Merge pull request #2567 from 1c3t3a/xmp-metadata

Provide a way to extract XMP metadata (png & webp & tiff only for now)

Author: 197g

src/codecs/png.rs

@@ -25,6 +25,7 @@ use crate::{
 // http://www.w3.org/TR/PNG-Structure.html
 // The first eight bytes of a PNG file always contain the following (decimal) values:
 pub(crate) const PNG_SIGNATURE: [u8; 8] = [137, 80, 78, 71, 13, 10, 26, 10];
+const XMP_KEY: &str = "XML:com.adobe.xmp";
 
 /// PNG decoder
 pub struct PngDecoder<R: BufRead + Seek> {
@@ -45,6 +46,7 @@ impl<R: BufRead + Seek> PngDecoder<R> {
 
         let max_bytes = usize::try_from(limits.max_alloc.unwrap_or(u64::MAX)).unwrap_or(usize::MAX);
         let mut decoder = png::Decoder::new_with_limits(r, png::Limits { bytes: max_bytes });
+        decoder.set_ignore_text_chunk(false);
 
         let info = decoder.read_header_info().map_err(ImageError::from_png)?;
         limits.check_dimensions(info.width, info.height)?;
@@ -186,6 +188,24 @@ impl<R: BufRead + Seek> ImageDecoder for PngDecoder<R> {
             .map(|x| x.to_vec()))
     }
 
+    fn xmp_metadata(&mut self) -> ImageResult<Option<Vec<u8>>> {
+        if let Some(mut itx_chunk) = self
+            .reader
+            .info()
+            .utf8_text
+            .iter()
+            .find(|chunk| chunk.keyword.contains(XMP_KEY))
+            .cloned()
+        {
+            itx_chunk.decompress_text().map_err(ImageError::from_png)?;
+            return itx_chunk
+                .get_text()
+                .map(|text| Some(text.as_bytes().to_vec()))
+                .map_err(ImageError::from_png);
+        }
+        Ok(None)
+    }
+
     fn read_image(mut self, buf: &mut [u8]) -> ImageResult<()> {
         use byteorder_lite::{BigEndian, ByteOrder, NativeEndian};
 

src/codecs/tiff.rs

@@ -20,6 +20,8 @@ use crate::error::{
 use crate::metadata::Orientation;
 use crate::{utils, ImageDecoder, ImageEncoder, ImageFormat};
 
+const TAG_XML_PACKET: Tag = Tag::Unknown(700);
+
 /// Decoder for TIFF images.
 pub struct TiffDecoder<R>
 where
@@ -264,6 +266,24 @@ impl<R: BufRead + Seek> ImageDecoder for TiffDecoder<R> {
         }
     }
 
+    fn xmp_metadata(&mut self) -> ImageResult<Option<Vec<u8>>> {
+        let Some(decoder) = &mut self.inner else {
+            return Ok(None);
+        };
+
+        let value = match decoder.get_tag(TAG_XML_PACKET) {
+            Ok(value) => value,
+            Err(tiff::TiffError::FormatError(tiff::TiffFormatError::RequiredTagNotFound(_))) => {
+                return Ok(None);
+            }
+            Err(err) => return Err(ImageError::from_tiff_decode(err)),
+        };
+        value
+            .into_u8_vec()
+            .map(Some)
+            .map_err(ImageError::from_tiff_decode)
+    }
+
     fn orientation(&mut self) -> ImageResult<Orientation> {
         if let Some(decoder) = &mut self.inner {
             Ok(decoder

src/codecs/webp/decoder.rs

@@ -84,6 +84,12 @@ impl<R: BufRead + Seek> ImageDecoder for WebPDecoder<R> {
         Ok(exif)
     }
 
+    fn xmp_metadata(&mut self) -> ImageResult<Option<Vec<u8>>> {
+        self.inner
+            .xmp_metadata()
+            .map_err(ImageError::from_webp_decode)
+    }
+
     fn orientation(&mut self) -> ImageResult<Orientation> {
         // `exif_metadata` caches the orientation, so call it if `orientation` hasn't been set yet.
         if self.orientation.is_none() {

src/io/decoder.rs

@@ -31,6 +31,14 @@ pub trait ImageDecoder {
         Ok(None)
     }
 
+    /// Returns the raw [XMP](https://en.wikipedia.org/wiki/Extensible_Metadata_Platform) chunk, if it is present.
+    /// A third-party crate such as [`roxmltree`](https://docs.rs/roxmltree/) is required to actually parse it.
+    ///
+    /// For formats that don't support embedded profiles this function should always return `Ok(None)`.
+    fn xmp_metadata(&mut self) -> ImageResult<Option<Vec<u8>>> {
+        Ok(None)
+    }
+
     /// Returns the orientation of the image.
     ///
     /// This is usually obtained from the Exif metadata, if present. Formats that don't support
@@ -134,6 +142,9 @@ impl<T: ?Sized + ImageDecoder> ImageDecoder for Box<T> {
     fn exif_metadata(&mut self) -> ImageResult<Option<Vec<u8>>> {
         (**self).exif_metadata()
     }
+    fn xmp_metadata(&mut self) -> ImageResult<Option<Vec<u8>>> {
+        (**self).xmp_metadata()
+    }
     fn orientation(&mut self) -> ImageResult<Orientation> {
         (**self).orientation()
     }

tests/images/png/transparency/tp1n3p08_xmp.png

@@ -0,0 +1,66 @@
+use std::fs;
+use std::path::PathBuf;
+use std::str::FromStr;
+
+use image::ImageDecoder;
+
+#[cfg(feature = "png")]
+use image::codecs::png::PngDecoder;
+#[cfg(feature = "tiff")]
+use image::codecs::tiff::TiffDecoder;
+#[cfg(feature = "webp")]
+use image::codecs::webp::WebPDecoder;
+
+extern crate glob;
+extern crate image;
+
+const XMP_PNG_PATH: &str = "tests/images/png/transparency/tp1n3p08_xmp.png";
+const EXPECTED_PNG_METADATA: &str = "<?xpacket begin='\u{feff}' id='W5M0MpCehiHzreSzNTczkc9d'?>\n<x:xmpmeta xmlns:x='adobe:ns:meta/' x:xmptk='Image::ExifTool 13.25'>\n<rdf:RDF xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#'>\n\n <rdf:Description rdf:about=''\n  xmlns:dc='http://purl.org/dc/elements/1.1/'>\n  <dc:subject>\n   <rdf:Bag>\n    <rdf:li>sunset, mountains, nature</rdf:li>\n   </rdf:Bag>\n  </dc:subject>\n </rdf:Description>\n</rdf:RDF>\n</x:xmpmeta>\n<?xpacket end='r'?>";
+
+const XMP_WEBP_PATH: &str = "tests/images/webp/lossless_images/simple_xmp.webp";
+const EXPECTED_WEBP_TIFF_METADATA: &str = "<?xpacket begin='\u{feff}' id='W5M0MpCehiHzreSzNTczkc9d'?>\n<x:xmpmeta xmlns:x='adobe:ns:meta/' x:xmptk='Image::ExifTool 13.25'>\n<rdf:RDF xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#'>\n\n <rdf:Description rdf:about=''\n  xmlns:dc='http://purl.org/dc/elements/1.1/'>\n  <dc:subject>\n   <rdf:Bag>\n    <rdf:li>sunset, mountains, nature</rdf:li>\n   </rdf:Bag>\n  </dc:subject>\n </rdf:Description>\n</rdf:RDF>\n</x:xmpmeta>\n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n                                                                                                    \n<?xpacket end='w'?>";
+
+const XMP_TIFF_PATH: &str = "tests/images/tiff/testsuite/l1_xmp.tiff";
+
+#[test]
+#[cfg(feature = "png")]
+fn test_read_xmp_png() -> Result<(), image::ImageError> {
+    let img_path = PathBuf::from_str(XMP_PNG_PATH).unwrap();
+
+    let data = fs::read(img_path)?;
+    let mut png_decoder = PngDecoder::new(std::io::Cursor::new(data))?;
+    let metadata = png_decoder.xmp_metadata()?;
+    assert!(metadata.is_some());
+    assert_eq!(EXPECTED_PNG_METADATA.as_bytes(), metadata.unwrap());
+
+    Ok(())
+}
+
+#[test]
+#[cfg(feature = "webp")]
+fn test_read_xmp_webp() -> Result<(), image::ImageError> {
+    let img_path = PathBuf::from_str(XMP_WEBP_PATH).unwrap();
+
+    let data = fs::read(img_path)?;
+    let mut webp_decoder = WebPDecoder::new(std::io::Cursor::new(data))?;
+    let metadata = webp_decoder.xmp_metadata()?;
+
+    assert!(metadata.is_some());
+    assert_eq!(EXPECTED_WEBP_TIFF_METADATA.as_bytes(), metadata.unwrap());
+
+    Ok(())
+}
+
+#[test]
+#[cfg(feature = "tiff")]
+fn test_read_xmp_tiff() -> Result<(), image::ImageError> {
+    let img_path = PathBuf::from_str(XMP_TIFF_PATH).unwrap();
+
+    let data = fs::read(img_path)?;
+    let mut tiff_decoder = TiffDecoder::new(std::io::Cursor::new(data))?;
+    let metadata = tiff_decoder.xmp_metadata()?;
+    assert!(metadata.is_some());
+    assert_eq!(EXPECTED_WEBP_TIFF_METADATA.as_bytes(), metadata.unwrap());
+
+    Ok(())
+}