dissect.target 3.18.dev2__py3-none-any.whl → 3.18.dev3__py3-none-any.whl

dissect/target/plugins/os/unix/locate/plocate.py

@@ -1,8 +1,7 @@
 from __future__ import annotations
 
 import platform
-from io import BytesIO
-from typing import BinaryIO, Iterable
+from typing import BinaryIO, Iterator
 
 from dissect.cstruct import cstruct
 from dissect.util.stream import RangeStream
@@ -13,7 +12,11 @@ from dissect.target.plugin import export
 from dissect.target.plugins.os.unix.locate.locate import BaseLocatePlugin
 
 try:
-    import zstandard  # noqa
+    from zstandard import (
+        DECOMPRESSION_RECOMMENDED_OUTPUT_SIZE,
+        ZstdCompressionDict,
+        ZstdDecompressor,
+    )
 
     HAS_ZSTD = True
 except ImportError:
@@ -32,7 +35,7 @@ struct header {
     uint64_t filename_index_offset_bytes;
 
     /* Version 1 and up only. */
-    uint32_t max_version;
+    uint32_t max_version;   // Nominally 1 or 2, but can be increased if more features are added in a backward-compatible way.
     uint32_t zstd_dictionary_length_bytes;
     uint64_t zstd_dictionary_offset_bytes;
 
@@ -44,6 +47,7 @@ struct header {
     uint64_t conf_block_length_bytes;
     uint64_t conf_block_offset_bytes;
 
+    // Only if max_version >= 2.
     uint8_t check_visibility;
     char padding[7];                         /* padding for alignment */
 };
@@ -51,7 +55,7 @@ struct header {
 struct file {
     char path[];
 };
-"""
+"""  # noqa : E501
 
 PLocateRecord = TargetRecordDescriptor(
     "linux/locate/plocate",
@@ -104,40 +108,46 @@ class PLocateFile:
         self.dict_data = None
 
         if self.header.zstd_dictionary_offset_bytes:
-            self.dict_data = zstandard.ZstdCompressionDict(self.fh.read(self.header.zstd_dictionary_length_bytes))
+            self.dict_data = ZstdCompressionDict(self.fh.read(self.header.zstd_dictionary_length_bytes))
 
         self.compressed_length_bytes = (
             self.header.filename_index_offset_bytes - self.HEADER_SIZE - self.header.zstd_dictionary_length_bytes
         )
-        self.ctx = zstandard.ZstdDecompressor(dict_data=self.dict_data)
+        self.ctx = ZstdDecompressor(dict_data=self.dict_data)
         self.buf = RangeStream(self.fh, self.fh.tell(), self.compressed_length_bytes)
 
-    def __iter__(self) -> Iterable[PLocateFile]:
+    def __iter__(self) -> Iterator[PLocateFile]:
         # NOTE: This is a workaround for a PyPy bug
         # We don't know what breaks, but PyPy + zstandard = unhappy times
         # You just get random garbage data back instead of the decompressed data
         # This weird dance of using a decompressobj and unused data is the only way that seems to work
         # It's more expensive on memory, but at least it doesn't break
         if platform.python_implementation() == "PyPy":
-            obj = self.ctx.decompressobj()
             buf = self.buf.read()
 
-            tmp = obj.decompress(buf)
-            while unused_data := obj.unused_data:
-                obj = self.ctx.decompressobj()
-                tmp += obj.decompress(unused_data)
+            def reader(ctx: ZstdDecompressor) -> Iterator[bytes]:
+                obj = ctx.decompressobj()
 
-            reader = BytesIO(tmp)
+                yield obj.decompress(buf)
+                while unused_data := obj.unused_data:
+                    obj = self.ctx.decompressobj()
+                    yield obj.decompress(unused_data)
+
+            it = reader(self.ctx)
         else:
-            reader = self.ctx.stream_reader(self.buf)
-
-        with reader:
-            try:
-                while True:
-                    file = c_plocate.file(reader)
-                    yield file.path.decode(errors="surrogateescape")
-            except EOFError:
-                return
+            # NOTE: The end of a zstandard frame does not include a final `0x00`.
+            # This causes the c_plocate `file` struct to parse the last path and the first path on the next frame as one
+            # since cstruct will read it across frame boundaries waiting for a `0x00`.
+            def reader() -> Iterator[bytes]:
+                with self.ctx.stream_reader(self.buf) as reader:
+                    while chunk := reader.read(DECOMPRESSION_RECOMMENDED_OUTPUT_SIZE):
+                        yield chunk
+
+            it = reader()
+
+        for chunk in it:
+            for path in chunk.split(b"\x00"):
+                yield path.decode(errors="surrogateescape")
 
     def filename_index(self) -> bytes:
         """Return the filename index of the plocate.db file."""

{dissect.target-3.18.dev2.dist-info → dissect.target-3.18.dev3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dissect.target
-Version: 3.18.dev2
+Version: 3.18.dev3
 Summary: This module ties all other Dissect modules together, it provides a programming API and command line tools which allow easy access to various data sources inside disk images or file collections (a.k.a. targets)
 Author-email: Dissect Team <dissect@fox-it.com>
 License: Affero General Public License v3

{dissect.target-3.18.dev2.dist-info → dissect.target-3.18.dev3.dist-info}/RECORD

@@ -243,7 +243,7 @@ dissect/target/plugins/os/unix/locate/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCe
 dissect/target/plugins/os/unix/locate/gnulocate.py,sha256=P-YbMFw901p2EBgTaZH6axShfIRRDrCx3APBy6Ii3lE,2934
 dissect/target/plugins/os/unix/locate/locate.py,sha256=uXFcWAqoz_3eNWHhsGoEtkkhmT5J3F1GYvr4uQxi308,122
 dissect/target/plugins/os/unix/locate/mlocate.py,sha256=DhrFgxDQF-fMZaA0WK8Z-5o9i9iDsuTHW7MHJtWwz6o,4485
-dissect/target/plugins/os/unix/locate/plocate.py,sha256=Skb24ba_MVzM4nuDaZHw-ZmomIEZ3TJ7g5kHCvQViko,6545
+dissect/target/plugins/os/unix/locate/plocate.py,sha256=ShU-F9_31rGfMYXqaR_KrHXVxgDDRZMJ_zEMuekw57w,7229
 dissect/target/plugins/os/unix/log/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dissect/target/plugins/os/unix/log/atop.py,sha256=DdiTf-gVJJvvPbR36khU4388lxQzABaWI-95jLCGgSw,16345
 dissect/target/plugins/os/unix/log/audit.py,sha256=OjorWTmCFvCI5RJq6m6WNW0Lhb-poB2VAggKOGZUHK4,3722
@@ -340,10 +340,10 @@ dissect/target/volumes/luks.py,sha256=OmCMsw6rCUXG1_plnLVLTpsvE1n_6WtoRUGQbpmu1z
 dissect/target/volumes/lvm.py,sha256=wwQVR9I3G9YzmY6UxFsH2Y4MXGBcKL9aayWGCDTiWMU,2269
 dissect/target/volumes/md.py,sha256=j1K1iKmspl0C_OJFc7-Q1BMWN2OCC5EVANIgVlJ_fIE,1673
 dissect/target/volumes/vmfs.py,sha256=-LoUbn9WNwTtLi_4K34uV_-wDw2W5hgaqxZNj4UmqAQ,1730
-dissect.target-3.18.dev2.dist-info/COPYRIGHT,sha256=m-9ih2RVhMiXHI2bf_oNSSgHgkeIvaYRVfKTwFbnJPA,301
-dissect.target-3.18.dev2.dist-info/LICENSE,sha256=DZak_2itbUtvHzD3E7GNUYSRK6jdOJ-GqncQ2weavLA,34523
-dissect.target-3.18.dev2.dist-info/METADATA,sha256=CQlktWeUyWCrkDq9WKqNFR3gxHwKVUZ-9KDPop4n6s8,11299
-dissect.target-3.18.dev2.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-dissect.target-3.18.dev2.dist-info/entry_points.txt,sha256=tvFPa-Ap-gakjaPwRc6Fl6mxHzxEZ_arAVU-IUYeo_s,447
-dissect.target-3.18.dev2.dist-info/top_level.txt,sha256=Mn-CQzEYsAbkxrUI0TnplHuXnGVKzxpDw_po_sXpvv4,8
-dissect.target-3.18.dev2.dist-info/RECORD,,
+dissect.target-3.18.dev3.dist-info/COPYRIGHT,sha256=m-9ih2RVhMiXHI2bf_oNSSgHgkeIvaYRVfKTwFbnJPA,301
+dissect.target-3.18.dev3.dist-info/LICENSE,sha256=DZak_2itbUtvHzD3E7GNUYSRK6jdOJ-GqncQ2weavLA,34523
+dissect.target-3.18.dev3.dist-info/METADATA,sha256=t5yUGDDtnDgTACRs1bUW1xJF2rlVMzxbFW0iWG5WRUw,11299
+dissect.target-3.18.dev3.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+dissect.target-3.18.dev3.dist-info/entry_points.txt,sha256=tvFPa-Ap-gakjaPwRc6Fl6mxHzxEZ_arAVU-IUYeo_s,447
+dissect.target-3.18.dev3.dist-info/top_level.txt,sha256=Mn-CQzEYsAbkxrUI0TnplHuXnGVKzxpDw_po_sXpvv4,8
+dissect.target-3.18.dev3.dist-info/RECORD,,