added Node2Type, its tests, and an bin to convert the .node2swhid.bin file to...

added Node2Type, its tests, and an bin to convert the .node2swhid.bin file to the new .node2type.bin

Cargo.lock

@@ -4,9 +4,9 @@ version = 3
 
 [[package]]
 name = "addr2line"
-version = "0.19.0"
+version = "0.20.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a76fd60b23679b7d19bd066031410fb7e458ccc5e958eb5c325888ce4baedc97"
+checksum = "f4fa78e18c64fce05e902adecd7a5eed15a5e0a3439f7b0e169f0252214865e3"
 dependencies = [
  "gimli",
 ]
@@ -124,9 +124,9 @@ checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa"
 
 [[package]]
 name = "backtrace"
-version = "0.3.67"
+version = "0.3.68"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "233d376d6d185f2a3093e58f283f60f880315b6c60075b01f36b3b85154564ca"
+checksum = "4319208da049c43661739c5fade2ba182f09d1dc2299b32298d3a31692b17e12"
 dependencies = [
  "addr2line",
  "cc",
@@ -156,7 +156,7 @@ dependencies = [
  "regex",
  "rustc-hash",
  "shlex",
- "syn 2.0.18",
+ "syn 2.0.23",
  "which",
 ]
 
@@ -168,9 +168,9 @@ checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a"
 
 [[package]]
 name = "bitflags"
-version = "2.3.2"
+version = "2.3.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6dbe3c979c178231552ecba20214a8272df4e09f232a87aef4320cf06539aded"
+checksum = "630be753d4e58660abd17930c71b647fe46c27ea6b63cc59e1e3851406972e42"
 
 [[package]]
 name = "bitvec"
@@ -259,9 +259,9 @@ dependencies = [
 
 [[package]]
 name = "clap"
-version = "4.3.5"
+version = "4.3.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2686c4115cb0810d9a984776e197823d08ec94f176549a89a9efded477c456dc"
+checksum = "1640e5cc7fb47dbb8338fd471b105e7ed6c3cb2aeb00c2e067127ffd3764a05d"
 dependencies = [
  "clap_builder",
  "clap_derive",
@@ -270,13 +270,12 @@ dependencies = [
 
 [[package]]
 name = "clap_builder"
-version = "4.3.5"
+version = "4.3.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2e53afce1efce6ed1f633cf0e57612fe51db54a1ee4fd8f8503d078fe02d69ae"
+checksum = "98c59138d527eeaf9b53f35a77fcc1fad9d883116070c63d5de1c7dc7b00c72b"
 dependencies = [
  "anstream",
  "anstyle",
- "bitflags 1.3.2",
  "clap_lex",
  "strsim",
 ]
@@ -290,7 +289,7 @@ dependencies = [
  "heck",
  "proc-macro2",
  "quote",
- "syn 2.0.18",
+ "syn 2.0.23",
 ]
 
 [[package]]
@@ -367,7 +366,7 @@ dependencies = [
 [[package]]
 name = "dsi-bitstream"
 version = "0.1.0"
-source = "git+https://github.com/vigna/dsi-bitstream-rs#29e5c3f40a42cccb00ce08600b0de31fb4d41b7c"
+source = "git+https://github.com/vigna/dsi-bitstream-rs#380b90846cf5d8229910fd96b5c3305af8b4deb0"
 dependencies = [
  "anyhow",
  "rand",
@@ -543,18 +542,9 @@ dependencies = [
 
 [[package]]
 name = "hermit-abi"
-version = "0.2.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ee512640fe35acbfb4bb779db6f0d80704c2cacfa2e39b601ef3e3f47d1ae4c7"
-dependencies = [
- "libc",
-]
-
-[[package]]
-name = "hermit-abi"
-version = "0.3.1"
+version = "0.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fed44880c466736ef9a5c5b5facefb5ed0785676d0c02d612db14e54f0d84286"
+checksum = "443144c8cdadd93ebf52ddb4056d257f5b52c04d3c804e657d19eb73fc33668b"
 
 [[package]]
 name = "iana-time-zone"
@@ -594,20 +584,19 @@ version = "1.0.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "eae7b9aee968036d54dce06cebaefd919e4472e753296daccd6d344e3e2df0c2"
 dependencies = [
- "hermit-abi 0.3.1",
+ "hermit-abi 0.3.2",
  "libc",
  "windows-sys",
 ]
 
 [[package]]
 name = "is-terminal"
-version = "0.4.7"
+version = "0.4.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "adcf93614601c8129ddf72e2d5633df827ba6551541c6d8c59520a371475be1f"
+checksum = "cb0889898416213fab133e1d33a0e5858a48177452750691bde3666d0fdbaf8b"
 dependencies = [
- "hermit-abi 0.3.1",
- "io-lifetimes",
- "rustix",
+ "hermit-abi 0.3.2",
+ "rustix 0.38.3",
  "windows-sys",
 ]
 
@@ -622,9 +611,9 @@ dependencies = [
 
 [[package]]
 name = "itoa"
-version = "1.0.6"
+version = "1.0.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "453ad9f582a441959e5f0d088b02ce04cfe8d51a8eaf077f12ac6d3e94164ca6"
+checksum = "62b02a5381cc465bd3041d84623d0fa3b66738b52b8e2fc3bab8ad63ab032f4a"
 
 [[package]]
 name = "java-properties"
@@ -689,6 +678,12 @@ version = "0.3.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ef53942eb7bf7ff43a617b3e2c1c4a5ecf5944a7c1bc12d7ee39bbb15e5c1519"
 
+[[package]]
+name = "linux-raw-sys"
+version = "0.4.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "09fc20d2ca12cb9f044c93e3bd6d32d523e6e2ec3db4f7b2939cd99026ecd3f0"
+
 [[package]]
 name = "log"
 version = "0.4.19"
@@ -736,9 +731,9 @@ checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
 
 [[package]]
 name = "miniz_oxide"
-version = "0.6.2"
+version = "0.7.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b275950c28b37e794e8c55d88aeb5e139d0ce23fdbbeda68f8d7174abdf9e8fa"
+checksum = "e7810e0be55b428ada41041c41f32c9f1a42817901b4ccf45fa3d4b6561e74c7"
 dependencies = [
  "adler",
 ]
@@ -814,19 +809,19 @@ dependencies = [
 
 [[package]]
 name = "num_cpus"
-version = "1.15.0"
+version = "1.16.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0fac9e2da13b5eb447a6ce3d392f23a29d8694bff781bf03a16cd9ac8697593b"
+checksum = "4161fcb6d602d4d2081af7c3a45852d875a03dd337a6bfdd6e06407b61342a43"
 dependencies = [
- "hermit-abi 0.2.6",
+ "hermit-abi 0.3.2",
  "libc",
 ]
 
 [[package]]
 name = "object"
-version = "0.30.4"
+version = "0.31.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "03b4680b86d9cfafba8fc491dc9b6df26b68cf40e9e6cd73909194759a63c385"
+checksum = "8bda667d9f2b5051b8833f59f3bf748b28ef54f850f4fcb389a252aa383866d1"
 dependencies = [
  "memchr",
 ]
@@ -873,28 +868,28 @@ checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"
 
 [[package]]
 name = "prettyplease"
-version = "0.2.9"
+version = "0.2.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9825a04601d60621feed79c4e6b56d65db77cdca55cef43b46b0de1096d1c282"
+checksum = "92139198957b410250d43fad93e630d956499a625c527eda65175c8680f83387"
 dependencies = [
  "proc-macro2",
- "syn 2.0.18",
+ "syn 2.0.23",
 ]
 
 [[package]]
 name = "proc-macro2"
-version = "1.0.60"
+version = "1.0.63"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dec2b086b7a862cf4de201096214fa870344cf922b2b30c167badb3af3195406"
+checksum = "7b368fba921b0dce7e60f5e04ec15e565b3303972b42bcfde1d0713b881959eb"
 dependencies = [
  "unicode-ident",
 ]
 
 [[package]]
 name = "quote"
-version = "1.0.28"
+version = "1.0.29"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1b9ab9c7eadfd8df19006f1cf1a4aed13540ed5cbc047010ece5826e10825488"
+checksum = "573015e8ab27661678357f27dc26460738fd2b6c86e46f386fde94cb5d913105"
 dependencies = [
  "proc-macro2",
 ]
@@ -968,9 +963,21 @@ dependencies = [
 
 [[package]]
 name = "regex"
-version = "1.8.4"
+version = "1.9.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d0ab3ca65655bb1e41f2a8c8cd662eb4fb035e67c3f78da1d61dffe89d07300f"
+checksum = "89089e897c013b3deb627116ae56a6955a72b8bed395c9526af31c9fe528b484"
+dependencies = [
+ "aho-corasick",
+ "memchr",
+ "regex-automata",
+ "regex-syntax",
+]
+
+[[package]]
+name = "regex-automata"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fa250384981ea14565685dea16a9ccc4d1c541a13f82b9c168572264d1df8c56"
 dependencies = [
  "aho-corasick",
  "memchr",
@@ -979,9 +986,9 @@ dependencies = [
 
 [[package]]
 name = "regex-syntax"
-version = "0.7.2"
+version = "0.7.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "436b050e76ed2903236f032a59761c1eb99e1b0aead2c257922771dab1fc8c78"
+checksum = "2ab07dc67230e4a4718e70fd5c20055a4334b121f1f9db8fe63ef39ce9b8c846"
 
 [[package]]
 name = "rustc-demangle"
@@ -997,15 +1004,28 @@ checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
 
 [[package]]
 name = "rustix"
-version = "0.37.20"
+version = "0.37.23"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b96e891d04aa506a6d1f318d2771bcb1c7dfda84e126660ace067c9b474bb2c0"
+checksum = "4d69718bf81c6127a49dc64e44a742e8bb9213c0ff8869a22c308f84c1d4ab06"
 dependencies = [
  "bitflags 1.3.2",
  "errno",
  "io-lifetimes",
  "libc",
- "linux-raw-sys",
+ "linux-raw-sys 0.3.8",
+ "windows-sys",
+]
+
+[[package]]
+name = "rustix"
+version = "0.38.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac5ffa1efe7548069688cd7028f32591853cd7b5b756d41bcffd2353e4fc75b4"
+dependencies = [
+ "bitflags 2.3.3",
+ "errno",
+ "libc",
+ "linux-raw-sys 0.4.3",
  "windows-sys",
 ]
 
@@ -1026,9 +1046,9 @@ checksum = "d29ab0c6d3fc0ee92fe66e2d99f700eab17a8d57d1c1d3b748380fb20baa78cd"
 
 [[package]]
 name = "serde"
-version = "1.0.164"
+version = "1.0.167"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9e8c8cf938e98f769bc164923b06dce91cea1751522f46f8466461af04c9027d"
+checksum = "7daf513456463b42aa1d94cff7e0c24d682b429f020b9afa4f5ba5c40a22b237"
 
 [[package]]
 name = "shlex"
@@ -1070,10 +1090,10 @@ checksum = "73473c0e59e6d5812c5dfe2a064a6444949f089e20eec9a2e5506596494e4623"
 [[package]]
 name = "sux"
 version = "0.1.0"
-source = "git+https://github.com/vigna/sux-rs#25fbdf42024b6cbe98741bd0d8135f3188293677"
+source = "git+https://github.com/vigna/sux-rs#9c47cceefb8e6b45118c99fd72a4cf040eb14ad3"
 dependencies = [
  "anyhow",
- "bitflags 2.3.2",
+ "bitflags 2.3.3",
  "bytemuck",
  "mmap-rs",
  "stable_deref_trait",
@@ -1086,6 +1106,7 @@ version = "0.1.0"
 dependencies = [
  "anyhow",
  "bitvec",
+ "clap",
  "dsi-progress-logger",
  "libc",
  "log",
@@ -1109,9 +1130,9 @@ dependencies = [
 
 [[package]]
 name = "syn"
-version = "2.0.18"
+version = "2.0.23"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "32d41677bcbe24c20c52e7c70b0d8db04134c5d1066bf98662e2871ad200ea3e"
+checksum = "59fb7d6d8281a51045d62b8eb3a7d1ce347b76f312af50cd3dc0af39c87c1737"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -1134,9 +1155,9 @@ dependencies = [
 
 [[package]]
 name = "sysinfo"
-version = "0.29.2"
+version = "0.29.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9557d0845b86eea8182f7b10dff120214fb6cd9fd937b6f4917714e546a38695"
+checksum = "751e810399bba86e9326f5762b7f32ac5a085542df78da6a78d94e07d14d7c11"
 dependencies = [
  "cfg-if",
  "core-foundation-sys",
@@ -1163,7 +1184,7 @@ dependencies = [
  "cfg-if",
  "fastrand",
  "redox_syscall",
- "rustix",
+ "rustix 0.37.23",
  "windows-sys",
 ]
 
@@ -1178,22 +1199,22 @@ dependencies = [
 
 [[package]]
 name = "thiserror"
-version = "1.0.40"
+version = "1.0.43"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "978c9a314bd8dc99be594bc3c175faaa9794be04a5a5e153caba6915336cebac"
+checksum = "a35fc5b8971143ca348fa6df4f024d4d55264f3468c71ad1c2f365b0a4d58c42"
 dependencies = [
  "thiserror-impl",
 ]
 
 [[package]]
 name = "thiserror-impl"
-version = "1.0.40"
+version = "1.0.43"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f9456a42c5b0d803c8cd86e73dd7cc9edd429499f37a3550d286d5e86720569f"
+checksum = "463fe12d7993d3b327787537ce8dd4dfa058de32fc2b195ef3cde03dc4771e8f"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.18",
+ "syn 2.0.23",
 ]
 
 [[package]]
@@ -1208,9 +1229,9 @@ dependencies = [
 
 [[package]]
 name = "unicode-ident"
-version = "1.0.9"
+version = "1.0.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b15811caf2415fb889178633e7724bad2509101cde276048e013b9def5e51fa0"
+checksum = "22049a19f4a68748a168c0fc439f9516686aa045927ff767eca0a85101fb6e73"
 
 [[package]]
 name = "utf8parse"
@@ -1255,7 +1276,7 @@ dependencies = [
  "once_cell",
  "proc-macro2",
  "quote",
- "syn 2.0.18",
+ "syn 2.0.23",
  "wasm-bindgen-shared",
 ]
 
@@ -1277,7 +1298,7 @@ checksum = "54681b18a46765f095758388f2d0cf16eb8d4169b639ab575a8f5693af210c7b"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.18",
+ "syn 2.0.23",
  "wasm-bindgen-backend",
  "wasm-bindgen-shared",
 ]
@@ -1291,7 +1312,7 @@ checksum = "ca6ad05a4870b2bf5fe995117d3728437bd27d7cd5f06f13c17443ef369775a1"
 [[package]]
 name = "webgraph"
 version = "0.1.0"
-source = "git+https://github.com/vigna/webgraph-rs#319a8a30fcf7acf42c052e593cb72b6d8f0f2f5f"
+source = "git+https://github.com/vigna/webgraph-rs#c0a89d22add27a2ae81bf8a7f024ee13422067d6"
 dependencies = [
  "anyhow",
  "bindgen",
@@ -1376,7 +1397,7 @@ version = "0.48.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e686886bc078bc1b0b600cac0147aadb815089b6e4da64016cbd754b6342700f"
 dependencies = [
- "windows-targets 0.48.0",
+ "windows-targets 0.48.1",
 ]
 
 [[package]]
@@ -1385,7 +1406,7 @@ version = "0.48.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "677d2418bec65e3338edb076e806bc1ec15693c5d0104683f2efe857f61056a9"
 dependencies = [
- "windows-targets 0.48.0",
+ "windows-targets 0.48.1",
 ]
 
 [[package]]
@@ -1405,9 +1426,9 @@ dependencies = [
 
 [[package]]
 name = "windows-targets"
-version = "0.48.0"
+version = "0.48.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7b1eb6f0cd7c80c79759c929114ef071b87354ce476d9d94271031c0497adfd5"
+checksum = "05d4b17490f70499f20b9e791dcf6a299785ce8af4d709018206dc5b4953e95f"
 dependencies = [
  "windows_aarch64_gnullvm 0.48.0",
  "windows_aarch64_msvc 0.48.0",

rust/Cargo.toml

@@ -12,16 +12,17 @@ categories = ["compression", "graph"]
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-anyhow = "1.0.71"
+anyhow = {version="1.0.71", features=["backtrace"]}
 mmap-rs = "0.5.0"
 sux = {git = "https://github.com/vigna/sux-rs"}
 webgraph = {git = "https://github.com/vigna/webgraph-rs"}
 log = "0.4.17"
 libc = "0.2.147"
-
-[dev-dependencies]
 stderrlog = "0.5.4"
 dsi-progress-logger = "0.1.0"
+clap = { version = "4.1.6", features = ["derive"] }
+
+[dev-dependencies]
 bitvec = { version = "1.0.1", features = ["atomic"] }
 zstd = "0.12"
 

rust/src/bin/node2type.rs

+use anyhow::{Context, Result};
+use clap::Parser;
+use dsi_progress_logger::ProgressLogger;
+use log::info;
+use swh_graph::map::{Node2SWHID, Node2Type};
+
+#[derive(Parser, Debug)]
+#[command(about = "Build `.node2type.bin` from `.node2swhid.bin`. Example usage:  cargo run --bin node2type -- swh/graph/example_dataset/compressed/example ", long_about = None)]
+struct Args {
+    /// The basename of the graph.
+    basename: String,
+
+    /// The basename of the file to create. By default it's the same as
+    /// `basename`.
+    dst_basename: Option<String>,
+}
+
+pub fn main() -> Result<()> {
+    let args = Args::parse();
+
+    stderrlog::new()
+        .verbosity(2)
+        .timestamp(stderrlog::Timestamp::Second)
+        .init()
+        .with_context(|| "While Initializing the stderrlog")?;
+
+    // load the node ID -> SWHID map so we can convert it to a node2file
+    let node2swhid_path = format!("{}.node2swhid.bin", args.basename);
+    info!("loading node ID -> SWHID map from {node2swhid_path} ...");
+    let node2swhid = Node2SWHID::load(&node2swhid_path).with_context(|| {
+        format!(
+            "While loading the .node2swhid.bin file: {}",
+            node2swhid_path
+        )
+    })?;
+    let num_nodes = node2swhid.len();
+
+    // compute the path of the file we are creating
+    let node2type_path = format!(
+        "{}.node2type.bin",
+        args.dst_basename.unwrap_or(args.basename)
+    );
+    // create a new node2type file that can index `num_nodes` nodes
+    let mut node2type = Node2Type::new(&node2type_path, num_nodes as u64)
+        .with_context(|| format!("While creating the .node2type.bin file: {}", node2type_path))?;
+
+    // init the progress logger
+    let mut pl = ProgressLogger::default().display_memory();
+    pl.item_name = "node";
+    pl.local_speed = true;
+    pl.expected_updates = Some(num_nodes);
+    pl.start("iterating over node ID -> SWHID map ...");
+
+    // build the file
+    for node_id in 0..num_nodes {
+        // get the SWHID of the node to get the type and write it inside node2type
+        node2type.set(node_id, node2swhid.get(node_id).unwrap().node_type);
+        pl.light_update();
+    }
+
+    pl.done();
+    Ok(())
+}

rust/src/map/mod.rs

@@ -9,5 +9,8 @@
 mod node2swhid;
 pub use node2swhid::Node2SWHID;
 
+mod node2type;
+pub use node2type::Node2Type;
+
 mod order;
 pub use order::Order;

rust/src/map/node2swhid.rs

@@ -34,6 +34,22 @@ impl Node2SWHID {
 
 impl Node2SWHID {
     /// Convert a node_it to a SWHID
+    ///
+    /// # Safety
+    /// This function is unsafe because it does not check that `node_id` is
+    /// within bounds of the array if debug asserts are disabled
+    #[inline]
+    pub unsafe fn get_unchecked(&self, node_id: usize) -> SWHID {
+        let offset = node_id * SWHID::BYTES_SIZE;
+        let bytes = self.data.get_unchecked(offset..offset + SWHID::BYTES_SIZE);
+        // this unwrap is always safe because we use the same const
+        let bytes: [u8; SWHID::BYTES_SIZE] = bytes.try_into().unwrap();
+        // this unwrap can only fail on a corrupted file, so it's ok to panic
+        SWHID::try_from(bytes).unwrap()
+    }
+
+    /// Convert a node_it to a SWHID
+    #[inline]
     pub fn get(&self, node_id: usize) -> Option<SWHID> {
         let offset = node_id * SWHID::BYTES_SIZE;
         let bytes = self.data.get(offset..offset + SWHID::BYTES_SIZE)?;
@@ -45,6 +61,7 @@ impl Node2SWHID {
 
     /// Return how many node_ids are in this map
     #[allow(clippy::len_without_is_empty)] // rationale: we don't care about empty maps
+    #[inline]
     pub fn len(&self) -> usize {
         self.data.len() / SWHID::BYTES_SIZE
     }

rust/src/map/node2type.rs

+use crate::SWHType;
+use anyhow::{Context, Result};
+use log::info;
+use mmap_rs::{Mmap, MmapMut};
+use std::path::Path;
+use sux::prelude::CompactArray;
+use sux::traits::*;
+
+/// Struct to create and load a `.node2type.bin` file and convert node ids to types.
+pub struct Node2Type<B: VSlice> {
+    data: CompactArray<B>,
+}
+
+impl<B: VSlice> Node2Type<B> {
+    #[inline]
+    /// Get the type of a node with id `node_id` without bounds checking
+    ///
+    /// # Safety
+    /// This function is unsafe because it does not check that `node_id` is
+    /// within bounds of the array if debug asserts are disabled
+    pub unsafe fn get_unchecked(&self, node_id: usize) -> SWHType {
+        SWHType::try_from(self.data.get_unchecked(node_id) as u8).unwrap()
+    }
+
+    #[inline]
+    /// Get the type of a node with id `node_id`
+    pub fn get(&self, node_id: usize) -> Option<SWHType> {
+        SWHType::try_from(self.data.get(node_id) as u8).ok()
+    }
+}
+
+impl<B: VSliceMut> Node2Type<B> {
+    #[inline]
+    /// Get the type of a node with id `node_id` without bounds checking
+    ///
+    /// # Safety
+    /// This function is unsafe because it does not check that `node_id` is
+    /// within bounds of the array if debug asserts are disabled
+    pub unsafe fn set_unchecked(&mut self, node_id: usize, node_type: SWHType) {
+        self.data.set_unchecked(node_id, node_type as u64);
+    }
+
+    #[inline]
+    /// Set the type of a node with id `node_id`
+    pub fn set(&mut self, node_id: usize, node_type: SWHType) {
+        self.data.set(node_id, node_type as u64);
+    }
+}
+
+impl Node2Type<MmapMut> {
+    /// Create a new `.node2type.bin` file
+    pub fn new<P: AsRef<Path>>(path: P, num_nodes: u64) -> Result<Self> {
+        let path = path.as_ref();
+        // compute the size of the file we are creating in bytes
+        let mut file_len = (num_nodes * SWHType::BITWIDTH as u64 + 7) / 8;
+        // make the file dimension a multiple of 8 bytes so CompactArray can
+        // read u64 words from it
+        file_len += 8 - (file_len % 8);
+        info!("The resulting file will be {} bytes long.", file_len);
+
+        // create the file
+        let node2type_file = std::fs::File::options()
+            .read(true)
+            .write(true)
+            .create(true)
+            .open(path)
+            .with_context(|| {
+                format!(
+                    "While creating the .node2type.bin file: {}",
+                    path.to_string_lossy()
+                )
+            })?;
+
+        // fallocate the file with zeros so we can fill it without ever resizing it
+        node2type_file
+            .set_len(file_len)
+            .with_context(|| "While fallocating the file with zeros")?;
+
+        // create a mutable mmap to the file so we can directly write it in place
+        let mmap = unsafe {
+            mmap_rs::MmapOptions::new(file_len as _)?
+                .with_file(node2type_file, 0)
+                .map_mut()
+                .with_context(|| "While mmapping the file")?
+        };
+        // use the CompactArray over the mmap
+        let node2type =
+            unsafe { CompactArray::from_raw_parts(mmap, SWHType::BITWIDTH, num_nodes as usize) };
+
+        Ok(Self { data: node2type })
+    }
+
+    /// Load a mutable `.node2type.bin` file
+    pub fn load_mut<P: AsRef<Path>>(path: P, num_nodes: u64) -> Result<Self> {
+        let path = path.as_ref();
+        let file_len = path.metadata()?.len();
+        let file = std::fs::File::open(path)?;
+        let data = unsafe {
+            mmap_rs::MmapOptions::new(file_len as _)?
+                .with_flags((sux::prelude::Flags::TRANSPARENT_HUGE_PAGES).mmap_flags())
+                .with_file(file, 0)
+                .map_mut()?
+        };
+        #[cfg(target_os = "linux")]
+        unsafe {
+            libc::madvise(data.as_ptr() as *mut _, data.len(), libc::MADV_RANDOM)
+        };
+        // use the CompactArray over the mmap
+        let node2type =
+            unsafe { CompactArray::from_raw_parts(data, SWHType::BITWIDTH, num_nodes as usize) };
+        Ok(Self { data: node2type })
+    }
+}
+
+impl Node2Type<Mmap> {
+    /// Load a read-only `.node2type.bin` file
+    pub fn load<P: AsRef<Path>>(path: P, num_nodes: u64) -> Result<Self> {
+        let path = path.as_ref();
+        let file_len = path.metadata()?.len();
+        let file = std::fs::File::open(path)?;
+        let data = unsafe {
+            mmap_rs::MmapOptions::new(file_len as _)?
+                .with_flags((sux::prelude::Flags::TRANSPARENT_HUGE_PAGES).mmap_flags())
+                .with_file(file, 0)
+                .map()?
+        };
+        #[cfg(target_os = "linux")]
+        unsafe {
+            libc::madvise(data.as_ptr() as *mut _, data.len(), libc::MADV_RANDOM)
+        };
+        // use the CompactArray over the mmap
+        let node2type =
+            unsafe { CompactArray::from_raw_parts(data, SWHType::BITWIDTH, num_nodes as usize) };
+        Ok(Self { data: node2type })
+    }
+}

rust/src/swhtype.rs

@@ -83,6 +83,21 @@ impl TryFrom<u8> for SWHType {
 }
 
 impl SWHType {
+    /// Get the number of possible types.
+    ///
+    /// To avoid having to update this when adding a new type
+    /// we can use the unstable function `std::mem::variant_count`
+    /// or the `variant_count` crate.
+    /// But for now we just hardcode it while we decide how to
+    /// deal with this.
+    pub const NUMBER_OF_TYPES: usize = 6;
+
+    /// The number of bits needed to store the node type as integers
+    /// This is `ceil(log2(NUMBER_OF_TYPES))`  which can be arithmetized into
+    /// `floor(log2(NUMBER_OF_TYPES))` plus one if it's not a power of two.
+    pub const BITWIDTH: usize = Self::NUMBER_OF_TYPES.ilog2() as usize
+        + (!Self::NUMBER_OF_TYPES.is_power_of_two()) as usize;
+
     /// Convert a type to the str used in the SWHID
     pub fn to_str(&self) -> &'static str {
         match self {

rust/tests/test_node2type.rs

+use anyhow::{Context, Result};
+use log::info;
+use swh_graph::map::{Node2SWHID, Node2Type};
+
+const BASENAME: &str = "../swh/graph/example_dataset/compressed/example";
+
+#[test]
+fn test_node2type() -> Result<()> {
+    // load the node ID -> SWHID map so we can convert it to a node2file
+    let node2swhid_path = format!("{}.node2swhid.bin", BASENAME);
+    info!("loading node ID -> SWHID map from {node2swhid_path} ...");
+    let node2swhid = Node2SWHID::load(&node2swhid_path).with_context(|| {
+        format!(
+            "While loading the .node2swhid.bin file: {}",
+            node2swhid_path
+        )
+    })?;
+    let num_nodes = node2swhid.len();
+
+    // load the node2type file
+    let node2type_path = format!("{}.node2type.bin", BASENAME);
+    info!("loading node ID -> type map from {node2type_path} ...");
+    let node2type = Node2Type::load(&node2type_path, num_nodes as u64)
+        .with_context(|| format!("While loading the .node2type.bin file: {}", node2type_path))?;
+
+    // check that the the node2type matches with the node2swhid
+    for node_id in 0..num_nodes {
+        assert_eq!(
+            node2swhid.get(node_id).unwrap().node_type,
+            node2type.get(node_id).unwrap()
+        )
+    }
+
+    Ok(())
+}

rust/tests/test_order_mph.rs

@@ -13,7 +13,6 @@ use webgraph::prelude::*;
 const BASENAME: &str = "../swh/graph/example_dataset/compressed/example";
 
 #[test]
-#[allow(dead_code)]
 fn test_order_mph() -> Result<()> {
     // Setup a stderr logger because ProgressLogger uses the `log` crate
     // to printout