From e70ea811c99c16f3382c08153eda573df0825536 Mon Sep 17 00:00:00 2001
From: jvoisin
Date: Thu, 18 Oct 2018 19:19:56 +0200
Subject: Implement support for .avi files, via ffmpeg

- This commit introduces optional dependencies (namely ffmpeg):
  mat2 will spit a warning when trying to process an .avi file
  if ffmpeg isn't installed.
- Since metadata are obtained via exiftool, this commit
  also refactors a bit our exfitool wrapper.
---
 libmat2/exiftool.py | 61 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 61 insertions(+)
 create mode 100644 libmat2/exiftool.py

(limited to 'libmat2/exiftool.py')

diff --git a/libmat2/exiftool.py b/libmat2/exiftool.py
new file mode 100644
index 0000000..e17d31b
--- /dev/null
+++ b/libmat2/exiftool.py
@@ -0,0 +1,61 @@
+import json
+import os
+import re
+import shutil
+import subprocess
+import tempfile
+
+from typing import Dict, Union, Set
+
+from . import abstract
+
+# Make pyflakes happy
+assert Set
+
+
+class ExiftoolParser(abstract.AbstractParser):
+    """ Exiftool is often the easiest way to get all the metadata
+    from a import file, hence why several parsers are re-using its `get_meta`
+    method.
+    """
+    meta_whitelist = set()  # type: Set[str]
+
+    @staticmethod
+    def __handle_problematic_filename(filename: str, callback) -> bytes:
+        """ This method takes a filename with a problematic name,
+        and safely applies it a `callback`."""
+        tmpdirname = tempfile.mkdtemp()
+        fname = os.path.join(tmpdirname, "temp_file")
+        shutil.copy(filename, fname)
+        out = callback(fname)
+        shutil.rmtree(tmpdirname)
+        return out
+
+    def get_meta(self) -> Dict[str, Union[str, dict]]:
+        """ There is no way to escape the leading(s) dash(es) of the current
+        self.filename to prevent parameter injections, so we need to take care
+        of this.
+        """
+        fun = lambda f: subprocess.check_output([_get_exiftool_path(), '-json', f])
+        if re.search('^[a-z0-9/]', self.filename) is None:
+            out = self.__handle_problematic_filename(self.filename, fun)
+        else:
+            out = fun(self.filename)
+        meta = json.loads(out.decode('utf-8'))[0]
+        for key in self.meta_whitelist:
+            meta.pop(key, None)
+        return meta
+
+def _get_exiftool_path() -> str:  # pragma: no cover
+    exiftool_path = '/usr/bin/exiftool'
+    if os.path.isfile(exiftool_path):
+        if os.access(exiftool_path, os.X_OK):
+            return exiftool_path
+
+    # ArchLinux
+    exiftool_path = '/usr/bin/vendor_perl/exiftool'
+    if os.path.isfile(exiftool_path):
+        if os.access(exiftool_path, os.X_OK):
+            return exiftool_path
+
+    raise RuntimeError("Unable to find exiftool")
-- 
cgit v1.3