fix: ensure automatically generated filename is a valid filename (#27)

* if filename is not valid replace unvalid charachter with -

* fix typo

* better file name for EmbedImages code by @fmartingr

* update unittests

* be sure filename came from EmbedImages not be duplicated

* update unittest

* add failover for detect extention

* add test for file without extention and fix other test

epub.go

@@ -34,7 +34,9 @@ import (
 	"fmt"
 	"io/fs"
 	"log"
+	"mime"
 	"net/http"
+	"net/url"
 	"os"
 	"path"
 	"path/filepath"
@@ -601,8 +603,28 @@ func (e *Epub) EmbedImages() {
 
 				firstSrcIndex := strings.Index(match[0], " src=")
 				match[0] = match[0][:firstSrcIndex+len(" src=")] + strings.ReplaceAll(match[0][firstSrcIndex+len(" src="):], " src=", " data-src=")
-
-				filePath, err := e.AddImage(string(imageURL), "")
+				parsedImageURL, err := url.Parse(imageURL)
+				if err != nil {
+					log.Printf("can't parse image URL: %s", err)
+					continue
+				}
+				extension := filepath.Ext(parsedImageURL.Path)
+				if extension == "" {
+					res, err := http.Head(imageURL)
+					if err != nil {
+						log.Printf("can't get image headers: %s", err)
+					} else {
+						// Get extension from the file content type
+						extensions, err := mime.ExtensionsByType(res.Header.Get("Content-Type"))
+						if err != nil {
+							log.Printf("can't get file type from content type: %s", err)
+						} else if len(extensions) > 0 {
+							extension = extensions[0]
+						}
+					}
+				}
+				filename := fmt.Sprintf("image%04d%s", len(e.images)+1, extension)
+				filePath, err := e.AddImage(string(imageURL), filename)
 				if err != nil {
 					log.Printf("can't add image to the epub: %s", err)
 					continue

epub_test.go

@@ -1015,9 +1015,21 @@ func TestEmbedImage(t *testing.T) {
 	testSectionBodyWithImage := `    <h1>Section 1</h1>
 	<p>This is a paragraph.</p>
 	<p><img src="` + server.URL + `/gophercolor16x16.png" loading="lazy"/></p>`
+	testSectionBodyWithImageUnvalidname := `    <h1>Section 1</h1>
+	<p>This is a paragraph.</p>
+	<p><img src="` + server.URL + `/gophercolor16x16.png?width=300" loading="lazy"/></p>`
+	testSectionBodyWithImageWithoutExtention := `    <h1>Section 1</h1>
+	<p>This is a paragraph.</p>
+	<p><img src="` + server.URL + `/gophercolor16x16withoutextention" loading="lazy"/></p>`
 	testSectionBodyWithImageExpect := `    <h1>Section 1</h1>
 	<p>This is a paragraph.</p>
-	<p><img src="../images/gophercolor16x16.png" loading="lazy"/></p>`
+	<p><img src="../images/image0001.png" loading="lazy"/></p>`
+	testSectionBodyWithImageUnvalidnameExpect := `    <h1>Section 1</h1>
+	<p>This is a paragraph.</p>
+	<p><img src="../images/image0003.png" loading="lazy"/></p>`
+	testSectionBodyWithImageWithoutExtentionExpected := `    <h1>Section 1</h1>
+	<p>This is a paragraph.</p>
+	<p><img src="../images/image0004.png" loading="lazy"/></p>`
 	e, err := NewEpub(testEpubTitle)
 	if err != nil {
 		t.Error(err)
@@ -1036,6 +1048,18 @@ func TestEmbedImage(t *testing.T) {
 	if err != nil {
 		t.Errorf("Error adding section: %s", err)
 	}
+	_, err = e.AddSection(testSectionBodyWithImage, testSectionTitle, "", "")
+	if err != nil {
+		t.Errorf("Error adding section: %s", err)
+	}
+	testSection4Path, err := e.AddSection(testSectionBodyWithImageUnvalidname, testSectionTitle, "", "")
+	if err != nil {
+		t.Errorf("Error adding section: %s", err)
+	}
+	testSection5Path, err := e.AddSection(testSectionBodyWithImageWithoutExtention, testSectionTitle, "", "")
+	if err != nil {
+		t.Errorf("Error adding section: %s", err)
+	}
 	e.EmbedImages()
 	tempDir := writeAndExtractEpub(t, e, testEpubFilename)
 
@@ -1083,6 +1107,34 @@ func TestEmbedImage(t *testing.T) {
 			contents,
 			testSection3Contents)
 	}
+	// test 4
+	contents, err = storage.ReadFile(filesystem, filepath.Join(tempDir, contentFolderName, xhtmlFolderName, testSection4Path))
+	if err != nil {
+		t.Errorf("Unexpected error reading section file: %s", err)
+	}
+	testSection4Contents := fmt.Sprintf(testSectionContentTemplate, testSectionTitle, testSectionBodyWithImageUnvalidnameExpect)
+	if trimAllSpace(string(contents)) != trimAllSpace(testSection4Contents) {
+		t.Errorf(
+			"Section file contents don't match\n"+
+				"Got: %s\n"+
+				"Expected: %s",
+			contents,
+			testSection4Contents)
+	}
+	// test 5
+	contents, err = storage.ReadFile(filesystem, filepath.Join(tempDir, contentFolderName, xhtmlFolderName, testSection5Path))
+	if err != nil {
+		t.Errorf("Unexpected error reading section file: %s", err)
+	}
+	testSection5Contents := fmt.Sprintf(testSectionContentTemplate, testSectionTitle, testSectionBodyWithImageWithoutExtentionExpected)
+	if trimAllSpace(string(contents)) != trimAllSpace(testSection5Contents) {
+		t.Errorf(
+			"Section file contents don't match\n"+
+				"Got: %s\n"+
+				"Expected: %s",
+			contents,
+			testSection5Contents)
+	}
 	cleanup(testEpubFilename, tempDir)
 }