zipora 3.2.0

High-performance Rust implementation providing advanced data structures and compression algorithms with memory safety guarantees. Features LRU page cache, sophisticated caching layer, fiber-based concurrency, real-time compression, secure memory pools, SIMD optimizations, and complete C FFI for migration from C++.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136

[package]
name = "zipora"
version = "3.2.0"
edition = "2024"
rust-version = "1.95"
authors = ["InfiniLabs"]
description = "High-performance Rust implementation providing advanced data structures and compression algorithms with memory safety guarantees. Features LRU page cache, sophisticated caching layer, fiber-based concurrency, real-time compression, secure memory pools, SIMD optimizations, and complete C FFI for migration from C++."
license-file = "LICENSE"
repository = "https://github.com/infinilabs/zipora"
keywords = ["compression", "data-structures", "trie", "succinct", "concurrency"]
categories = ["compression", "data-structures", "algorithms", "concurrency", "memory-management"]

[lib]
name = "zipora"
crate-type = ["rlib"]

[features]
default = ["simd", "mmap", "zstd", "serde", "lz4", "async", "avx512"]
simd = []
# AVX-512 code paths (runtime-detected, safe on all x86_64 CPUs)
# Disable to reduce binary size on platforms without AVX-512 hardware
avx512 = ["simd"]
mmap = ["dep:memmap2"]
zstd = ["dep:zstd"]
lz4 = ["dep:lz4_flex"]
ffi = ["dep:cbindgen"]
serde = ["dep:serde", "dep:serde_json", "dep:bincode"]
async = ["dep:tokio"]
criterion = []

[dependencies]
crossbeam-epoch = "0.9"
# Core functionality
memmap2 = { version = "0.9.9", optional = true }
bytemuck = "1.24"
thiserror = "2.0"
log = "0.4"

# Performance
ahash = "0.8.12"
rayon = "1.11"
thread_local = "1.1"
fastrand = "2.1"

# Compression (optional)
zstd = { version = "0.13.3", optional = true }
lz4_flex = { version = "0.13", optional = true }

# Serialization (optional)
serde = { version = "1.0", features = ["derive"], optional = true }
serde_json = { version = "1.0", optional = true }
bincode = { version = "1.3", optional = true }
bitflags = "2.10"

libc = "0.2"

# Async runtime (optional)
tokio = { version = "1.43", features = ["full"], optional = true }

# System utilities dependencies
raw-cpuid = "11"
base64 = "0.22"

# FFI support (optional)
cbindgen = { version = "0.29", optional = true }

[dev-dependencies]
criterion = { version = "0.8", features = ["html_reports"] }
proptest = "1.9"
tempfile = "3.14"
arbitrary = { version = "1.4", features = ["derive"] }
rand = "0.9"
# Note: rayon is already in main dependencies

[build-dependencies]
cbindgen = { version = "0.29", optional = true }

[[example]]
name = "production_benchmark"
path = "examples/production_benchmark.rs"

[profile.release]
opt-level = 3           # Maximum optimization level
lto = "thin"           # Thin LTO: ~95% of fat LTO quality, much faster link
codegen-units = 16     # Parallel codegen — use all cores during compilation
panic = "unwind"       # Unwind for test compatibility
strip = false          # Keep debug symbols for tests
overflow-checks = false # Disable overflow checks in release for performance
debug-assertions = false # Disable debug assertions in release

# Shipping profile: use for final binaries (`cargo build --profile shipping`)
# Keeps codegen-units=1 + fat LTO for maximum runtime performance.
[profile.shipping]
inherits = "release"
lto = "fat"            # Full link-time optimization
codegen-units = 1      # Single codegen unit for best optimization
strip = true           # Strip symbols for smaller binary

[profile.bench]
opt-level = 3           # Maximum optimization for benchmarks
lto = "thin"           # Thin LTO: fast link, accurate benchmarks
codegen-units = 8      # Parallel codegen
overflow-checks = false # Optimize for performance
debug = false          # Remove debug info for accurate benchmarks
debug-assertions = false # No debug assertions in benchmarks

[profile.test]
opt-level = 1           # Moderate optimization: fast compile, reasonable test speed
debug = true           # Enable debug info for test failures
debug-assertions = true # Enable debug assertions in tests
overflow-checks = true  # Enable overflow checks for safety in tests

[package.metadata.docs.rs]
features = ["simd", "mmap", "zstd", "lz4", "serde", "ffi"]
all-features = true

# Miri configuration for enhanced memory safety testing
[package.metadata.miri]
# Flags for comprehensive memory safety verification
flags = [
    "-Zmiri-strict-provenance",      # Strict pointer provenance checking
    "-Zmiri-symbolic-alignment-check", # Check alignment symbolically  
    "-Zmiri-check-number-validity",   # Check for invalid bit patterns
    "-Zmiri-disable-isolation",       # Allow file system access for tests
]
# Exclude tests that require specific hardware features not available in Miri
exclude = [
    "tests/simd_specific_tests.rs",
    "tests/avx512_tests.rs", 
    "benchmarks/*",
]


[[bench]]
name = "avx512_bench"
harness = false