feat: add rpa extractor

src/rpycdec/rpa.py

@@ -0,0 +1,61 @@
+import os
+import pickle
+import zlib
+from io import BufferedIOBase
+
+
+def read_str(data: BufferedIOBase, eol: int = 0x00) -> bytes:
+    content = bytearray()
+    while True:
+        c = data.read(1)
+        if c[0] == eol:
+            break
+        content += c
+    return content
+
+
+def to_start(start: str | None | bytes) -> bytes:
+    if not start:
+        return b""
+    if isinstance(start, bytes):
+        return start
+    return start.encode("latin-1")
+
+
+def extract_rpa(r: BufferedIOBase, dir: str):
+    header = read_str(r).decode("utf-8")
+    lines = header.splitlines()
+    metadata = lines[0].split()
+    magic, metadata = metadata[0], metadata[1:]
+    index_offset, metadata = int(metadata[0], 16), metadata[1:]
+    if magic == "RPA-3.0":
+        key, metadata = int(metadata[0], 16), metadata[1:0]
+    else:
+        raise Exception("magic %s not supported" % magic)
+
+    # read index
+    r.seek(index_offset)
+    index = pickle.loads(zlib.decompress(r.read()))
+    for k, v in index.items():
+        index[k] = [
+            (offset ^ key, dlen ^ key, b"" if len(left) == 0 else to_start(left[0]))
+            for offset, dlen, *left in v
+        ]
+
+    for filename, entries in index.items():
+        data = bytearray()
+        for offset, dlen, start in entries:
+            r.seek(offset)
+            block = r.read(dlen)
+            if start:
+                if block.startswith(start):
+                    block = block[len(start) :]
+                else:
+                    print("Warning: %s does not start with %s" % (filename, start))
+            data += block
+
+        filename = os.path.join(dir, filename)
+        os.makedirs(os.path.dirname(filename), exist_ok=True)
+        with open(filename, "wb") as f:
+            print("extracting: ", filename)
+            f.write(data)