Metadata: Add support for XMP sidecar CreateDate and Keywords (#1161)

* Metadata: Read title, description, date and keywords from apple xmp

* Metadata: Add testfiles and tests

* Metadata: Add support for XMP sidecar CreateDate and Keywords #1151

Co-authored-by: Michael Mayer <michael@lastzero.net>

internal/meta/testdata/apple-test-2.xmp

@@ -0,0 +1,25 @@
+<x:xmpmeta xmlns:x="adobe:ns:meta/" x:xmptk="XMP Core 6.0.0">
+   <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
+      <rdf:Description rdf:about=""
+            xmlns:exif="http://ns.adobe.com/exif/1.0/"
+            xmlns:dc="http://purl.org/dc/elements/1.1/"
+            xmlns:photoshop="http://ns.adobe.com/photoshop/1.0/">
+         <exif:GPSLongitude>13.369367199999999</exif:GPSLongitude>
+         <exif:GPSLongitudeRef>E</exif:GPSLongitudeRef>
+         <exif:GPSHPositioningError>1</exif:GPSHPositioningError>
+         <exif:GPSLatitude>52.5250816</exif:GPSLatitude>
+         <exif:GPSLatitudeRef>N</exif:GPSLatitudeRef>
+         <exif:GPSTimeStamp>2021-03-26T09:18:59Z</exif:GPSTimeStamp>
+         <dc:title>Botanischer Garten</dc:title>
+         <dc:description>Tulpen am See</dc:description>
+         <dc:subject>
+            <rdf:Seq>
+               <rdf:li>Krokus</rdf:li>
+               <rdf:li>Blume</rdf:li>
+               <rdf:li>Schöne Wiese</rdf:li>
+            </rdf:Seq>
+         </dc:subject>
+         <photoshop:DateCreated>2021-03-24T13:07:29+01:00</photoshop:DateCreated>
+      </rdf:Description>
+   </rdf:RDF>
+</x:xmpmeta>

internal/meta/xmp.go

@@ -57,5 +57,13 @@ func (data *Data) XMP(fileName string) (err error) {
 		data.LensModel = doc.LensModel()
 	}
 
+	if takenAt := doc.TakenAt(); !takenAt.IsZero() {
+		data.TakenAt = takenAt
+	}
+
+	if len(doc.Keywords()) != 0 {
+		data.AddKeywords(doc.Keywords())
+	}
+
 	return nil
 }

internal/meta/xmp_document.go

@@ -3,6 +3,8 @@ package meta
 import (
 	"encoding/xml"
 	"io/ioutil"
+	"strings"
+	"time"
 )
 
 // XmpDocument represents an XMP sidecar file.
@@ -75,6 +77,10 @@ type XmpDocument struct {
 					Text string   `xml:",chardata" json:"text,omitempty"`
 					Li   []string `xml:"li"` // desk, coffee, computer
 				} `xml:"Bag" json:"bag,omitempty"`
+				Seq struct {
+					Text string   `xml:",chardata" json:"text,omitempty"`
+					Li   []string `xml:"li"` // desk, coffee, computer
+				} `xml:"Seq" json:"seq,omitempty"`
 			} `xml:"subject" json:"subject,omitempty"`
 			Rights struct {
 				Text string `xml:",chardata" json:"text,omitempty"`
@@ -192,6 +198,7 @@ type XmpDocument struct {
 	} `xml:"RDF" json:"rdf,omitempty"`
 }
 
+// Load parses an XMP file and populates document values with its contents.
 func (doc *XmpDocument) Load(filename string) error {
 	data, err := ioutil.ReadFile(filename)
 
@@ -202,30 +209,81 @@ func (doc *XmpDocument) Load(filename string) error {
 	return xml.Unmarshal(data, doc)
 }
 
+// Title returns the XMP document title.
 func (doc *XmpDocument) Title() string {
-	return SanitizeTitle(doc.RDF.Description.Title.Alt.Li.Text)
+	t := doc.RDF.Description.Title.Alt.Li.Text
+	t2 := doc.RDF.Description.Title.Text
+	if t != "" {
+		return SanitizeTitle(t)
+	} else if t2 != "" {
+		return SanitizeTitle(t2)
+	}
+	return ""
 }
 
+// Artist returns the XMP document artist.
 func (doc *XmpDocument) Artist() string {
 	return SanitizeString(doc.RDF.Description.Creator.Seq.Li)
 }
 
+// Description returns the XMP document description.
 func (doc *XmpDocument) Description() string {
-	return SanitizeDescription(doc.RDF.Description.Description.Alt.Li.Text)
+	d := doc.RDF.Description.Description.Alt.Li.Text
+	d2 := doc.RDF.Description.Description.Text
+	if d != "" {
+		return SanitizeDescription(d)
+	} else if d2 != "" {
+		return SanitizeTitle(d2)
+	}
+	return ""
 }
 
+// Copyright returns the XMP document copyright info.
 func (doc *XmpDocument) Copyright() string {
 	return SanitizeString(doc.RDF.Description.Rights.Alt.Li.Text)
 }
 
+// CameraMake returns the XMP document camera make name.
 func (doc *XmpDocument) CameraMake() string {
 	return SanitizeString(doc.RDF.Description.Make)
 }
 
+// CameraModel returns the XMP document camera model name.
 func (doc *XmpDocument) CameraModel() string {
 	return SanitizeString(doc.RDF.Description.Model)
 }
 
+// LensModel returns the XMP document lens model name.
 func (doc *XmpDocument) LensModel() string {
 	return SanitizeString(doc.RDF.Description.LensModel)
 }
+
+// TakenAt returns the XMP document taken date.
+func (doc *XmpDocument) TakenAt() time.Time {
+	taken := time.Time{} // Unknown
+
+	s := SanitizeString(doc.RDF.Description.DateCreated)
+
+	if s == "" {
+		return taken
+	}
+
+	if t, err := time.Parse(time.RFC3339, s); err == nil {
+		taken = t
+	} else if t, err := time.Parse("2006-01-02T15:04:05.999999999", s); err == nil {
+		taken = t
+	} else if t, err := time.Parse("2006-01-02T15:04:05-07:00", s); err == nil {
+		taken = t
+	} else if t, err := time.Parse("2006-01-02T15:04:05", s[:19]); err == nil {
+		taken = t
+	}
+
+	return taken
+}
+
+// Keywords returns the XMP document keywords.
+func (doc *XmpDocument) Keywords() string {
+	s := doc.RDF.Description.Subject.Seq.Li
+
+	return strings.Join(s, ", ")
+}

internal/meta/xmp_test.go

@@ -2,11 +2,25 @@ package meta
 
 import (
 	"testing"
+	"time"
 
 	"github.com/stretchr/testify/assert"
 )
 
 func TestXMP(t *testing.T) {
+	t.Run("apple xmp 2", func(t *testing.T) {
+		data, err := XMP("testdata/apple-test-2.xmp")
+
+		if err != nil {
+			t.Fatal(err)
+		}
+
+		assert.Equal(t, "Botanischer Garten", data.Title)
+		assert.Equal(t, time.Date(2021, 3, 24, 13, 07, 29, 0, time.FixedZone("", +3600)), data.TakenAt)
+		assert.Equal(t, "Tulpen am See", data.Description)
+		assert.Equal(t, Keywords{"blume", "krokus", "schöne", "wiese"}, data.Keywords)
+	})
+
 	t.Run("photoshop", func(t *testing.T) {
 		data, err := XMP("testdata/photoshop.xmp")
 
@@ -15,6 +29,8 @@ func TestXMP(t *testing.T) {
 		}
 
 		assert.Equal(t, "Night Shift / Berlin / 2020", data.Title)
+		t.Log(data.TakenAt)
+		assert.Equal(t, time.Date(2020, 1, 1, 17, 28, 25, 729626112, time.UTC), data.TakenAt)
 		assert.Equal(t, "Michael Mayer", data.Artist)
 		assert.Equal(t, "Example file for development", data.Description)
 		assert.Equal(t, "This is an (edited) legal notice", data.Copyright)

internal/photoprism/index_related_test.go

@@ -13,61 +13,129 @@ import (
 )
 
 func TestIndexRelated(t *testing.T) {
-	conf := config.TestConfig()
+	t.Run("2018-04-12 19_24_49.gif", func(t *testing.T) {
+		conf := config.TestConfig()
 
-	testFile, err := NewMediaFile("testdata/2018-04-12 19_24_49.gif")
+		testFile, err := NewMediaFile("testdata/2018-04-12 19_24_49.gif")
 
-	if err != nil {
+		if err != nil {
-		t.Fatal(err)
+			t.Fatal(err)
-	}
-
-	testRelated, err := testFile.RelatedFiles(true)
-
-	if err != nil {
-		t.Fatal(err)
-	}
-
-	testToken := rnd.Token(8)
-	testPath := filepath.Join(conf.OriginalsPath(), testToken)
-
-	for _, f := range testRelated.Files {
-		dest := filepath.Join(testPath, f.BaseName())
-
-		if err := f.Copy(dest); err != nil {
-			t.Fatalf("COPY FAILED: %s", err)
 		}
-	}
 
-	mainFile, err := NewMediaFile(filepath.Join(testPath, "2018-04-12 19_24_49.gif"))
+		testRelated, err := testFile.RelatedFiles(true)
 
-	if err != nil {
+		if err != nil {
-		t.Fatal(err)
+			t.Fatal(err)
-	}
+		}
 
-	related, err := mainFile.RelatedFiles(true)
+		testToken := rnd.Token(8)
+		testPath := filepath.Join(conf.OriginalsPath(), testToken)
 
-	if err != nil {
+		for _, f := range testRelated.Files {
-		t.Fatal(err)
+			dest := filepath.Join(testPath, f.BaseName())
-	}
 
-	tf := classify.New(conf.AssetsPath(), conf.DisableTensorFlow())
+			if err := f.Copy(dest); err != nil {
-	nd := nsfw.New(conf.NSFWModelPath())
+				t.Fatalf("copying test file failed: %s", err)
-	convert := NewConvert(conf)
+			}
+		}
 
-	ind := NewIndex(conf, tf, nd, convert, NewFiles(), NewPhotos())
+		mainFile, err := NewMediaFile(filepath.Join(testPath, "2018-04-12 19_24_49.gif"))
-	opt := IndexOptionsAll()
 
-	result := IndexRelated(related, ind, opt)
+		if err != nil {
+			t.Fatal(err)
+		}
 
-	assert.False(t, result.Failed())
+		related, err := mainFile.RelatedFiles(true)
-	assert.False(t, result.Stacked())
-	assert.True(t, result.Success())
-	assert.Equal(t, IndexAdded, result.Status)
 
-	if photo, err := query.PhotoByUID(result.PhotoUID); err != nil {
+		if err != nil {
-		t.Fatal(err)
+			t.Fatal(err)
-	} else {
+		}
-		assert.Equal(t, "2018-04-12 19:24:49 +0000 UTC", photo.TakenAt.String())
+
-		assert.Equal(t, "name", photo.TakenSrc)
+		tf := classify.New(conf.AssetsPath(), conf.DisableTensorFlow())
-	}
+		nd := nsfw.New(conf.NSFWModelPath())
+		convert := NewConvert(conf)
+
+		ind := NewIndex(conf, tf, nd, convert, NewFiles(), NewPhotos())
+		opt := IndexOptionsAll()
+
+		result := IndexRelated(related, ind, opt)
+
+		assert.False(t, result.Failed())
+		assert.False(t, result.Stacked())
+		assert.True(t, result.Success())
+		assert.Equal(t, IndexAdded, result.Status)
+
+		if photo, err := query.PhotoByUID(result.PhotoUID); err != nil {
+			t.Fatal(err)
+		} else {
+			assert.Equal(t, "2018-04-12 19:24:49 +0000 UTC", photo.TakenAt.String())
+			assert.Equal(t, "name", photo.TakenSrc)
+		}
+	})
+
+	t.Run("apple-test-2.jpg", func(t *testing.T) {
+		conf := config.TestConfig()
+
+		testFile, err := NewMediaFile("testdata/apple-test-2.jpg")
+
+		if err != nil {
+			t.Fatal(err)
+		}
+
+		testRelated, err := testFile.RelatedFiles(true)
+
+		if err != nil {
+			t.Fatal(err)
+		}
+
+		testToken := rnd.Token(8)
+		testPath := filepath.Join(conf.OriginalsPath(), testToken)
+
+		for _, f := range testRelated.Files {
+			dest := filepath.Join(testPath, f.BaseName())
+
+			if err := f.Copy(dest); err != nil {
+				t.Fatalf("copying test file failed: %s", err)
+			}
+		}
+
+		mainFile, err := NewMediaFile(filepath.Join(testPath, "apple-test-2.jpg"))
+
+		if err != nil {
+			t.Fatal(err)
+		}
+
+		related, err := mainFile.RelatedFiles(true)
+
+		if err != nil {
+			t.Fatal(err)
+		}
+
+		tf := classify.New(conf.AssetsPath(), conf.DisableTensorFlow())
+		nd := nsfw.New(conf.NSFWModelPath())
+		convert := NewConvert(conf)
+
+		ind := NewIndex(conf, tf, nd, convert, NewFiles(), NewPhotos())
+		opt := IndexOptionsAll()
+
+		result := IndexRelated(related, ind, opt)
+
+		assert.False(t, result.Failed())
+		assert.False(t, result.Stacked())
+		assert.True(t, result.Success())
+		assert.Equal(t, IndexAdded, result.Status)
+
+		if photo, err := query.PhotoByUID(result.PhotoUID); err != nil {
+			t.Fatal(err)
+		} else {
+			assert.Equal(t, "Botanischer Garten", photo.PhotoTitle)
+			assert.Equal(t, "Tulpen am See", photo.PhotoDescription)
+			assert.Contains(t, photo.Details.Keywords, "krokus")
+			assert.Contains(t, photo.Details.Keywords, "blume")
+			assert.Contains(t, photo.Details.Keywords, "schöne")
+			assert.Contains(t, photo.Details.Keywords, "wiese")
+			assert.Equal(t, "2021-03-24 12:07:29 +0000 UTC", photo.TakenAt.String())
+			assert.Equal(t, "xmp", photo.TakenSrc)
+		}
+	})
 }

internal/photoprism/testdata/apple-test-2.xmp

@@ -0,0 +1,25 @@
+<x:xmpmeta xmlns:x="adobe:ns:meta/" x:xmptk="XMP Core 6.0.0">
+   <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
+      <rdf:Description rdf:about=""
+            xmlns:exif="http://ns.adobe.com/exif/1.0/"
+            xmlns:dc="http://purl.org/dc/elements/1.1/"
+            xmlns:photoshop="http://ns.adobe.com/photoshop/1.0/">
+         <exif:GPSLongitude>13.369367199999999</exif:GPSLongitude>
+         <exif:GPSLongitudeRef>E</exif:GPSLongitudeRef>
+         <exif:GPSHPositioningError>1</exif:GPSHPositioningError>
+         <exif:GPSLatitude>52.5250816</exif:GPSLatitude>
+         <exif:GPSLatitudeRef>N</exif:GPSLatitudeRef>
+         <exif:GPSTimeStamp>2021-03-26T09:18:59Z</exif:GPSTimeStamp>
+         <dc:title>Botanischer Garten</dc:title>
+         <dc:description>Tulpen am See</dc:description>
+         <dc:subject>
+            <rdf:Seq>
+               <rdf:li>Krokus</rdf:li>
+               <rdf:li>Blume</rdf:li>
+               <rdf:li>Schöne Wiese</rdf:li>
+            </rdf:Seq>
+         </dc:subject>
+         <photoshop:DateCreated>2021-03-24T13:07:29+01:00</photoshop:DateCreated>
+      </rdf:Description>
+   </rdf:RDF>
+</x:xmpmeta>