Basic Compression

Compresses a string and reads the header back.

zig build run-basic_compression

Source: examples/basic_compression.zig

The program

const std = @import("std");
const bzip2 = @import("bzip2");

pub fn main(init: std.process.Init) !void {
    const gpa = init.gpa;

    const sentence = "The quick brown fox jumps over the lazy dog. ";
    var input: [sentence.len * 20]u8 = undefined;
    for (&input, 0..) |*byte, i| byte.* = sentence[i % sentence.len];

    const compressed = try bzip2.compress(gpa, &input, .{});
    defer gpa.free(compressed);

    std.debug.print("input:      {d} bytes\n", .{input.len});
    std.debug.print("compressed: {d} bytes\n", .{compressed.len});
    std.debug.print("ratio:      {d:.2}x\n", .{
        @as(f64, @floatFromInt(input.len)) / @as(f64, @floatFromInt(compressed.len)),
    });
    std.debug.print("block size: {d} bytes\n", .{try bzip2.blockSizeOfStream(compressed)});
}

Output

input:      900 bytes
compressed: 124 bytes
ratio:      7.26x
block size: 900000 bytes

What to notice

The block size is 900,000 because omitting options uses level 9. The same call with .{ .level = 1 } reports 100,000.

Both buffers belong to the caller. defer gpa.free releases the compressed slice; the input array is on the stack and needs nothing.

blockSizeOfStream reads the header without decoding, so it costs nothing beyond the first four bytes.

Related

• Compression
• Decoder Semantics shows the round trip with statistics
• compression_levels.zig prints the whole level table