archae 2026.1.0__tar.gz → 2026.1.0b2__tar.gz

{archae-2026.1.0 → archae-2026.1.0b2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: archae
-Version: 2026.1.0
+Version: 2026.1.0b2
 Summary: Archae explodes archives.
 Keywords: 
 Author: Shawn McNaughton
@@ -65,6 +65,25 @@ Archae explodes archives.
 
 <!-- end docs-include-index -->
 
+## Why
+
+Every once and a while, I run into an issue: multiple layers of archives. The reasons vary, but examples would include:
+
+- Searching for something in a ZIP of folders that contained a ZIP with a CD image in it
+- Running a malware scan and finding an obscure archive format was missed and not even flagged
+  Meanwhile, I want to make sure I don't fill my disk, especially if an archive bomb (more commonly known as a ZIP bomb) has been jammed in somewhere. They're only funny the first time. :D
+
+## Features
+
+- Uses 7z/peazip/unar (not 7za/7zr) to try to extract archives
+- No substantial limit on the number of archive layers
+- Identifies file types via libmagic
+- Detects duplicate archives
+- Basic archive bomb protections
+    - min_archive_size_bytes - ensures the uncompressed size of an archive is limited
+    - min_total_size_bytes - ensures the total extracted footprint isn't above a certain size
+    - min_archive_ratio - ensures very-high-compression-ratio archives are stopped
+
 ## Installation
 
 <!-- start docs-include-installation -->
@@ -129,19 +148,15 @@ archae --help
 
 <!-- end docs-include-usage -->
 
-## Features
-
-- Uses 7z/peazip/unar (not 7za/7zr) to try to extract archives
-- No substantial limit on the number of archive layers
-- Identifies file types via libmagic
-- Basic archive bomb protections
-
 ## TODOs
 
-- Protect against/detect archive bombs (malicious disk usage)
+- More archive bomb protections
+    - min_total_size_bytes - (NYI) ensures the total extracted footprint isn't above a certain size
+    - min_free_space - minimum free space at the extraction location
+    - delete_archives_as_exploded - remove archive files to reduce duplication (boolean)
+    - max_archive_depth - allow setting a maximum archive depth
 - Improve archive type detection
 - Separate between extractable and non-extractable archive types
-- Protect against/detect depth attacks (excessively nested archives)
 - Detect password-protected archives
 - Allow supplying archive passwords by hash
 - Add custom magic to detect obscure archive formats

{archae-2026.1.0 → archae-2026.1.0b2}/README.md

@@ -15,6 +15,25 @@ Archae explodes archives.
 
 <!-- end docs-include-index -->
 
+## Why
+
+Every once and a while, I run into an issue: multiple layers of archives. The reasons vary, but examples would include:
+
+- Searching for something in a ZIP of folders that contained a ZIP with a CD image in it
+- Running a malware scan and finding an obscure archive format was missed and not even flagged
+  Meanwhile, I want to make sure I don't fill my disk, especially if an archive bomb (more commonly known as a ZIP bomb) has been jammed in somewhere. They're only funny the first time. :D
+
+## Features
+
+- Uses 7z/peazip/unar (not 7za/7zr) to try to extract archives
+- No substantial limit on the number of archive layers
+- Identifies file types via libmagic
+- Detects duplicate archives
+- Basic archive bomb protections
+    - min_archive_size_bytes - ensures the uncompressed size of an archive is limited
+    - min_total_size_bytes - ensures the total extracted footprint isn't above a certain size
+    - min_archive_ratio - ensures very-high-compression-ratio archives are stopped
+
 ## Installation
 
 <!-- start docs-include-installation -->
@@ -79,19 +98,15 @@ archae --help
 
 <!-- end docs-include-usage -->
 
-## Features
-
-- Uses 7z/peazip/unar (not 7za/7zr) to try to extract archives
-- No substantial limit on the number of archive layers
-- Identifies file types via libmagic
-- Basic archive bomb protections
-
 ## TODOs
 
-- Protect against/detect archive bombs (malicious disk usage)
+- More archive bomb protections
+    - min_total_size_bytes - (NYI) ensures the total extracted footprint isn't above a certain size
+    - min_free_space - minimum free space at the extraction location
+    - delete_archives_as_exploded - remove archive files to reduce duplication (boolean)
+    - max_archive_depth - allow setting a maximum archive depth
 - Improve archive type detection
 - Separate between extractable and non-extractable archive types
-- Protect against/detect depth attacks (excessively nested archives)
 - Detect password-protected archives
 - Allow supplying archive passwords by hash
 - Add custom magic to detect obscure archive formats

{archae-2026.1.0 → archae-2026.1.0b2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "archae"
-version = "2026.1.0"
+version = "2026.1.0b2"
 description = "Archae explodes archives."
 authors = [{name = "Shawn McNaughton", email = "shawngmc@gmail.com"}]
 readme = "README.md"

{archae-2026.1.0 → archae-2026.1.0b2}/src/archae/cli.py

@@ -64,28 +64,19 @@ class FileSizeParamType(click.ParamType):
             pass
 
         # Regex to split number and unit
-        match = re.match(r"^(\d+(?:\.\d+)?)\s*([KMGT]B?)$", str(value), re.IGNORECASE)
+        match = re.match(r"^(\d+(?:\.\d+)?)\s*([KMGTP]B?)$", str(value), re.IGNORECASE)
         if not match:
             msg = f"{value} is not a valid file size (e.g., 10G, 500M)"
             raise ValueError(msg)
 
         number, unit = match.groups()
         number = float(number)
-        unit = unit.upper()
-
-        units = {
-            "K": 1024,
-            "KB": 1024,
-            "M": 1024**2,
-            "MB": 1024**2,
-            "G": 1024**3,
-            "GB": 1024**3,
-            "T": 1024**4,
-            "TB": 1024**4,
-        }
+        unit = unit[0].upper()
+
+        byte_scale = 1024 ** (ByteScale.from_prefix_letter(unit).value)
 
         # Default to bytes if no specific unit multiplier, or assume B
-        return int(number * units.get(unit, 1))
+        return int(number * byte_scale)
 
     def convert(self, value: click.Option, param: str, ctx: click.Context) -> int:
         """Convert a FileSizeParam to an int.
@@ -211,6 +202,13 @@ def handle_file(file_path: Path) -> None:
                 click.echo(
                     f"Skipped archive {file_path} because expected size {extracted_size} is greater than max_archive_size_bytes {config['max_archive_size_bytes']}"
                 )
+            elif (
+                get_tracked_file_size() + extracted_size
+                > config["max_total_size_bytes"]
+            ):
+                click.echo(
+                    f"Skipped archive {file_path} because expected size {extracted_size} + current tracked files {get_tracked_file_size()} is greater than max_total_size_bytes {config['max_total_size_bytes']}"
+                )
             elif compression_ratio < config["min_archive_ratio"]:
                 click.echo(
                     f"Skipped archive {file_path} because compression ratio {compression_ratio:.5f} is less than min_archive_ratio {config['min_archive_ratio']}"
@@ -372,3 +370,12 @@ def add_metadata_to_hash(hash: str, key: str, value: Any) -> None:
         value (Any): The metadata value.
     """
     tracked_files[hash]["metadata"][key] = value
+
+
+def get_tracked_file_size() -> int:
+    """Get the total size of all tracked files.
+
+    Returns:
+        int: The total size in bytes.
+    """
+    return sum(tracked_files[hash].get("size", 0) for hash in tracked_files)

{archae-2026.1.0 → archae-2026.1.0b2}/src/archae/util/enum/byte_scale.py

@@ -44,3 +44,12 @@ class ByteScale(Enum):
     def prefix_letter(self, value: str) -> None:
         """Setter for prefix letter."""
         self._prefix_letter = value
+
+    @staticmethod
+    def from_prefix_letter(prefix_letter: str) -> Self:  # type: ignore[misc]
+        """Static method to look up from a prefix_letter."""
+        for member in ByteScale:
+            if member.prefix_letter == prefix_letter.upper():
+                return member
+        msg = f"'{prefix_letter}' is not a valid byte scale prefix letter."
+        raise ValueError(msg)