192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304
/// Collects statistics about the SKS packet dump using the openpgp
/// crate, Sequoia's low-level API.
///
/// Note that to achieve reasonable performance, you need to compile
/// Sequoia and this program with optimizations:
///
///     % cargo run -p sequoia-openpgp --example statistics --release \
///           -- <packet-dump>

use std::env;
extern crate sequoia_openpgp as openpgp;
use openpgp::Packet;
use openpgp::constants::SignatureType;
use openpgp::packet::{user_attribute, BodyLength, Tag};
use openpgp::parse::{Parse, PacketParserResult, PacketParser};

fn main() {
    let args: Vec<String> = env::args().collect();
    if args.len() < 2 {
        panic!("Collects statistics about OpenPGP packet dumps.\n\n\
                Usage: {} <packet-dump> [<packet-dump>...]\n", args[0]);
    }

    // Global stats.
    let mut packet_count = 0;
    let mut packet_size = 0 as usize;

    // Per-tag statistics.
    let mut tags_count = vec![0; 64];
    let mut tags_unknown = vec![0; 64];
    let mut tags_size_bytes = vec![0 as usize; 64];
    let mut tags_size_count = vec![0; 64];
    let mut tags_size_min = vec![::std::u32::MAX; 64];
    let mut tags_size_max = vec![0; 64];
    let mut sigs_count = vec![0; 256];

    // Per-TPK statistics.
    let mut tpk_count = 0;
    let mut tpk = PerTPK::min();
    let mut tpk_min = PerTPK::max();
    let mut tpk_max = PerTPK::min();

    // UserAttribute statistics.
    let mut ua_image_count = vec![0; 256];
    let mut ua_unknown_count = vec![0; 256];
    let mut ua_invalid_count = 0;

    // For each input file, create a parser.
    for input in &args[1..] {
        eprintln!("Parsing {}...", input);
        let mut ppr = PacketParser::from_file(input)
            .expect("Failed to create reader");

        // Iterate over all packets.
        while let PacketParserResult::Some(pp) = ppr {
            // While the packet is in the parser, get some data for later.
            let size = match pp.header().length {
                BodyLength::Full(n) => Some(n),
                _ => None,
            };

            // Get the packet and advance the parser.
            let (packet, tmp) = pp.next().expect("Failed to get next packet");
            ppr = tmp;

            packet_count += 1;
            if let Some(n) = size {
                packet_size += n as usize;
            }
            let i = u8::from(packet.tag()) as usize;
            tags_count[i] += 1;

            match packet {
                // If a new TPK starts, update TPK statistics.
                Packet::PublicKey(_) | Packet::SecretKey(_) => {
                    if tpk_count > 0 {
                        tpk.update_min_max(&mut tpk_min, &mut tpk_max);
                    }
                    tpk_count += 1;
                    tpk = PerTPK::min();
                },

                Packet::Signature(ref sig) => {
                    sigs_count[u8::from(sig.sigtype()) as usize] += 1;
                    tpk.sigs[u8::from(sig.sigtype()) as usize] += 1;
                },

                Packet::UserAttribute(ref ua) => {
                    use user_attribute::Subpacket;
                    use user_attribute::Image;
                    for subpacket in ua.subpackets() {
                        match subpacket {
                            Ok(Subpacket::Image(i)) => match i {
                                Image::JPEG(_) =>
                                    ua_image_count[1] += 1,
                                Image::Private(n, _) =>
                                    ua_image_count[n as usize] += 1,
                                Image::Unknown(n, _) =>
                                    ua_image_count[n as usize] += 1,
                            },
                            Ok(Subpacket::Unknown(n, _)) =>
                                ua_unknown_count[n as usize] += 1,
                            Err(_) => ua_invalid_count += 1,
                        }
                    }
                },

                _ => (),
            }

            if let Packet::Unknown(_) = packet {
                tags_unknown[i] += 1;
            } else {
                // Only record size statistics of packets we successfully
                // parsed.
                if let Some(n) = size {
                    tags_size_bytes[i] += n as usize;
                    tags_size_count[i] += 1;
                    if n < tags_size_min[i] {
                        tags_size_min[i] = n;
                    }
                    if n > tags_size_max[i] {
                        tags_size_max[i] = n;
                    }

                    tpk.bytes += n as usize;
                }

                tpk.packets += 1;
                tpk.tags[i] += 1;
            }
        }
        tpk.update_min_max(&mut tpk_min, &mut tpk_max);
    }

    // Print statistics.
    println!("# Packet statistics");
    println!();
    println!("{:>14} {:>9} {:>9} {:>9} {:>9} {:>9} {:>12}",
             "", "count", "unknown",
             "min size", "mean size", "max size", "sum size");
    println!("-------------------------------------------------------\
              -----------------------");

    for t in 0..64 {
        let count = tags_count[t];
        if count > 0 {
            println!("{:>14} {:>9} {:>9} {:>9} {:>9} {:>9} {:>12}",
                     format!("{:?}", Tag::from(t as u8)),
                     count,
                     tags_unknown[t],
                     tags_size_min[t],
                     tags_size_bytes[t] / tags_size_count[t],
                     tags_size_max[t],
                     tags_size_bytes[t]);
        }
    }
    println!();

    println!("# Signature statistics");
    println!();
    println!("{:>22} {:>9}",
             "", "count",);
    println!("--------------------------------");
    for t in 0..256 {