an efficient binary archive format
add verification step to tests
+42
-1
+1
-1
README.md
+1
-1
README.md
···
1
1
# bindle-file
2
2
3
-
[bindle](https://en.wikipedia.org/wiki/Bindle) is an efficient, general purpose binary archive format for collecting files.
3
+
[bindle](https://en.wikipedia.org/wiki/Bindle) is an fast/efficient, binary archive format.
4
4
5
5
The format uses memory-mapped I/O for fast reads, optional zstd compression, and supports append-only writes with shadowing for updates. Files can be added incrementally without rewriting the entire archive.
6
6
+41
bench/bench.py
+41
bench/bench.py
···
37
37
return f"{seconds:.3f} s"
38
38
39
39
40
+
def verify_extraction(src_dir: Path, extract_dir: Path) -> None:
41
+
"""Verify extracted files match source files."""
42
+
src_files = {f.relative_to(src_dir): f for f in src_dir.rglob("*") if f.is_file()}
43
+
extract_files = {f.relative_to(extract_dir): f for f in extract_dir.rglob("*") if f.is_file()}
44
+
45
+
# Check file count
46
+
if len(src_files) != len(extract_files):
47
+
raise ValueError(f"File count mismatch: {len(src_files)} source, {len(extract_files)} extracted")
48
+
49
+
# Check each file exists and has correct size
50
+
for rel_path, src_file in src_files.items():
51
+
if rel_path not in extract_files:
52
+
raise ValueError(f"Missing file in extraction: {rel_path}")
53
+
54
+
extract_file = extract_files[rel_path]
55
+
src_size = src_file.stat().st_size
56
+
extract_size = extract_file.stat().st_size
57
+
58
+
if src_size != extract_size:
59
+
raise ValueError(f"Size mismatch for {rel_path}: {src_size} vs {extract_size}")
60
+
61
+
# Verify content matches
62
+
if src_file.read_bytes() != extract_file.read_bytes():
63
+
raise ValueError(f"Content mismatch for {rel_path}")
64
+
65
+
40
66
def create_test_data(base_dir: Path) -> None:
41
67
"""Create a variety of test files."""
42
68
base_dir.mkdir(parents=True, exist_ok=True)
···
90
116
)
91
117
unpack_time = time.perf_counter() - start
92
118
119
+
# Verify extraction (not timed)
120
+
verify_extraction(src_dir, extract_dir)
121
+
93
122
return pack_time, size, unpack_time
94
123
95
124
···
119
148
)
120
149
unpack_time = time.perf_counter() - start
121
150
151
+
# Verify extraction (not timed)
152
+
verify_extraction(src_dir, extract_dir)
153
+
122
154
return pack_time, size, unpack_time
123
155
124
156
···
146
178
)
147
179
unpack_time = time.perf_counter() - start
148
180
181
+
# Verify extraction (not timed)
182
+
verify_extraction(src_dir, extract_dir)
183
+
149
184
return pack_time, size, unpack_time
150
185
151
186
···
173
208
)
174
209
unpack_time = time.perf_counter() - start
175
210
211
+
# Verify extraction (not timed)
212
+
verify_extraction(src_dir, extract_dir)
213
+
176
214
return pack_time, size, unpack_time
177
215
178
216
···
199
237
check=True,
200
238
)
201
239
unpack_time = time.perf_counter() - start
240
+
241
+
# Verify extraction (not timed)
242
+
verify_extraction(src_dir, extract_dir)
202
243
203
244
return pack_time, size, unpack_time
204
245