From aea61a6192949d36adff0b369a4fd2c03502441b Mon Sep 17 00:00:00 2001
From: Helmut Grohne <helmut@subdivi.de>
Date: Thu, 9 May 2024 12:06:28 +0200
Subject: add linuxnamespaces.tarutils

Move the generic tar utilities from the chroottar.py example into a
linuxnamespaces module as dealing with tar archives is a fairly common
thing when dealing with namespaces.
---
 linuxnamespaces/tarutils.py | 77 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 77 insertions(+)
 create mode 100644 linuxnamespaces/tarutils.py

(limited to 'linuxnamespaces')

diff --git a/linuxnamespaces/tarutils.py b/linuxnamespaces/tarutils.py
new file mode 100644
index 0000000..c7a065c
--- /dev/null
+++ b/linuxnamespaces/tarutils.py
@@ -0,0 +1,77 @@
+#!/usr/bin/python3
+# Copyright 2024 Helmut Grohne <helmut@subdivi.de>
+# SPDX-License-Identifier: GPL-3
+
+"""Extensions to the tarfile module.
+ * ZstdTarFile extends TarFile to deal with zstd-compressed archives.
+ * get_comptype guesses the compression used for an open TarFile.
+"""
+
+import tarfile
+import typing
+
+
+class ZstdTarFile(tarfile.TarFile):
+    """Subclass of tarfile.TarFile that can read zstd compressed archives."""
+
+    OPEN_METH = {"zst": "zstopen"} | tarfile.TarFile.OPEN_METH
+
+    @classmethod
+    def zstopen(
+        cls,
+        name: str,
+        mode: typing.Literal["r", "w", "x"] = "r",
+        fileobj: typing.BinaryIO | None = None,
+        **kwargs: typing.Any,
+    ) -> tarfile.TarFile:
+        if mode not in ("r", "w", "x"):
+            raise ValueError("mode must be 'r', 'w' or 'x'")
+        openobj: str | typing.BinaryIO = name if fileobj is None else fileobj
+        try:
+            import zstandard
+        except ImportError as err:
+            raise tarfile.CompressionError(
+                "zstandard module not available"
+            ) from err
+        if mode == "r":
+            zfobj = zstandard.open(openobj, "rb")
+        else:
+            zfobj = zstandard.open(
+                openobj,
+                mode + "b",
+                cctx=zstandard.ZstdCompressor(write_checksum=True, threads=-1),
+            )
+        try:
+            tarobj = cls.taropen(name, mode, zfobj, **kwargs)
+        except (OSError, EOFError, zstandard.ZstdError) as exc:
+            zfobj.close()
+            if mode == "r":
+                raise tarfile.ReadError("not a zst file") from exc
+            raise
+        except:
+            zfobj.close()
+            raise
+        # Setting the _extfileobj attribute is important to signal a need to
+        # close this object and thus flush the compressed stream.
+        # Unfortunately, tarfile.pyi doesn't know about it.
+        tarobj._extfileobj = False  # type: ignore
+        return tarobj
+
+
+def get_comptype(tarobj: tarfile.TarFile) -> str:
+    """Return the compression type used to compress the given TarFile."""
+    # The tarfile module does not expose the compression method selected
+    # for open mode "r:*" in any way. We can guess it from the module that
+    # implements the fileobj.
+    compmodule = tarobj.fileobj.__class__.__module__
+    try:
+        return {
+            "bz2": "bz2",
+            "gzip": "gz",
+            "lzma": "xz",
+            "_io": "tar",
+            "zstd": "zst",
+        }[compmodule]
+    except KeyError:
+        # pylint: disable=raise-missing-from  # no value in chaining
+        raise ValueError(f"cannot guess comptype for module {compmodule}")
-- 
cgit v1.2.3