from '../config'
 import type { AttributeChangePayload, ChangePayload, SaveBatchRequest } from '../editor/types'
 import type { ManifestWriter } from '../manifest-writer'
@@ -532,16 +533,11 @@ export function applyTextChange(
 	const updatedSnippet = sourceSnippet.replace(resolvedOriginal, resolvedNewText)
 
 	if (updatedSnippet === sourceSnippet) {
-		// Try with <br> tag normalization (browser normalizes <br /> to <br>)
-		const brNorm = (s: string) => s.replace(/<br\s*\/?>/gi, '<br>')
-		if (brNorm(sourceSnippet).includes(brNorm(resolvedOriginal))) {
-			const parts = resolvedOriginal.split(/<br\s*\/?>/gi)
-			const regexStr = parts.map((p) => escapeRegExp(p)).join('<br\\s*\\/?>')
-			const brRegex = new RegExp(regexStr)
-			const updatedWithBr = sourceSnippet.replace(brRegex, resolvedNewText)
-			if (updatedWithBr !== sourceSnippet) {
-				return { success: true, content: content.replace(sourceSnippet, updatedWithBr) }
-			}
+		// Try AST-based <br> normalization (browser normalizes <br class="..." /> to <br>
+		// and collapses surrounding whitespace/indentation)
+		const brResult = tryBrNormalizedChange(sourceSnippet, resolvedOriginal, resolvedNewText)
+		if (brResult !== null) {
+			return { success: true, content: content.replace(sourceSnippet, brResult) }
 		}
 
 		// resolvedOriginal wasn't found in snippet - try HTML entity handling
@@ -657,7 +653,7 @@ function findTextInSnippet(snippet: string, decodedText: string): string | null
 
 	// Try matching with <br> tags stripped from snippet
 	const chars = [...decodedText].map((ch) => escapeRegExp(ch))
-	const brAwarePattern = chars.join('(?:<br\\s*\\/?>)*')
+	const brAwarePattern = chars.join('(?:<br\\b[^>]*\\/?>)*')
 	const brRegex = new RegExp(brAwarePattern)
 	const brMatch = snippet.match(brRegex)
 
@@ -737,6 +733,110 @@ function findExpressionSrcAttribute(text: string): { index: number; length: numb
 	}
 }
 
+/**
+ * Extract visible text from an HTML string the way a browser would render it.
+ * Text nodes contribute their content, <br> elements become '\n',
+ * and whitespace around '\n' is collapsed (matching browser behavior).
+ */
+function getVisibleText(html: string): string {
+	const root = parseHtml(html, { blockTextElements: {} })
+	let text = ''
+	const walk = (node: ReturnType<typeof parseHtml>) => {
+		for (const child of node.childNodes) {
+			if (child.nodeType === NodeType.TEXT_NODE) {
+				text += child.rawText
+			} else if (child.nodeType === NodeType.ELEMENT_NODE && (child as any).rawTagName === 'br') {
+				text += '\n'
+			} else {
+				walk(child as any)
+			}
+		}
+	}
+	walk(root)
+	// Collapse whitespace around newlines (browser behavior around <br>)
+	text = text.replace(/[ \t]*\n[ \t]*/g, '\n')
+	return text.trim()
+}
+
+/**
+ * Try to apply a text change when the mismatch is due to <br> normalization.
+ * The browser normalizes <br class="..." /> to plain <br> and collapses surrounding whitespace.
+ * This function preserves the original <br> elements (with attributes) and surrounding indentation.
+ * Returns the updated snippet, or null if this approach doesn't apply.
+ */
+function tryBrNormalizedChange(
+	sourceSnippet: string,
+	resolvedOriginal: string,
+	resolvedNewText: string,
+): string | null {
+	// Only applies when the browser text contains <br>
+	if (!resolvedOriginal.includes('<br>')) return null
+
+	// Verify that the visible text matches after normalization
+	const sourceVisible = getVisibleText(sourceSnippet)
+	const originalVisible = getVisibleText(resolvedOriginal)
+	if (sourceVisible !== originalVisible) return null
+
+	// Split browser text by <br> into segments
+	const originalSegments = resolvedOriginal.split('<br>')
+	const newSegments = resolvedNewText.split('<br>')
+
+	// If segment count changed, user added/removed line breaks — let other fallbacks handle it
+	if (originalSegments.length !== newSegments.length) return null
+
+	// Parse the source snippet and identify text nodes and br elements
+	const root = parseHtml(sourceSnippet, { blockTextElements: {} })
+
+	// Find the outer element (e.g., <h1>, <p>)
+	const outerElement = root.childNodes.find(
+		(n) => n.nodeType === NodeType.ELEMENT_NODE,
+	) as any
+	if (!outerElement) return null
+
+	// Collect text nodes between br boundaries
+	const groups: Array<Array<{ node: any; index: number }>> = [[]]
+	for (let i = 0; i < outerElement.childNodes.length; i++) {
+		const child = outerElement.childNodes[i]
+		if (child.nodeType === NodeType.ELEMENT_NODE && (child as any).rawTagName === 'br') {
+			groups.push([])
+		} else if (child.nodeType === NodeType.TEXT_NODE) {
+			groups[groups.length - 1]!.push({ node: child, index: i })
+		}
+	}
+
+	// Number of groups should match number of segments
+	if (groups.length !== originalSegments.length) return null
+
+	// Replace text content in each group
+	let result = sourceSnippet
+	for (let g = groups.length - 1; g >= 0; g--) {
+		const group = groups[g]!
+		const origSegment = originalSegments[g]!.trim()
+		const newSegment = newSegments[g]!.trim()
+
+		if (origSegment === newSegment) continue
+
+		// Find the text node in this group that contains the meaningful text
+		for (const { node } of group) {
+			const raw: string = node.rawText
+			const trimmed = raw.trim()
+			if (!trimmed) continue
+
+			// Check if this text node's trimmed content matches the original segment
+			if (trimmed === origSegment) {
+				// Replace the meaningful text, preserving surrounding whitespace
+				const leadingWs = raw.slice(0, raw.indexOf(trimmed))
+				const trailingWs = raw.slice(raw.indexOf(trimmed) + trimmed.length)
+				const newRaw = leadingWs + newSegment + trailingWs
+				result = result.replace(raw, newRaw)
+				break
+			}
+		}
+	}
+
+	return result !== sourceSnippet ? result : null
+}
+
 function escapeRegExp(string: string): string {
 	return string.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')
 }