Version 2.0.0b2 (#108)

Author: cdgriffith

README.rst

@@ -157,9 +157,8 @@ Acknowledgements
 
 Gary C. Kessler
 
-
 For use of his File Signature Tables, available at:
-http://www.garykessler.net/library/file_sigs.html
+https://filesig.search.org/
 
 Freedesktop.org
 

puremagic/main.py

@@ -9,11 +9,9 @@
 Acknowledgements
 Gary C. Kessler
     For use of his File Signature Tables, available at:
-    http://www.garykessler.net/library/file_sigs.html
+    https://filesig.search.org/
 """
 
-from __future__ import annotations
-
 import json
 import os
 from binascii import unhexlify
@@ -22,10 +20,10 @@
 from pathlib import Path
 
 if os.getenv("PUREMAGIC_DEEPSCAN") != "0":
-    from puremagic.scanners import zip_scanner, pdf_scanner, text_scanner
+    from puremagic.scanners import zip_scanner, pdf_scanner, text_scanner, json_scanner, python_scanner
 
 __author__ = "Chris Griffith"
-__version__ = "2.0.0b1"
+__version__ = "2.0.0b2"
 __all__ = [
     "magic_file",
     "magic_string",
@@ -262,7 +260,7 @@ def ext_from_filename(filename: os.PathLike | str) -> str:
     all_exts = [x.extension for x in chain(magic_header_array, magic_footer_array)]
 
     if base[-4:].startswith("."):
-        # For double extensions like like .tar.gz
+        # For double extensions like .tar.gz
         long_ext = base[-4:] + ext
         if long_ext in all_exts:
             return long_ext
@@ -320,7 +318,7 @@ def from_stream(stream, mime: bool = False, filename: os.PathLike | str | None =
 def magic_file(filename: os.PathLike | str) -> list[PureMagicWithConfidence]:
     """
     Returns list of (num_of_matches, array_of_matches)
-    arranged highest confidence match first.
+    arranged by highest confidence match first.
 
     :param filename: path to file
     :return: list of possible matches, highest confidence first
@@ -341,7 +339,7 @@ def magic_file(filename: os.PathLike | str) -> list[PureMagicWithConfidence]:
 def magic_string(string, filename: os.PathLike | str | None = None) -> list[PureMagicWithConfidence]:
     """
     Returns tuple of (num_of_matches, array_of_matches)
-    arranged highest confidence match first
+    arranged by highest confidence match first
     If filename is provided it will be used in the computation.
 
     :param string: string representation to check
@@ -362,7 +360,7 @@ def magic_stream(
     filename: os.PathLike | None = None,
 ) -> list[PureMagicWithConfidence]:
     """Returns tuple of (num_of_matches, array_of_matches)
-    arranged highest confidence match first
+    arranged by highest confidence match first
     If filename is provided it will be used in the computation.
 
     :param stream: stream representation to check
@@ -393,14 +391,27 @@ def _single_deep_scan(
             return zip_scanner.main(filename, head, foot)
         case pdf_scanner.match_bytes:
             return pdf_scanner.main(filename, head, foot)
-        case None | b"":
-            for scanner in (text_scanner, pdf_scanner):
-                result = scanner.main(filename, head, foot)
-                if result:
-                    return result
+
+    # First match wins, so text_scanner should always be last
+    for scanner in (pdf_scanner, python_scanner, json_scanner):
+        result = scanner.main(filename, head, foot)
+        if result:
+            return result
     return None
 
 
+def _catch_all_deep_scan(
+    filename: os.PathLike | str,
+    head=None,
+    foot=None,
+):
+    if os.getenv("PUREMAGIC_DEEPSCAN") == "0":
+        return None
+    if not isinstance(filename, os.PathLike):
+        filename = Path(filename)
+    return text_scanner.main(filename, head, foot)
+
+
 def _run_deep_scan(
     matches: list[PureMagicWithConfidence],
     filename: os.PathLike | str,
@@ -425,10 +436,18 @@ def _run_deep_scan(
                         name=result.name,
                     )
                 ]
+        try:
+            result = _catch_all_deep_scan(filename, head, foot)
+        except Exception:
+            pass
+        else:
+            if result:
+                return [result]
         if raise_on_none:
             raise PureError("Could not identify file")
 
     for pure_magic_match in matches:
+        # noinspection PyBroadException
         try:
             result = _single_deep_scan(pure_magic_match.byte_match, filename, head, foot)
         except Exception:
@@ -484,7 +503,7 @@ def command_line_entry(*args):
                 if i == 0:
                     print("\n\tBest Match")
                 else:
-                    print(f"\tAlertnative Match #{i}")
+                    print(f"\tAlternative Match #{i}")
                 print(f"\tName: {result.name}")
                 print(f"\tConfidence: {int(result.confidence * 100)}%")
                 print(f"\tExtension: {result.extension}")
@@ -526,14 +545,14 @@ def what(file: os.PathLike | str | None, h: bytes | None = None, imghdr_strict:
 
     imghdr_strict enables bug-for-bug compatibility between imghdr.what() and puremagic.what() when the imghdr returns
     a match but puremagic returns None.  We believe that imghdr is delivering a "false positive" in each of these
-    scenerios but we want puremagic.what()'s default behavior to match imghdr.what()'s false positives so we do not
+    scenarios, but we want puremagic.what()'s default behavior to match imghdr.what()'s false positives so we do not
     break existing applications.
 
     If imghdr_strict is True (the default) then a lookup will be done to deliver a matching result on all known false
     positives.  If imghdr_strict is False then puremagic's algorithms will determine the image type.  True is more
     compatible while False is more correct.
 
-    NOTE: This compatibility effort only deals false positives and we are not interested to track the opposite
+    NOTE: This compatibility effort only deals false positives, and we are not interested to track the opposite
     situation where puremagic's deliver a match while imghdr would have returned None.  Also, puremagic.what() can
     recognize many more file types than the twelve image file types that imghdr focused on.
     """

puremagic/scanners/json_scanner.py

@@ -0,0 +1,22 @@
+import os
+import json
+
+from puremagic.scanners.helpers import Match
+
+match_bytes = b"{"
+
+
+def main(file_path: os.PathLike | str, head: bytes, foot: bytes) -> Match | None:
+    if not (head.strip().startswith(b"{") and foot.strip().endswith(b"}")):
+        return None
+    try:
+        with open(file_path, "rb") as file:
+            json.load(file)
+    except (json.decoder.JSONDecodeError, OSError):
+        return None
+    return Match(
+        extension=".json",
+        name="JSON File",
+        mime_type="application/json",
+        confidence=1.0,
+    )

puremagic/scanners/pdf_scanner.py

@@ -1,13 +1,9 @@
-from __future__ import annotations
-
-from typing import Optional
-
 from puremagic.scanners.helpers import Match
 
 match_bytes = b"%PDF"
 
 
-def main(_, head: bytes, foot: bytes) -> Optional[Match]:
+def main(_, head: bytes, foot: bytes) -> Match | None:
     if b"%PDF-" in head and b"startxref" in foot:
         return Match(".pdf", "PDF document", "application/pdf")
     return None

puremagic/scanners/python_scanner.py

@@ -0,0 +1,25 @@
+import ast
+import os
+
+from puremagic.scanners.helpers import Match
+
+
+def main(file_path: os.PathLike | str, *_, **__) -> Match | None:
+    file_size = os.path.getsize(file_path)
+    if file_size > 1_000_000:
+        return None
+    if not str(file_path).endswith(".py") and file_size < 100:
+        return None
+
+    try:
+        with open(file_path, "r") as file:
+            content = file.read()
+        ast.parse(content)
+    except Exception:
+        return None
+    return Match(
+        extension=".py",
+        name="Python Script",
+        mime_type="text/x-python",
+        confidence=1.0,
+    )

puremagic/scanners/text_scanner.py

@@ -1,8 +1,5 @@
-from __future__ import annotations
-
 import os
 import re
-from typing import Optional
 
 from puremagic.scanners.helpers import Match
 
@@ -11,7 +8,7 @@
 cr_pattern = re.compile(rb"\r(?!\n)")
 
 
-def main(file_path: os.PathLike | str, _, __) -> Optional[Match]:
+def main(file_path: os.PathLike | str, _, __) -> Match | None:
     with open(file_path, "rb") as file:
         head = file.read(1_000_000)
         if len(head) < 8:

puremagic/scanners/zip_scanner.py

@@ -1,8 +1,5 @@
-from __future__ import annotations
-
 import re
 import os
-from typing import Optional
 from zipfile import ZipFile
 
 from puremagic.scanners.helpers import Match
@@ -13,7 +10,7 @@
 application_re = re.compile(b"<Application>(.*)</Application>")
 
 
-def open_office_check(internal_files: list[str], zip_file: ZipFile, extension: str | None = None) -> Optional[Match]:
+def open_office_check(internal_files: list[str], zip_file: ZipFile, extension: str | None = None) -> Match | None:
     if "content.xml" not in internal_files:
         return None
     if "mimetype" not in internal_files:
@@ -35,7 +32,7 @@ def open_office_check(internal_files: list[str], zip_file: ZipFile, extension: s
     return None
 
 
-def office_check(internal_files: list[str], zip_file: ZipFile, extension: str | None = None) -> Optional[Match]:
+def office_check(internal_files: list[str], zip_file: ZipFile, extension: str | None = None) -> Match | None:
     if "[Content_Types].xml" not in internal_files:
         return None
     if "docProps/app.xml" not in internal_files:
@@ -107,7 +104,7 @@ def office_check(internal_files: list[str], zip_file: ZipFile, extension: str |
     return None
 
 
-def jar_check(internal_files: list[str], zip_file: ZipFile) -> Optional[Match]:
+def jar_check(internal_files: list[str], zip_file: ZipFile) -> Match | None:
 
     if "META-INF/MANIFEST.MF" not in internal_files:
         return None
@@ -119,21 +116,21 @@ def jar_check(internal_files: list[str], zip_file: ZipFile) -> Optional[Match]:
     return None
 
 
-def apk_check(internal_files: list[str]) -> Optional[Match]:
+def apk_check(internal_files: list[str]) -> Match | None:
     if "META-INF/MANIFEST.MF" not in internal_files:
         return None
     if "AndroidManifest.xml" in internal_files:
         return Match(".apk", "Android Package", "application/vnd.android.package-archive")
     return None
 
 
-def xpi_check(internal_files: list[str], zip_file: ZipFile) -> Optional[Match]:
+def xpi_check(internal_files: list[str], zip_file: ZipFile) -> Match | None:
     if "install.rdf" in internal_files and b"mozilla:install-manifest" in zip_file.read("install.rdf"):
         return Match(".xpi", "Mozilla Firefox Add-on", "application/x-xpinstall")
     return None
 
 
-def fb2_check(internal_files: list[str], zip_file: ZipFile, file_path: os.PathLike) -> Optional[Match]:
+def fb2_check(internal_files: list[str], zip_file: ZipFile, file_path: os.PathLike) -> Match | None:
     if (
         len(internal_files) == 1
         and internal_files[0].endswith(".fb2")
@@ -148,7 +145,7 @@ def fb2_check(internal_files: list[str], zip_file: ZipFile, file_path: os.PathLi
     return None
 
 
-def cbz_check(internal_files: list[str], extension: str) -> Optional[Match]:
+def cbz_check(internal_files: list[str], extension: str) -> Match | None:
     if extension != "cbz":
         return None
     image_extensions = (".jpg", ".jpeg", ".png", ".gif", ".bmp", ".tiff", ".tif")
@@ -158,7 +155,7 @@ def cbz_check(internal_files: list[str], extension: str) -> Optional[Match]:
     return Match(".cbz", "Comic Book Archive", "application/vnd.comicbook+zip")
 
 
-def main(file_path: os.PathLike, _, __) -> Optional[Match]:
+def main(file_path: os.PathLike, _, __) -> Match | None:
     extension = str(file_path).split(".")[-1].lower()
     if extension == "zip" and not str(file_path).endswith(".fb2.zip"):
         return Match(".zip", "ZIP archive", "application/zip")

pyproject.toml

@@ -38,9 +38,11 @@ lint.extend-ignore = [
   "EXE001",
   "F401",
   "F403",
+  "FA102",
   "FBT",
   "FIX002",
   "I001",
+  "INP001",
   "N817",
   "PERF401",
   "PGH003",

test/resources/system/test.json

@@ -0,0 +1 @@
+{"Test": "Script"}

test/resources/system/test.py

@@ -0,0 +1,9 @@
+# Very simple test python file
+
+
+def main():
+    print("Hello World")
+
+
+if __name__ == "__main__":
+    main()