feat(handler): support NSIS Installers

docs/handlers.md

@@ -46,6 +46,7 @@
     | [`NETGEAR TRX V2`](#netgear-trx-v2) | ARCHIVE | :octicons-check-16: |
     | [`NTFS`](#ntfs) | FILESYSTEM | :octicons-check-16: |
     | [`PARTCLONE`](#partclone) | ARCHIVE | :octicons-check-16: |
+    | [`PE`](#pe) | EXECUTABLE | :octicons-check-16: |
     | [`QNAP NAS`](#qnap-nas) | ARCHIVE | :octicons-check-16: |
     | [`RAR`](#rar) | ARCHIVE | :octicons-alert-fill-12: |
     | [`ROMFS`](#romfs) | FILESYSTEM | :octicons-check-16: |
@@ -810,6 +811,23 @@
 
         - [Partclone GitHub Repository](https://github.com/Thomas-Tsai/partclone){ target="_blank" }
         - [Clonezilla Official Documentation](https://clonezilla.org/){ target="_blank" }
+## pe
+
+!!! success "Fully supported"
+
+    === "Description"
+
+        The PE (Portable Executable) is a binary file format used for executable code on 32-bit and 64-bit Windows operating systems as well as in UEFI environments.
+
+        ---
+
+        - **Handler type:** Executable
+        - **Vendor:** Microsoft
+
+    === "References"
+
+        - [PE Format](https://learn.microsoft.com/en-us/windows/win32/debug/pe-format){ target="_blank" }
+        - [Portable Executable Wikipedia](https://en.wikipedia.org/wiki/Portable_Executable){ target="_blank" }
 ## QNAP NAS
 
 !!! success "Fully supported"

python/unblob/handlers/__init__.py

@@ -37,7 +37,7 @@
     zlib,
     zstd,
 )
-from .executable import elf
+from .executable import elf, pe
 from .filesystem import (
     cramfs,
     extfs,
@@ -115,6 +115,7 @@
     zstd.ZSTDHandler,
     elf.ELF32Handler,
     elf.ELF64Handler,
+    pe.PEHandler,
     zlib.ZlibHandler,
     engenius.EngeniusHandler,
     ecc.AutelECCHandler,

python/unblob/handlers/executable/pe.py

@@ -0,0 +1,104 @@
+import io
+from pathlib import Path
+from typing import Optional
+
+import lief
+from structlog import get_logger
+
+from unblob.extractors.command import Command
+
+from ...models import (
+    Extractor,
+    ExtractResult,
+    File,
+    Handler,
+    HandlerDoc,
+    HandlerType,
+    HexString,
+    Reference,
+    ValidChunk,
+)
+
+lief.logging.disable()
+
+logger = get_logger()
+
+
+class PEExtractor(Extractor):
+    def extract(self, inpath: Path, outdir: Path) -> Optional[ExtractResult]:
+        binary = lief.PE.parse(inpath)
+        if binary and self.is_nsis(binary):
+            return Command("7z", "x", "-y", "{inpath}", "-o{outdir}").extract(
+                inpath, outdir
+            )
+        return None
+
+    def is_nsis(self, binary: lief.PE.Binary) -> bool:
+        # Test if binary appears to be a Nullsoft Installer self-extracting archive
+        # see https://github.com/file/file/blob/7ed3febfcd616804a2ec6495b3e5f9ccb6fc5f8f/magic/Magdir/msdos#L383
+
+        if binary.has_resources:
+            resource_manager = binary.resources_manager
+            if (
+                isinstance(resource_manager, lief.PE.ResourcesManager)
+                and resource_manager.has_manifest
+            ):
+                manifest = (
+                    resource_manager.manifest
+                    if isinstance(resource_manager.manifest, str)
+                    else resource_manager.manifest.decode(errors="ignore")
+                )
+                if "Nullsoft.NSIS.exehead" in manifest:
+                    return True
+        return False
+
+
+class PEHandler(Handler):
+    NAME = "pe"
+
+    PATTERNS = [
+        HexString(
+            """
+            // MZ header
+            4d 5a
+            """
+        ),
+        HexString(
+            """
+            // PE header
+            50 45 00 00
+            """
+        ),
+    ]
+
+    EXTRACTOR = PEExtractor()
+
+    DOC = HandlerDoc(
+        name="pe",
+        description="The PE (Portable Executable) is a binary file format used for executable code on 32-bit and 64-bit Windows operating systems as well as in UEFI environments.",
+        handler_type=HandlerType.EXECUTABLE,
+        vendor="Microsoft",
+        references=[
+            Reference(
+                title="PE Format",
+                url="https://learn.microsoft.com/en-us/windows/win32/debug/pe-format",
+            ),
+            Reference(
+                title="Portable Executable Wikipedia",
+                url="https://en.wikipedia.org/wiki/Portable_Executable",
+            ),
+        ],
+        limitations=[],
+    )
+
+    def calculate_chunk(self, file: File, start_offset: int) -> Optional[ValidChunk]:
+        file.seek(start_offset, io.SEEK_SET)
+
+        binary = lief.PE.parse(file[start_offset:])
+        if not binary:
+            return None
+
+        return ValidChunk(
+            start_offset=start_offset,
+            end_offset=start_offset + binary.original_size,
+        )