gh-115988: Add missing ARM64 and RISCV filter in lzma module

Doc/library/lzma.rst

@@ -322,6 +322,25 @@ Miscellaneous
    feature set.
 
 
+Information about the version of the lzma library in use is available through
+the following constants:
+
+
+.. data:: LZMA_VERSION
+
+   The version string of the lzma library that was used for building the module.
+   This may be different from the lzma library actually used at runtime, which
+   is available as :const:`LZMA_RUNTIME_VERSION`.
+
+   .. versionadded:: 3.13
+
+
+.. data:: LZMA_RUNTIME_VERSION
+
+   The version string of the lzma library actually loaded by the interpreter.
+
+   .. versionadded:: 3.13
+
 .. _filter-chain-specs:
 
 Specifying custom filter chains
@@ -343,12 +362,26 @@ options. Valid filter IDs are as follows:
 
 * Branch-Call-Jump (BCJ) filters:
 
-  * :const:`FILTER_X86`
-  * :const:`FILTER_IA64`
-  * :const:`FILTER_ARM`
-  * :const:`FILTER_ARMTHUMB`
-  * :const:`FILTER_POWERPC`
-  * :const:`FILTER_SPARC`
+  * :const:`!FILTER_X86`
+  * :const:`!FILTER_IA64`
+  * :const:`!FILTER_ARM`
+  * :const:`!FILTER_ARMTHUMB`
+  * :const:`!FILTER_POWERPC`
+  * :const:`!FILTER_SPARC`
+
+    The above work on all lzma runtime library versions.
+
+  * :const:`!FILTER_ARM64`
+
+    Only works if :data:`LZMA_RUNTIME_VERSION` is 5.4.0 or later.
+
+    .. versionadded:: 3.13
+
+  * :const:`!FILTER_RISCV`
+
+    Only works if :data:`LZMA_RUNTIME_VERSION` is 5.6.0 or later.
+
+    .. versionadded:: 3.13
 
 A filter chain can consist of up to 4 filters, and cannot be empty. The last
 filter in the chain must be a compression filter, and any other filters must be

Doc/whatsnew/3.13.rst

@@ -329,6 +329,20 @@ itertools
   than the specified batch size.
   (Contributed by Raymond Hettinger in :gh:`113202`.)
 
+lzma
+----
+
+* Add :const:`lzma.LZMA_VERSION` and :const:`lzma.LZMA_RUNTIME_VERSION`. The former
+  reports the version string of the underlying ``lzma`` library during build while
+  the later reports runtime version.
+  (Contributed by Chien Wong in :gh:`115988`.)
+
+* Add support of new BCJ filters ARM64 and RISC-V via :const:`!lzma.FILTER_ARM64` and
+  :const:`!lzma.FILTER_RISCV`.
+  Note that the new filters will work only if runtime library supports them. ARM64 filter
+  requires ``lzma`` 5.4.0 or newer while RISC-V requires 5.6.0 or newer.
+  (Contributed by Chien Wong in :gh:`115988`.)
+
 marshal
 -------
 

Lib/lzma.py

@@ -13,6 +13,7 @@
     "CHECK_ID_MAX", "CHECK_UNKNOWN",
     "FILTER_LZMA1", "FILTER_LZMA2", "FILTER_DELTA", "FILTER_X86", "FILTER_IA64",
     "FILTER_ARM", "FILTER_ARMTHUMB", "FILTER_POWERPC", "FILTER_SPARC",
+    "FILTER_ARM64", "FILTER_RISCV",
     "FORMAT_AUTO", "FORMAT_XZ", "FORMAT_ALONE", "FORMAT_RAW",
     "MF_HC3", "MF_HC4", "MF_BT2", "MF_BT3", "MF_BT4",
     "MODE_FAST", "MODE_NORMAL", "PRESET_DEFAULT", "PRESET_EXTREME",

Lib/test/test_lzma.py

@@ -5,6 +5,7 @@
 import pathlib
 import pickle
 import random
+import re
 import sys
 from test import support
 import unittest
@@ -19,6 +20,16 @@
 from lzma import LZMACompressor, LZMADecompressor, LZMAError, LZMAFile
 
 
+def _lzma_runtime_version_tuple(lzma_version=lzma.LZMA_RUNTIME_VERSION):
+    v = lzma_version.split('.', 3)
+    if not v[-1].isnumeric():
+        v[-1] = re.search(r'\d+', v[-1]).group()
+    return tuple(map(int, v))
+
+
+LZMA_RUNTIME_VERSION_TUPLE = _lzma_runtime_version_tuple()
+
+
 class CompressorDecompressorTestCase(unittest.TestCase):
 
     # Test error cases.
@@ -384,6 +395,21 @@ def test_uninitialized_LZMADecompressor_crash(self):
         self.assertEqual(LZMADecompressor.__new__(LZMADecompressor).
                          decompress(bytes()), b'')
 
+    # Test the existence of the relatively new BCJ filters.  These just
+    # ensure that the constant was found at compile time and exposed.
+
+    @unittest.skipUnless(
+            LZMA_RUNTIME_VERSION_TUPLE >= (5, 6, 0),
+            "RISC-V filter is only available on lzma 5.6.0 or later")
+    def test_riscv_filter_exists(self):
+        self.assertTrue(lzma.FILTER_RISCV)
+
+    @unittest.skipUnless(
+            LZMA_RUNTIME_VERSION_TUPLE >= (5, 4, 0),
+            "ARM64 filter is only available on lzma 5.4.0 or later")
+    def test_arm64_filter_exists(self):
+        self.assertTrue(lzma.FILTER_ARM64)
+
 
 class CompressDecompressFunctionTestCase(unittest.TestCase):
 

Misc/NEWS.d/next/Library/2024-03-17-22-16-53.gh-issue-115988.EshSDM.rst

@@ -0,0 +1,2 @@
+The lzma module adds build and runtime version info of the underlying C library
+and supports new BCJ filters: ARM64 and RISC-V.

Modules/_lzmamodule.c

@@ -24,6 +24,20 @@
     #error "The maximum block size accepted by liblzma is SIZE_MAX."
 #endif
 
+
+/*
+ * If the lzma.h we're building against is so old as not to define these, this
+ * provides their equivalent values so that the names remain defined in Python
+ * regardless.  lzma.LZMA_RUNTIME_VERSION is exposed to Python and is what
+ * people can use to decide if they can use them at runtime.
+ */
+#ifndef LZMA_FILTER_ARM64
+#define LZMA_FILTER_ARM64       LZMA_VLI_C(0x0A)
+#endif
+#ifndef LZMA_FILTER_RISCV
+#define LZMA_FILTER_RISCV       LZMA_VLI_C(0x0B)
+#endif
+
 /* On success, return value >= 0
    On failure, return -1 */
 static inline Py_ssize_t
@@ -372,6 +386,8 @@ lzma_filter_converter(_lzma_state *state, PyObject *spec, void *ptr)
         case LZMA_FILTER_ARM:
         case LZMA_FILTER_ARMTHUMB:
         case LZMA_FILTER_SPARC:
+        case LZMA_FILTER_ARM64:
+        case LZMA_FILTER_RISCV:
             f->options = parse_filter_spec_bcj(state, spec);
             return f->options != NULL;
         default:
@@ -490,7 +506,9 @@ build_filter_spec(const lzma_filter *f)
         case LZMA_FILTER_IA64:
         case LZMA_FILTER_ARM:
         case LZMA_FILTER_ARMTHUMB:
-        case LZMA_FILTER_SPARC: {
+        case LZMA_FILTER_SPARC:
+        case LZMA_FILTER_ARM64:
+        case LZMA_FILTER_RISCV: {
             lzma_options_bcj *options = f->options;
             if (options) {
                 ADD_FIELD(options, start_offset);
@@ -1551,6 +1569,8 @@ lzma_exec(PyObject *module)
     ADD_INT_PREFIX_MACRO(module, FILTER_ARMTHUMB);
     ADD_INT_PREFIX_MACRO(module, FILTER_SPARC);
     ADD_INT_PREFIX_MACRO(module, FILTER_POWERPC);
+    ADD_INT_PREFIX_MACRO(module, FILTER_ARM64);
+    ADD_INT_PREFIX_MACRO(module, FILTER_RISCV);
     ADD_INT_PREFIX_MACRO(module, MF_HC3);
     ADD_INT_PREFIX_MACRO(module, MF_HC4);
     ADD_INT_PREFIX_MACRO(module, MF_BT2);
@@ -1591,6 +1611,15 @@ lzma_exec(PyObject *module)
         return -1;
     }
 
+    if (PyModule_Add(module, "LZMA_VERSION",
+                     PyUnicode_FromString(LZMA_VERSION_STRING)) < 0) {
+        return -1;
+    }
+    if (PyModule_Add(module, "LZMA_RUNTIME_VERSION",
+                     PyUnicode_FromString(lzma_version_string())) < 0) {
+        return -1;
+    }
+
     return 0;
 }