Peak throughput: 41,000 pages/sec (Intel SDM, parallel)

Build with zig build -Doptimize=ReleaseFast for these results.

Note: MuPDF's threading (-T flag) is for rendering/rasterization only. Text extraction via mutool convert -F text is single-threaded by design. zpdf parallelizes text extraction across pages.

zig build              # Build library and CLI
zig build test         # Run tests

const zpdf = @import("zpdf");

pub fn main() !void {
    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
    defer _ = gpa.deinit();
    const allocator = gpa.allocator();

    const doc = try zpdf.Document.open(allocator, "file.pdf");
    defer doc.close();

    var buf: [4096]u8 = undefined;
    var writer = std.fs.File.stdout().writer(&buf);
    defer writer.interface.flush() catch {};

    for (0..doc.pages.items.len) |page_num| {
        try doc.extractText(page_num, &writer.interface);
    }
}

zpdf extract document.pdf           # Extract all pages to stdout
zpdf extract -p 1-10 document.pdf   # Extract pages 1-10
zpdf extract -o out.txt document.pdf # Output to file
zpdf info document.pdf              # Show document info
zpdf bench document.pdf             # Run benchmark

src/
├── root.zig         # Document API and core types
├── parser.zig       # PDF object parser
├── xref.zig         # XRef table/stream parsing
├── pagetree.zig     # Page tree resolution
├── decompress.zig   # Stream decompression filters
├── encoding.zig     # Font encoding and CMap parsing
├── interpreter.zig  # Content stream interpreter
├── simd.zig         # SIMD string operations
└── main.zig         # CLI

Implemented:

• XRef table and stream parsing
• Incremental PDF updates (follows /Prev chain for modified documents)
• Object parser
• Page tree resolution
• Content stream interpretation (Tj, TJ, Tm, Td, etc.)
• Font encoding (WinAnsi, MacRoman, ToUnicode CMap)
• CID font handling (Type0 composite fonts, Identity-H/V encoding, UTF-16BE)
• Stream decompression (FlateDecode, ASCII85, ASCIIHex, LZW, RunLength)

MIT