Added BigTIFF reading

Tests/test_file_tiff.py

@@ -87,6 +87,10 @@ def test_mac_tiff(self):
 
             assert_image_similar_tofile(im, "Tests/images/pil136.png", 1)
 
+    def test_bigtiff(self):
+        with Image.open("Tests/images/hopper_bigtiff.tif") as im:
+            assert_image_equal_tofile(im, "Tests/images/hopper.tif")
+
     @pytest.mark.parametrize(
         "file_name,mode,size,offset",
         [

src/PIL/Image.py

@@ -49,7 +49,7 @@
 # PILLOW_VERSION was removed in Pillow 9.0.0.
 # Use __version__ instead.
 from . import ImageMode, TiffTags, UnidentifiedImageError, __version__, _plugins
-from ._binary import i32le
+from ._binary import i32le, o32be, o32le
 from ._util import deferred_error, isPath
 
 
@@ -1416,6 +1416,7 @@ def getexif(self):
                     "".join(self.info["Raw profile type exif"].split("\n")[3:])
                 )
             elif hasattr(self, "tag_v2"):
+                self._exif.bigtiff = self.tag_v2._bigtiff
                 self._exif.endian = self.tag_v2._endian
                 self._exif.load_from_fp(self.fp, self.tag_v2._offset)
         if exif_info is not None:
@@ -3423,6 +3424,7 @@ def _apply_env_variables(env=None):
 
 class Exif(MutableMapping):
     endian = None
+    bigtiff = False
 
     def __init__(self):
         self._data = {}
@@ -3458,10 +3460,15 @@ def _get_ifd_dict(self, offset):
             return self._fixup_dict(info)
 
     def _get_head(self):
+        version = b"\x2B" if self.bigtiff else b"\x2A"
         if self.endian == "<":
-            return b"II\x2A\x00\x08\x00\x00\x00"
+            head = b"II" + version + b"\x00" + o32le(8)
         else:
-            return b"MM\x00\x2A\x00\x00\x00\x08"
+            head = b"MM\x00" + version + o32be(8)
+        if self.bigtiff:
+            head += o32le(8) if self.endian == "<" else o32be(8)
+            head += b"\x00\x00\x00\x00"
+        return head
 
     def load(self, data):
         # Extract EXIF information.  This is highly experimental,

src/PIL/TiffImagePlugin.py

@@ -260,6 +260,8 @@
     b"II\x2A\x00",  # Valid TIFF header with little-endian byte order
     b"MM\x2A\x00",  # Invalid TIFF header, assume big-endian
     b"II\x00\x2A",  # Invalid TIFF header, assume little-endian
+    b"MM\x00\x2B",  # BigTIFF with big-endian byte order
+    b"II\x2B\x00",  # BigTIFF with little-endian byte order
 ]
 
 
@@ -502,11 +504,14 @@ def __init__(self, ifh=b"II\052\0\0\0\0\0", prefix=None, group=None):
             self._endian = "<"
         else:
             raise SyntaxError("not a TIFF IFD")
+        self._bigtiff = ifh[2] == 43
         self.group = group
         self.tagtype = {}
         """ Dictionary of tag types """
         self.reset()
-        (self.next,) = self._unpack("L", ifh[4:])
+        (self.next,) = (
+            self._unpack("Q", ifh[8:]) if self._bigtiff else self._unpack("L", ifh[4:])
+        )
         self._legacy_api = False
 
     prefix = property(lambda self: self._prefix)
@@ -699,6 +704,7 @@ def _register_basic(idx_fmt_name):
                 (TiffTags.FLOAT, "f", "float"),
                 (TiffTags.DOUBLE, "d", "double"),
                 (TiffTags.IFD, "L", "long"),
+                (TiffTags.LONG8, "Q", "long8"),
             ],
         )
     )
@@ -776,8 +782,17 @@ def load(self, fp):
         self._offset = fp.tell()
 
         try:
-            for i in range(self._unpack("H", self._ensure_read(fp, 2))[0]):
-                tag, typ, count, data = self._unpack("HHL4s", self._ensure_read(fp, 12))
+            tag_count = (
+                self._unpack("Q", self._ensure_read(fp, 8))
+                if self._bigtiff
+                else self._unpack("H", self._ensure_read(fp, 2))
+            )[0]
+            for i in range(tag_count):
+                tag, typ, count, data = (
+                    self._unpack("HHQ8s", self._ensure_read(fp, 20))
+                    if self._bigtiff
+                    else self._unpack("HHL4s", self._ensure_read(fp, 12))
+                )
 
                 tagname = TiffTags.lookup(tag, self.group).name
                 typname = TYPES.get(typ, "unknown")
@@ -789,9 +804,9 @@ def load(self, fp):
                     logger.debug(msg + f" - unsupported type {typ}")
                     continue  # ignore unsupported type
                 size = count * unit_size
-                if size > 4:
+                if size > (8 if self._bigtiff else 4):
                     here = fp.tell()
-                    (offset,) = self._unpack("L", data)
+                    (offset,) = self._unpack("Q" if self._bigtiff else "L", data)
                     msg += f" Tag Location: {here} - Data Location: {offset}"
                     fp.seek(offset)
                     data = ImageFile._safe_read(fp, size)
@@ -820,7 +835,11 @@ def load(self, fp):
                 )
                 logger.debug(msg)
 
-            (self.next,) = self._unpack("L", self._ensure_read(fp, 4))
+            (self.next,) = (
+                self._unpack("Q", self._ensure_read(fp, 8))
+                if self._bigtiff
+                else self._unpack("L", self._ensure_read(fp, 4))
+            )
         except OSError as msg:
             warnings.warn(str(msg))
             return
@@ -1042,6 +1061,8 @@ def _open(self):
 
         # Header
         ifh = self.fp.read(8)
+        if ifh[2] == 43:
+            ifh += self.fp.read(8)
 
         self.tag_v2 = ImageFileDirectory_v2(ifh)
 

src/PIL/TiffTags.py

@@ -74,6 +74,7 @@ def lookup(tag, group=None):
 FLOAT = 11
 DOUBLE = 12
 IFD = 13
+LONG8 = 16
 
 TAGS_V2 = {
     254: ("NewSubfileType", LONG, 1),