Merge pull request galaxyproject#18503 from mvdbeek/error_on_binary_d…

…ata_display

Raise Message Exception when displaying binary data

lib/galaxy/datatypes/sequence.py

@@ -41,6 +41,7 @@
     get_headers,
     iter_headers,
 )
+from galaxy.exceptions import InvalidFileFormatError
 from galaxy.util import (
     compression_utils,
     nice_size,
@@ -775,15 +776,17 @@ def display_data(
         headers = kwd.get("headers", {})
         if preview:
             with compression_utils.get_fileobj(dataset.get_file_name()) as fh:
-                max_peek_size = 1000000  # 1 MB
-                if os.stat(dataset.get_file_name()).st_size < max_peek_size:
+                max_peek_size = 100000
+                try:
+                    chunk = fh.read(max_peek_size + 1)
+                except UnicodeDecodeError:
+                    raise InvalidFileFormatError("Dataset appears to contain binary data, cannot display.")
+                if len(chunk) <= max_peek_size:
                     mime = "text/plain"
                     self._clean_and_set_mime_type(trans, mime, headers)
-                    return fh.read(), headers
+                    return chunk[:-1], headers
                 return (
-                    trans.fill_template_mako(
-                        "/dataset/large_file.mako", truncated_data=fh.read(max_peek_size), data=dataset
-                    ),
+                    trans.fill_template_mako("/dataset/large_file.mako", truncated_data=chunk[:-1], data=dataset),
                     headers,
                 )
         else:

lib/galaxy/datatypes/tabular.py

@@ -65,6 +65,7 @@
     iter_headers,
     validate_tabular,
 )
+from galaxy.exceptions import InvalidFileFormatError
 from galaxy.util import compression_utils
 from galaxy.util.compression_utils import (
     FileObjType,
@@ -156,12 +157,15 @@ def get_chunk(self, trans, dataset: HasFileName, offset: int = 0, ck_size: Optio
     def _read_chunk(self, trans, dataset: HasFileName, offset: int, ck_size: Optional[int] = None):
         with compression_utils.get_fileobj(dataset.get_file_name()) as f:
             f.seek(offset)
-            ck_data = f.read(ck_size or trans.app.config.display_chunk_size)
-            if ck_data and ck_data[-1] != "\n":
-                cursor = f.read(1)
-                while cursor and cursor != "\n":
-                    ck_data += cursor
+            try:
+                ck_data = f.read(ck_size or trans.app.config.display_chunk_size)
+                if ck_data and ck_data[-1] != "\n":
                     cursor = f.read(1)
+                    while cursor and cursor != "\n":
+                        ck_data += cursor
+                        cursor = f.read(1)
+            except UnicodeDecodeError:
+                raise InvalidFileFormatError("Dataset appears to contain binary data, cannot display.")
             last_read = f.tell()
         return ck_data, last_read