mirror of
https://github.com/arsvendg/Stirling-PDF.git
synced 2026-07-16 19:33:11 +02:00
fix(markdown): markdown conversion image handling and zip support (#5677)
This commit is contained in:
Balázs Szücs
@@ -107,56 +107,65 @@ public class PDFToFile {
|
||||
File[] outputFiles =
|
||||
Objects.requireNonNull(tempOutputDir.getPath().toFile().listFiles());
|
||||
List<File> markdownFiles = new ArrayList<>();
|
||||
List<File> imageFiles = new ArrayList<>();
|
||||
|
||||
// Convert HTML files to Markdown
|
||||
// Convert HTML files to Markdown and collect image files
|
||||
for (File outputFile : outputFiles) {
|
||||
if (outputFile.getName().endsWith(".html")) {
|
||||
String html = Files.readString(outputFile.toPath());
|
||||
String markdown = htmlToMarkdownConverter.convert(html);
|
||||
|
||||
// Update image references to point to images/ folder
|
||||
markdown = updateImageReferences(markdown);
|
||||
|
||||
String mdFileName = outputFile.getName().replace(".html", ".md");
|
||||
File mdFile = new File(tempOutputDir.getPath().toFile(), mdFileName);
|
||||
Files.writeString(mdFile.toPath(), markdown);
|
||||
markdownFiles.add(mdFile);
|
||||
} else if (!outputFile.getName().endsWith(".md")) {
|
||||
// Collect non-HTML, non-MD files as images/assets
|
||||
imageFiles.add(outputFile);
|
||||
}
|
||||
}
|
||||
|
||||
// If there's only one markdown file, return it directly
|
||||
if (markdownFiles.size() == 1) {
|
||||
fileName = pdfBaseName + ".md";
|
||||
fileBytes = Files.readAllBytes(markdownFiles.get(0).toPath());
|
||||
} else {
|
||||
// Multiple files - create a zip
|
||||
fileName = pdfBaseName + "ToMarkdown.zip";
|
||||
ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
|
||||
// Always create a ZIP file
|
||||
fileName = pdfBaseName + "ToMarkdown.zip";
|
||||
ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
|
||||
|
||||
try (ZipOutputStream zipOutputStream = new ZipOutputStream(byteArrayOutputStream)) {
|
||||
// Add markdown files
|
||||
for (File mdFile : markdownFiles) {
|
||||
ZipEntry mdEntry = new ZipEntry(mdFile.getName());
|
||||
zipOutputStream.putNextEntry(mdEntry);
|
||||
Files.copy(mdFile.toPath(), zipOutputStream);
|
||||
zipOutputStream.closeEntry();
|
||||
}
|
||||
|
||||
// Add images and other assets
|
||||
for (File file : outputFiles) {
|
||||
if (!file.getName().endsWith(".html") && !file.getName().endsWith(".md")) {
|
||||
ZipEntry assetEntry = new ZipEntry(file.getName());
|
||||
zipOutputStream.putNextEntry(assetEntry);
|
||||
Files.copy(file.toPath(), zipOutputStream);
|
||||
zipOutputStream.closeEntry();
|
||||
}
|
||||
}
|
||||
try (ZipOutputStream zipOutputStream = new ZipOutputStream(byteArrayOutputStream)) {
|
||||
// Add markdown files to root of ZIP
|
||||
for (File mdFile : markdownFiles) {
|
||||
ZipEntry mdEntry = new ZipEntry(mdFile.getName());
|
||||
zipOutputStream.putNextEntry(mdEntry);
|
||||
Files.copy(mdFile.toPath(), zipOutputStream);
|
||||
zipOutputStream.closeEntry();
|
||||
}
|
||||
|
||||
fileBytes = byteArrayOutputStream.toByteArray();
|
||||
// Add images and other assets to images/ folder
|
||||
for (File imageFile : imageFiles) {
|
||||
ZipEntry assetEntry = new ZipEntry("images/" + imageFile.getName());
|
||||
zipOutputStream.putNextEntry(assetEntry);
|
||||
Files.copy(imageFile.toPath(), zipOutputStream);
|
||||
zipOutputStream.closeEntry();
|
||||
}
|
||||
}
|
||||
|
||||
fileBytes = byteArrayOutputStream.toByteArray();
|
||||
}
|
||||
return WebResponseUtils.bytesToWebResponse(
|
||||
fileBytes, fileName, MediaType.APPLICATION_OCTET_STREAM);
|
||||
}
|
||||
|
||||
/**
|
||||
* Updates image references in markdown to point to the images/ folder. Matches patterns like
|
||||
*  and converts to 
|
||||
*/
|
||||
private String updateImageReferences(String markdown) {
|
||||
// Match markdown image syntax: 
|
||||
// Only update if the path doesn't already start with images/
|
||||
return markdown.replaceAll("(!\\[.*?\\])\\((?!images/)([^/)][^)]*?)\\)", "$1(images/$2)");
|
||||
}
|
||||
|
||||
public ResponseEntity<byte[]> processPdfToHtml(MultipartFile inputFile)
|
||||
throws IOException, InterruptedException {
|
||||
if (!MediaType.APPLICATION_PDF_VALUE.equals(inputFile.getContentType())) {
|
||||
|
||||
@@ -153,11 +153,12 @@ class PDFToFileTest {
|
||||
MediaType.APPLICATION_PDF_VALUE,
|
||||
"Fake PDF content".getBytes());
|
||||
|
||||
// Create a mock HTML output file
|
||||
// Create a mock HTML output file with image references
|
||||
Path htmlOutputFile = tempDir.resolve("test.html");
|
||||
Files.write(
|
||||
htmlOutputFile,
|
||||
"<html><body><h1>Test</h1><p>This is a test.</p></body></html>".getBytes());
|
||||
"<html><body><h1>Test</h1><p>This is a test.</p><img src=\"image1.png\" /></body></html>"
|
||||
.getBytes());
|
||||
|
||||
// Setup ProcessExecutor mock
|
||||
mockedStaticProcessExecutor
|
||||
@@ -174,18 +175,61 @@ class PDFToFileTest {
|
||||
Files.copy(
|
||||
htmlOutputFile, Path.of(outputDir.getPath(), "test.html"));
|
||||
|
||||
// Create a mock image file
|
||||
Files.write(
|
||||
Path.of(outputDir.getPath(), "image1.png"),
|
||||
"Fake image data".getBytes());
|
||||
|
||||
return mockExecutorResult;
|
||||
});
|
||||
|
||||
// Execute the method
|
||||
ResponseEntity<byte[]> response = pdfToFile.processPdfToMarkdown(pdfFile);
|
||||
|
||||
// Verify
|
||||
// Verify - should now return a ZIP file instead of plain markdown
|
||||
assertEquals(HttpStatus.OK, response.getStatusCode());
|
||||
assertNotNull(response.getBody());
|
||||
assertTrue(response.getBody().length > 0);
|
||||
|
||||
// Verify content disposition indicates a ZIP file
|
||||
assertTrue(
|
||||
response.getHeaders().getContentDisposition().toString().contains("test.md"));
|
||||
response.getHeaders()
|
||||
.getContentDisposition()
|
||||
.toString()
|
||||
.contains("ToMarkdown.zip"));
|
||||
|
||||
// Verify the content by unzipping it
|
||||
try (ZipInputStream zipStream =
|
||||
ZipSecurity.createHardenedInputStream(
|
||||
new java.io.ByteArrayInputStream(response.getBody()))) {
|
||||
ZipEntry entry;
|
||||
boolean foundMdFile = false;
|
||||
boolean foundImageInFolder = false;
|
||||
String markdownContent = null;
|
||||
|
||||
while ((entry = zipStream.getNextEntry()) != null) {
|
||||
if (entry.getName().endsWith(".md")) {
|
||||
foundMdFile = true;
|
||||
// Read markdown content to verify image references
|
||||
markdownContent =
|
||||
new String(
|
||||
zipStream.readAllBytes(),
|
||||
java.nio.charset.StandardCharsets.UTF_8);
|
||||
} else if (entry.getName().startsWith("images/")
|
||||
&& entry.getName().endsWith(".png")) {
|
||||
foundImageInFolder = true;
|
||||
}
|
||||
zipStream.closeEntry();
|
||||
}
|
||||
|
||||
assertTrue(foundMdFile, "ZIP should contain Markdown file");
|
||||
assertTrue(foundImageInFolder, "ZIP should contain image in images/ folder");
|
||||
assertNotNull(markdownContent, "Markdown content should be present");
|
||||
// Verify markdown references images with images/ prefix
|
||||
assertTrue(
|
||||
markdownContent.contains("images/"),
|
||||
"Markdown should reference images with images/ prefix");
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -256,14 +300,15 @@ class PDFToFileTest {
|
||||
while ((entry = zipStream.getNextEntry()) != null) {
|
||||
if (entry.getName().endsWith(".md")) {
|
||||
foundMdFiles = true;
|
||||
} else if (entry.getName().endsWith(".png")) {
|
||||
} else if (entry.getName().startsWith("images/")
|
||||
&& entry.getName().endsWith(".png")) {
|
||||
foundImage = true;
|
||||
}
|
||||
zipStream.closeEntry();
|
||||
}
|
||||
|
||||
assertTrue(foundMdFiles, "ZIP should contain Markdown files");
|
||||
assertTrue(foundImage, "ZIP should contain image files");
|
||||
assertTrue(foundImage, "ZIP should contain image files in images/ folder");
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user