formatscaper/cli/formatscaper.py

@@ -3,8 +3,10 @@
 import argparse
 import os
 import re
+import shutil
 import subprocess
 import sys
+import tempfile
 import threading
 
 import progressbar as pb
@@ -79,7 +81,7 @@ def scape_formats(config):
         m = re.match(r"siegfried ((\d+\.?)+)", sf_output)
         if m and m.group(1):
             ver_nums = [int(num) for num in m.group(1).split(".")]
-            if not (ver_nums[0] >= 1 and ver_nums[1] >= 10):
+            if not (ver_nums[0] >= 1 and ver_nums[1] >= 11):
                 print(
                     f"WARN: siegfried version too old ({m.group(1)})", file=sys.stderr
                 )
@@ -159,9 +161,14 @@ def scape_formats(config):
         pb.Timer(),
     ]
     progress_bar = pb.ProgressBar(max_value=len(record_files), widgets=pb_ws)
+    base_dir = tempfile.mkdtemp()
 
     def process_record_file(record_file: RecordFile) -> None:
         with sem:
+            # link the files under investigation into a scoped directory
+            file_dir = os.path.join(base_dir, record_file.record)
+            file_path = os.path.join(file_dir, record_file.filename)
+
             try:
                 # if we already have an overridden result for the record file
                 # in question, we skip it
@@ -179,15 +186,22 @@ def scape_formats(config):
                     file_infos = []
 
                 else:
+                    # create a symlink to the file with a proper name to help siegfried
+                    # with file format identification as the file name plays a role
+                    # (this will be deleted afterwards)
+                    os.makedirs(file_dir, exist_ok=True)
+                    os.symlink(record_file.uri, file_path)
+
                     sf_output = subprocess.check_output(
                         [
                             config.sf_binary,
+                            "-sym",
                             "-z",
                             "-multi",
                             "1",
                             "-name",
                             record_file.filename,
-                            record_file.uri,
+                            file_path,
                         ],
                         stderr=sf_error_log,
                     )
@@ -217,7 +231,7 @@ def scape_formats(config):
 
                                     # replace first occurrence of the URI with filename
                                     filename = file_info["filename"].replace(
-                                        record_file.uri, record_file.filename, 1
+                                        (file_dir + os.path.sep), "", 1
                                     )
 
                                     result = Result(
@@ -239,6 +253,13 @@ def scape_formats(config):
             except (subprocess.CalledProcessError, ValueError) as e:
                 print("WARN: error during sf execution:", str(e), file=sys.stderr)
 
+            finally:
+                try:
+                    # in any case, remove the symlink to the file we generated
+                    os.remove(file_path)
+                except FileNotFoundError:
+                    pass
+
     # analyze all files in parallel, and create the summary after all threads complete
     threads = []
     for record_file in record_files or []:
@@ -252,6 +273,8 @@ def scape_formats(config):
     except KeyboardInterrupt:
         pass
 
+    # clean up
+    shutil.rmtree(base_dir)
     if sf_error_log is not None:
         sf_error_log.close()
 

formatscaper/models.py

 import dataclasses
 from typing import List, Optional
 
-from sqlalchemy import ForeignKey, create_engine
+from sqlalchemy import ForeignKey, UniqueConstraint, create_engine
 from sqlalchemy.orm import DeclarativeBase, Mapped, Session, mapped_column, relationship
 
 
@@ -98,6 +98,9 @@ class Result(ModelBase):
     format_id: Mapped[Optional[int]] = mapped_column(ForeignKey("format.id"))
     format: Mapped[Format] = relationship(back_populates="results")
 
+    # filenames are unique per record
+    __table_args__ = UniqueConstraint("record", "filename")
+
     @property
     def risk(self):
         """Calculate the risk assessment for the file."""