Version v2.3.1

Fixes #119

.github/workflows/build.yaml

@@ -11,6 +11,7 @@ on:
 env:
   BUILD_PROFILE: release-lto
   CARGO_TARGET_DIR: target
+  CARGO_INCREMENTAL: 0
 
 jobs:
   check:
@@ -25,37 +26,14 @@ jobs:
           sudo apt-get -y install libgtk-3-dev
       - name: Checkout
         uses: actions/checkout@v4
-      - name: Check git tag against Cargo version
-        if: startsWith(github.ref, 'refs/tags/')
-        shell: bash
-        run: |
-          set -eou pipefail
-          tag='${{github.ref}}'
-          tag="${tag#refs/tags/}"
-          for file in */Cargo.toml; do
-            version=$(grep '^version' $file | head -1 | awk -F' = ' '{print $2}' | tr -d '"')
-            version="v$version"
-            if [ "$tag" != "$version" ]; then
-              echo "::error::Git tag doesn't match the Cargo version! ($tag != $version)"
-              exit 1
-            fi
-          done
       - name: Setup Rust toolchain
         uses: dtolnay/rust-toolchain@stable
         with:
           components: clippy
-      - name: Setup sccache
-        uses: mozilla-actions/sccache-action@v0.0.4
       - name: Cargo check
-        env:
-          RUSTC_WRAPPER: sccache
-          SCCACHE_GHA_ENABLED: "true"
-        run: cargo check
+        run: cargo check --all-features --all-targets
       - name: Cargo clippy
-        env:
-          RUSTC_WRAPPER: sccache
-          SCCACHE_GHA_ENABLED: "true"
-        run: cargo clippy
+        run: cargo clippy --all-features --all-targets
 
   fmt:
     name: Format
@@ -107,13 +85,8 @@ jobs:
         uses: actions/checkout@v4
       - name: Setup Rust toolchain
         uses: dtolnay/rust-toolchain@stable
-      - name: Setup sccache
-        uses: mozilla-actions/sccache-action@v0.0.4
       - name: Cargo test
-        env:
-          RUSTC_WRAPPER: sccache
-          SCCACHE_GHA_ENABLED: "true"
-        run: cargo test --release
+        run: cargo test --release --all-features
 
   build-cli:
     name: Build objdiff-cli
@@ -137,10 +110,10 @@ jobs:
             name: linux-aarch64
             build: zigbuild
             features: default
-          - platform: ubuntu-latest
-            target: armv7-unknown-linux-musleabi
-            name: linux-armv7l
-            build: zigbuild
+          - platform: windows-latest
+            target: i686-pc-windows-msvc
+            name: windows-x86
+            build: build
             features: default
           - platform: windows-latest
             target: x86_64-pc-windows-msvc
@@ -169,7 +142,11 @@ jobs:
         uses: actions/checkout@v4
       - name: Install cargo-zigbuild
         if: matrix.build == 'zigbuild'
-        run: pip install ziglang==0.13.0 cargo-zigbuild==0.19.1
+        run: |
+          python3 -m venv .venv
+          . .venv/bin/activate
+          echo PATH=$PATH >> $GITHUB_ENV
+          pip install ziglang==0.13.0 cargo-zigbuild==0.19.1
       - name: Setup Rust toolchain
         uses: dtolnay/rust-toolchain@stable
         with:
@@ -225,12 +202,7 @@ jobs:
         uses: dtolnay/rust-toolchain@stable
         with:
           targets: ${{ matrix.target }}
-      - name: Setup sccache
-        uses: mozilla-actions/sccache-action@v0.0.4
       - name: Cargo build
-        env:
-          RUSTC_WRAPPER: sccache
-          SCCACHE_GHA_ENABLED: "true"
         run: >
           cargo build --profile ${{ env.BUILD_PROFILE }} --target ${{ matrix.target }} 
           --bin ${{ env.CARGO_BIN_NAME }} --features ${{ matrix.features }}
@@ -247,10 +219,24 @@ jobs:
     name: Release
     if: startsWith(github.ref, 'refs/tags/')
     runs-on: ubuntu-latest
-    needs: [ check, build-cli, build-gui ]
+    needs: [ build-cli, build-gui ]
     permissions:
       contents: write
     steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Check git tag against Cargo version
+        shell: bash
+        run: |
+          set -eou pipefail
+          tag='${{github.ref}}'
+          tag="${tag#refs/tags/}"
+          version=$(grep '^version' Cargo.toml | head -1 | awk -F' = ' '{print $2}' | tr -d '"')
+          version="v$version"
+          if [ "$tag" != "$version" ]; then
+            echo "::error::Git tag doesn't match the Cargo version! ($tag != $version)"
+            exit 1
+          fi
       - name: Download artifacts
         uses: actions/download-artifact@v4
         with:
@@ -278,6 +264,8 @@ jobs:
           done
           ls -R ../out
       - name: Release
-        uses: softprops/action-gh-release@v1
+        uses: softprops/action-gh-release@v2
         with:
           files: out/*
+          draft: true
+          generate_release_notes: true

Cargo.toml

@@ -8,5 +8,14 @@ resolver = "2"
 
 [profile.release-lto]
 inherits = "release"
-lto = "thin"
+lto = "fat"
 strip = "debuginfo"
+codegen-units = 1
+
+[workspace.package]
+version = "2.3.1"
+authors = ["Luke Street <luke@street.dev>"]
+edition = "2021"
+license = "MIT OR Apache-2.0"
+repository = "https://github.com/encounter/objdiff"
+rust-version = "1.74"

README.md

@@ -6,6 +6,7 @@
 A local diffing tool for decompilation projects. Inspired by [decomp.me](https://decomp.me) and [asm-differ](https://github.com/simonlindholm/asm-differ).
 
 Features:
+
 - Compare entire object files: functions and data.
 - Built-in symbol demangling for C++. (CodeWarrior, Itanium & MSVC)
 - Automatic rebuild on source file changes.
@@ -14,6 +15,7 @@ Features:
 - Click to highlight all instances of values and registers.
 
 Supports:
+
 - PowerPC 750CL (GameCube, Wii)
 - MIPS (N64, PS1, PS2, PSP)
 - x86 (COFF only at the moment)
@@ -21,6 +23,25 @@ Supports:
 
 See [Usage](#usage) for more information.
 
+## Downloads
+
+To build from source, see [Building](#building).
+
+### GUI
+
+- [Windows (x86_64)](https://github.com/encounter/objdiff/releases/latest/download/objdiff-windows-x86_64.exe)
+- [Linux (x86_64)](https://github.com/encounter/objdiff/releases/latest/download/objdiff-linux-x86_64)
+- [macOS (arm64)](https://github.com/encounter/objdiff/releases/latest/download/objdiff-macos-arm64)
+- [macOS (x86_64)](https://github.com/encounter/objdiff/releases/latest/download/objdiff-macos-x86_64)
+
+For Linux and macOS, run `chmod +x objdiff-*` to make the binary executable.
+
+### CLI
+
+CLI binaries can be found on the [releases page](https://github.com/encounter/objdiff/releases).
+
+## Screenshots
+
 ![Symbol Screenshot](assets/screen-symbols.png)
 ![Diff Screenshot](assets/screen-diff.png)
 
@@ -141,16 +162,22 @@ Install Rust via [rustup](https://rustup.rs).
 $ git clone https://github.com/encounter/objdiff.git
 $ cd objdiff
 $ cargo run --release
-# or, for wgpu backend (recommended on macOS)
-$ cargo run --release --features wgpu
 ```
 
+Or using `cargo install`.
+
+```shell
+$ cargo install --locked --git https://github.com/encounter/objdiff.git objdiff-gui objdiff-cli
+```
+
+The binaries will be installed to `~/.cargo/bin` as `objdiff` and `objdiff-cli`.
+
 ## License
 
 Licensed under either of
 
-* Apache License, Version 2.0, ([LICENSE-APACHE](LICENSE-APACHE) or http://www.apache.org/licenses/LICENSE-2.0)
-* MIT license ([LICENSE-MIT](LICENSE-MIT) or http://opensource.org/licenses/MIT)
+- Apache License, Version 2.0, ([LICENSE-APACHE](LICENSE-APACHE) or <http://www.apache.org/licenses/LICENSE-2.0>)
+- MIT license ([LICENSE-MIT](LICENSE-MIT) or <http://opensource.org/licenses/MIT>)
 
 at your option.
 

config.schema.json

@@ -133,6 +133,13 @@
         },
         "metadata": {
           "ref": "#/$defs/metadata"
+        },
+        "symbol_mappings": {
+          "type": "object",
+          "description": "Manual symbol mappings from target to base.",
+          "additionalProperties": {
+            "type": "string"
+          }
         }
       }
     },

deny.toml

@@ -9,6 +9,11 @@
 # The values provided in this template are the default values that will be used
 # when any section or field is not specified in your own configuration
 
+# Root options
+
+# The graph table configures how the dependency graph is constructed and thus
+# which crates the checks are performed against
+[graph]
 # If 1 or more target triples (and optionally, target_features) are specified,
 # only the specified targets will be checked when running `cargo deny check`.
 # This means, if a particular package is only ever used as a target specific
@@ -20,51 +25,67 @@
 targets = [
     # The triple can be any string, but only the target triples built in to
     # rustc (as of 1.40) can be checked against actual config expressions
-    #{ triple = "x86_64-unknown-linux-musl" },
+    #"x86_64-unknown-linux-musl",
     # You can also specify which target_features you promise are enabled for a
     # particular target. target_features are currently not validated against
     # the actual valid features supported by the target architecture.
     #{ triple = "wasm32-unknown-unknown", features = ["atomics"] },
 ]
+# When creating the dependency graph used as the source of truth when checks are
+# executed, this field can be used to prune crates from the graph, removing them
+# from the view of cargo-deny. This is an extremely heavy hammer, as if a crate
+# is pruned from the graph, all of its dependencies will also be pruned unless
+# they are connected to another crate in the graph that hasn't been pruned,
+# so it should be used with care. The identifiers are [Package ID Specifications]
+# (https://doc.rust-lang.org/cargo/reference/pkgid-spec.html)
+#exclude = []
+# If true, metadata will be collected with `--all-features`. Note that this can't
+# be toggled off if true, if you want to conditionally enable `--all-features` it
+# is recommended to pass `--all-features` on the cmd line instead
+all-features = false
+# If true, metadata will be collected with `--no-default-features`. The same
+# caveat with `all-features` applies
+no-default-features = false
+# If set, these feature will be enabled when collecting metadata. If `--features`
+# is specified on the cmd line they will take precedence over this option.
+#features = []
+
+# The output table provides options for how/if diagnostics are outputted
+[output]
+# When outputting inclusion graphs in diagnostics that include features, this
+# option can be used to specify the depth at which feature edges will be added.
+# This option is included since the graphs can be quite large and the addition
+# of features from the crate(s) to all of the graph roots can be far too verbose.
+# This option can be overridden via `--feature-depth` on the cmd line
+feature-depth = 1
 
 # This section is considered when running `cargo deny check advisories`
 # More documentation for the advisories section can be found here:
 # https://embarkstudios.github.io/cargo-deny/checks/advisories/cfg.html
 [advisories]
-# The path where the advisory database is cloned/fetched into
-db-path = "~/.cargo/advisory-db"
+# The path where the advisory databases are cloned/fetched into
+#db-path = "$CARGO_HOME/advisory-dbs"
 # The url(s) of the advisory databases to use
-db-urls = ["https://github.com/rustsec/advisory-db"]
-# The lint level for security vulnerabilities
-vulnerability = "deny"
-# The lint level for unmaintained crates
-unmaintained = "warn"
-# The lint level for crates that have been yanked from their source registry
-yanked = "warn"
-# The lint level for crates with security notices. Note that as of
-# 2019-12-17 there are no security notice advisories in
-# https://github.com/rustsec/advisory-db
-notice = "warn"
+#db-urls = ["https://github.com/rustsec/advisory-db"]
 # A list of advisory IDs to ignore. Note that ignored advisories will still
 # output a note when they are encountered.
-ignore = []
-# Threshold for security vulnerabilities, any vulnerability with a CVSS score
-# lower than the range specified will be ignored. Note that ignored advisories
-# will still output a note when they are encountered.
-# * None - CVSS Score 0.0
-# * Low - CVSS Score 0.1 - 3.9
-# * Medium - CVSS Score 4.0 - 6.9
-# * High - CVSS Score 7.0 - 8.9
-# * Critical - CVSS Score 9.0 - 10.0
-#severity-threshold =
+ignore = [
+    "RUSTSEC-2024-0370",
+    #{ id = "RUSTSEC-0000-0000", reason = "you can specify a reason the advisory is ignored" },
+    #"a-crate-that-is-yanked@0.1.1", # you can also ignore yanked crate versions if you wish
+    #{ crate = "a-crate-that-is-yanked@0.1.1", reason = "you can specify why you are ignoring the yanked crate" },
+]
+# If this is true, then cargo deny will use the git executable to fetch advisory database.
+# If this is false, then it uses a built-in git library.
+# Setting this to true can be helpful if you have special authentication requirements that cargo-deny does not support.
+# See Git Authentication for more information about setting up git authentication.
+#git-fetch-with-cli = true
 
 # This section is considered when running `cargo deny check licenses`
 # More documentation for the licenses section can be found here:
 # https://embarkstudios.github.io/cargo-deny/checks/licenses/cfg.html
 [licenses]
-# The lint level for crates which do not have a detectable license
-unlicensed = "deny"
-# List of explictly allowed licenses
+# List of explicitly allowed licenses
 # See https://spdx.org/licenses/ for list of possible licenses
 # [possible values: any SPDX 3.11 short identifier (+ optional exception)].
 allow = [
@@ -83,28 +104,7 @@ allow = [
     "OFL-1.1",
     "LicenseRef-UFL-1.0",
     "OpenSSL",
-    "GPL-3.0",
 ]
-# List of explictly disallowed licenses
-# See https://spdx.org/licenses/ for list of possible licenses
-# [possible values: any SPDX 3.11 short identifier (+ optional exception)].
-deny = [
-    #"Nokia",
-]
-# Lint level for licenses considered copyleft
-copyleft = "warn"
-# Blanket approval or denial for OSI-approved or FSF Free/Libre licenses
-# * both - The license will be approved if it is both OSI-approved *AND* FSF
-# * either - The license will be approved if it is either OSI-approved *OR* FSF
-# * osi-only - The license will be approved if is OSI-approved *AND NOT* FSF
-# * fsf-only - The license will be approved if is FSF *AND NOT* OSI-approved
-# * neither - This predicate is ignored and the default lint level is used
-allow-osi-fsf-free = "neither"
-# Lint level used when no other predicates are matched
-# 1. License isn't in the allow or deny lists
-# 2. License isn't copyleft
-# 3. License isn't OSI/FSF, or allow-osi-fsf-free = "neither"
-default = "deny"
 # The confidence threshold for detecting a license from license text.
 # The higher the value, the more closely the license text must be to the
 # canonical license text of a valid SPDX license file.
@@ -115,17 +115,15 @@ confidence-threshold = 0.8
 exceptions = [
     # Each entry is the crate and version constraint, and its specific allow
     # list
-    #{ allow = ["Zlib"], name = "adler32", version = "*" },
+    #{ allow = ["Zlib"], crate = "adler32" },
 ]
 
 # Some crates don't have (easily) machine readable licensing information,
 # adding a clarification entry for it allows you to manually specify the
 # licensing information
 [[licenses.clarify]]
-# The name of the crate the clarification applies to
-name = "ring"
-# The optional version constraint for the crate
-version = "*"
+# The package spec the clarification applies to
+crate = "ring"
 # The SPDX expression for the license requirements of the crate
 expression = "MIT AND ISC AND OpenSSL"
 # One or more files in the crate's source used as the "source of truth" for
@@ -140,7 +138,9 @@ license-files = [
 
 [licenses.private]
 # If true, ignores workspace crates that aren't published, or are only
-# published to private registries
+# published to private registries.
+# To see how to mark a crate as unpublished (to the official registry),
+# visit https://doc.rust-lang.org/cargo/reference/manifest.html#the-publish-field.
 ignore = false
 # One or more private registries that you might publish crates to, if a crate
 # is only published to private registries, and ignore is true, the crate will
@@ -163,30 +163,63 @@ wildcards = "allow"
 # * simplest-path - The path to the version with the fewest edges is highlighted
 # * all - Both lowest-version and simplest-path are used
 highlight = "all"
+# The default lint level for `default` features for crates that are members of
+# the workspace that is being checked. This can be overridden by allowing/denying
+# `default` on a crate-by-crate basis if desired.
+workspace-default-features = "allow"
+# The default lint level for `default` features for external crates that are not
+# members of the workspace. This can be overridden by allowing/denying `default`
+# on a crate-by-crate basis if desired.
+external-default-features = "allow"
 # List of crates that are allowed. Use with care!
 allow = [
-    #{ name = "ansi_term", version = "=0.11.0" },
+    #"ansi_term@0.11.0",
+    #{ crate = "ansi_term@0.11.0", reason = "you can specify a reason it is allowed" },
 ]
 # List of crates to deny
 deny = [
-    # Each entry the name of a crate and a version range. If version is
-    # not specified, all versions will be matched.
-    #{ name = "ansi_term", version = "=0.11.0" },
-    #
+    #"ansi_term@0.11.0",
+    #{ crate = "ansi_term@0.11.0", reason = "you can specify a reason it is banned" },
     # Wrapper crates can optionally be specified to allow the crate when it
     # is a direct dependency of the otherwise banned crate
-    #{ name = "ansi_term", version = "=0.11.0", wrappers = [] },
+    #{ crate = "ansi_term@0.11.0", wrappers = ["this-crate-directly-depends-on-ansi_term"] },
 ]
+
+# List of features to allow/deny
+# Each entry the name of a crate and a version range. If version is
+# not specified, all versions will be matched.
+#[[bans.features]]
+#crate = "reqwest"
+# Features to not allow
+#deny = ["json"]
+# Features to allow
+#allow = [
+#    "rustls",
+#    "__rustls",
+#    "__tls",
+#    "hyper-rustls",
+#    "rustls",
+#    "rustls-pemfile",
+#    "rustls-tls-webpki-roots",
+#    "tokio-rustls",
+#    "webpki-roots",
+#]
+# If true, the allowed features must exactly match the enabled feature set. If
+# this is set there is no point setting `deny`
+#exact = true
+
 # Certain crates/versions that will be skipped when doing duplicate detection.
 skip = [
-    #{ name = "ansi_term", version = "=0.11.0" },
+    #"ansi_term@0.11.0",
+    #{ crate = "ansi_term@0.11.0", reason = "you can specify a reason why it can't be updated/removed" },
 ]
 # Similarly to `skip` allows you to skip certain crates during duplicate
 # detection. Unlike skip, it also includes the entire tree of transitive
 # dependencies starting at the specified crate, up to a certain depth, which is
-# by default infinite
+# by default infinite.
 skip-tree = [
-    #{ name = "ansi_term", version = "=0.11.0", depth = 20 },
+    #"ansi_term@0.11.0", # will be skipped along with _all_ of its direct and transitive dependencies
+    #{ crate = "ansi_term@0.11.0", depth = 20 },
 ]
 
 # This section is considered when running `cargo deny check sources`.
@@ -206,9 +239,9 @@ allow-registry = ["https://github.com/rust-lang/crates.io-index"]
 allow-git = []
 
 [sources.allow-org]
-# 1 or more github.com organizations to allow git sources for
+# github.com organizations to allow git sources for
 github = ["encounter"]
-# 1 or more gitlab.com organizations to allow git sources for
-#gitlab = [""]
-# 1 or more bitbucket.org organizations to allow git sources for
-#bitbucket = [""]
+# gitlab.com organizations to allow git sources for
+gitlab = []
+# bitbucket.org organizations to allow git sources for
+bitbucket = []

objdiff-cli/Cargo.toml

@@ -1,31 +1,33 @@
 [package]
 name = "objdiff-cli"
-version = "2.0.0-beta.6"
-edition = "2021"
-rust-version = "1.70"
-authors = ["Luke Street <luke@street.dev>"]
-license = "MIT OR Apache-2.0"
-repository = "https://github.com/encounter/objdiff"
+version.workspace = true
+edition.workspace = true
+rust-version.workspace = true
+authors.workspace = true
+license.workspace = true
+repository.workspace = true
 readme = "../README.md"
 description = """
 A local diffing tool for decompilation projects.
 """
 publish = false
-build = "build.rs"
 
 [dependencies]
-anyhow = "1.0.82"
-argp = "0.3.0"
-crossterm = "0.27.0"
-enable-ansi-support = "0.2.1"
-memmap2 = "0.9.4"
+anyhow = "1.0"
+argp = "0.3"
+crossterm = "0.28"
+enable-ansi-support = "0.2"
+memmap2 = "0.9"
 objdiff-core = { path = "../objdiff-core", features = ["all"] }
-prost = "0.13.1"
-ratatui = "0.26.2"
-rayon = "1.10.0"
-serde = { version = "1", features = ["derive"] }
-serde_json = "1.0.116"
-supports-color = "3.0.0"
-time = { version = "0.3.36", features = ["formatting", "local-offset"] }
-tracing = "0.1.40"
-tracing-subscriber = { version = "0.3.18", features = ["env-filter"] }
+prost = "0.13"
+ratatui = "0.28"
+rayon = "1.10"
+serde = { version = "1.0", features = ["derive"] }
+serde_json = "1.0"
+supports-color = "3.0"
+time = { version = "0.3", features = ["formatting", "local-offset"] }
+tracing = "0.1"
+tracing-subscriber = { version = "0.3", features = ["env-filter"] }
+
+[target.'cfg(target_env = "musl")'.dependencies]
+mimalloc = "0.1"

objdiff-cli/build.rs

@@ -1,9 +0,0 @@
-fn main() {
-    let output = std::process::Command::new("git")
-        .args(["rev-parse", "HEAD"])
-        .output()
-        .expect("Failed to execute git");
-    let rev = String::from_utf8(output.stdout).expect("Failed to parse git output");
-    println!("cargo:rustc-env=GIT_COMMIT_SHA={rev}");
-    println!("cargo:rustc-rerun-if-changed=.git/HEAD");
-}

objdiff-cli/src/argp_version.rs

@@ -31,10 +31,9 @@ where T: FromArgs
             Ok(v) => {
                 if v.version {
                     println!(
-                        "{} {} {}",
+                        "{} {}",
                         command_name.first().unwrap_or(&""),
                         env!("CARGO_PKG_VERSION"),
-                        env!("GIT_COMMIT_SHA"),
                     );
                     std::process::exit(0);
                 } else {

objdiff-cli/src/cmd/diff.rs

@@ -102,26 +102,32 @@ pub fn run(args: Args) -> Result<()> {
                     let unit_path =
                         PathBuf::from_str(u).ok().and_then(|p| fs::canonicalize(p).ok());
 
-                    let Some(object) = project_config.objects.iter_mut().find_map(|obj| {
-                        if obj.name.as_deref() == Some(u) {
+                    let Some(object) = project_config
+                        .units
+                        .as_deref_mut()
+                        .unwrap_or_default()
+                        .iter_mut()
+                        .find_map(|obj| {
+                            if obj.name.as_deref() == Some(u) {
+                                resolve_paths(obj);
+                                return Some(obj);
+                            }
+
+                            let up = unit_path.as_deref()?;
+
                             resolve_paths(obj);
-                            return Some(obj);
-                        }
 
-                        let up = unit_path.as_deref()?;
+                            if [&obj.base_path, &obj.target_path]
+                                .into_iter()
+                                .filter_map(|p| p.as_ref().and_then(|p| p.canonicalize().ok()))
+                                .any(|p| p == up)
+                            {
+                                return Some(obj);
+                            }
 
-                        resolve_paths(obj);
-
-                        if [&obj.base_path, &obj.target_path]
-                            .into_iter()
-                            .filter_map(|p| p.as_ref().and_then(|p| p.canonicalize().ok()))
-                            .any(|p| p == up)
-                        {
-                            return Some(obj);
-                        }
-
-                        None
-                    }) else {
+                            None
+                        })
+                    else {
                         bail!("Unit not found: {}", u)
                     };
 
@@ -129,7 +135,13 @@ pub fn run(args: Args) -> Result<()> {
                 } else if let Some(symbol_name) = &args.symbol {
                     let mut idx = None;
                     let mut count = 0usize;
-                    for (i, obj) in project_config.objects.iter_mut().enumerate() {
+                    for (i, obj) in project_config
+                        .units
+                        .as_deref_mut()
+                        .unwrap_or_default()
+                        .iter_mut()
+                        .enumerate()
+                    {
                         resolve_paths(obj);
 
                         if obj
@@ -148,7 +160,7 @@ pub fn run(args: Args) -> Result<()> {
                     }
                     match (count, idx) {
                         (0, None) => bail!("Symbol not found: {}", symbol_name),
-                        (1, Some(i)) => &mut project_config.objects[i],
+                        (1, Some(i)) => &mut project_config.units_mut()[i],
                         (2.., Some(_)) => bail!(
                             "Multiple instances of {} were found, try specifying a unit",
                             symbol_name
@@ -303,7 +315,7 @@ fn find_function(obj: &ObjInfo, name: &str) -> Option<SymbolRef> {
     None
 }
 
-#[allow(dead_code)]
+#[expect(dead_code)]
 struct FunctionDiffUi {
     relax_reloc_diffs: bool,
     left_highlight: HighlightKind,
@@ -345,7 +357,7 @@ enum EventControlFlow {
 
 impl FunctionDiffUi {
     fn draw(&mut self, f: &mut Frame, result: &mut EventResult) {
-        let chunks = Layout::vertical([Constraint::Length(1), Constraint::Fill(1)]).split(f.size());
+        let chunks = Layout::vertical([Constraint::Length(1), Constraint::Fill(1)]).split(f.area());
         let header_chunks = Layout::horizontal([
             Constraint::Fill(1),
             Constraint::Length(3),
@@ -415,7 +427,7 @@ impl FunctionDiffUi {
             get_symbol_diff(self.diff_result.left.as_ref(), self.left_sym),
         ) {
             let mut text = Text::default();
-            let rect = content_chunks[0].inner(&Margin::new(0, 1));
+            let rect = content_chunks[0].inner(Margin::new(0, 1));
             left_highlight = self.print_sym(
                 &mut text,
                 symbol,
@@ -437,7 +449,7 @@ impl FunctionDiffUi {
             get_symbol_diff(self.diff_result.right.as_ref(), self.right_sym),
         ) {
             let mut text = Text::default();
-            let rect = content_chunks[2].inner(&Margin::new(0, 1));
+            let rect = content_chunks[2].inner(Margin::new(0, 1));
             right_highlight = self.print_sym(
                 &mut text,
                 symbol,
@@ -452,7 +464,7 @@ impl FunctionDiffUi {
 
             // Render margin
             let mut text = Text::default();
-            let rect = content_chunks[1].inner(&Margin::new(1, 1));
+            let rect = content_chunks[1].inner(Margin::new(1, 1));
             self.print_margin(&mut text, symbol_diff, rect);
             margin_text = Some(text);
         }
@@ -465,7 +477,7 @@ impl FunctionDiffUi {
                 get_symbol_diff(self.diff_result.prev.as_ref(), self.prev_sym),
             ) {
                 let mut text = Text::default();
-                let rect = content_chunks[4].inner(&Margin::new(0, 1));
+                let rect = content_chunks[4].inner(Margin::new(0, 1));
                 self.print_sym(
                     &mut text,
                     symbol,
@@ -480,7 +492,7 @@ impl FunctionDiffUi {
 
                 // Render margin
                 let mut text = Text::default();
-                let rect = content_chunks[3].inner(&Margin::new(1, 1));
+                let rect = content_chunks[3].inner(Margin::new(1, 1));
                 self.print_margin(&mut text, symbol_diff, rect);
                 prev_margin_text = Some(text);
             }
@@ -498,18 +510,30 @@ impl FunctionDiffUi {
             // Render left column
             f.render_widget(
                 Paragraph::new(text)
-                    .block(Block::new().borders(Borders::TOP).gray().title("TARGET".bold()))
+                    .block(
+                        Block::new()
+                            .borders(Borders::TOP)
+                            .border_style(Style::new().fg(Color::Gray))
+                            .title_style(Style::new().bold())
+                            .title("TARGET"),
+                    )
                     .scroll((0, self.scroll_x as u16)),
                 content_chunks[0],
             );
         }
         if let Some(text) = margin_text {
-            f.render_widget(text, content_chunks[1].inner(&Margin::new(1, 1)));
+            f.render_widget(text, content_chunks[1].inner(Margin::new(1, 1)));
         }
         if let Some(text) = right_text {
             f.render_widget(
                 Paragraph::new(text)
-                    .block(Block::new().borders(Borders::TOP).gray().title("CURRENT".bold()))
+                    .block(
+                        Block::new()
+                            .borders(Borders::TOP)
+                            .border_style(Style::new().fg(Color::Gray))
+                            .title_style(Style::new().bold())
+                            .title("CURRENT"),
+                    )
                     .scroll((0, self.scroll_x as u16)),
                 content_chunks[2],
             );
@@ -517,9 +541,13 @@ impl FunctionDiffUi {
 
         if self.three_way {
             if let Some(text) = prev_margin_text {
-                f.render_widget(text, content_chunks[3].inner(&Margin::new(1, 1)));
+                f.render_widget(text, content_chunks[3].inner(Margin::new(1, 1)));
             }
-            let block = Block::new().borders(Borders::TOP).gray().title("SAVED".bold());
+            let block = Block::new()
+                .borders(Borders::TOP)
+                .border_style(Style::new().fg(Color::Gray))
+                .title_style(Style::new().bold())
+                .title("SAVED");
             if let Some(text) = prev_text {
                 f.render_widget(
                     Paragraph::new(text).block(block.clone()).scroll((0, self.scroll_x as u16)),
@@ -533,7 +561,7 @@ impl FunctionDiffUi {
         // Render scrollbars
         f.render_stateful_widget(
             Scrollbar::new(ScrollbarOrientation::VerticalRight).begin_symbol(None).end_symbol(None),
-            chunks[1].inner(&Margin::new(0, 1)),
+            chunks[1].inner(Margin::new(0, 1)),
             &mut self.scroll_state_y,
         );
         f.render_stateful_widget(
@@ -589,7 +617,7 @@ impl FunctionDiffUi {
             Constraint::Percentage(percent_y),
             Constraint::Percentage((100 - percent_y) / 2),
         ])
-        .split(f.size())[1];
+        .split(f.area())[1];
         let popup_rect = Layout::horizontal([
             Constraint::Percentage((100 - percent_x) / 2),
             Constraint::Percentage(percent_x),
@@ -742,7 +770,7 @@ impl FunctionDiffUi {
         self.scroll_y += self.per_page / if half { 2 } else { 1 };
     }
 
-    #[allow(clippy::too_many_arguments)]
+    #[expect(clippy::too_many_arguments)]
     fn print_sym(
         &self,
         out: &mut Text<'static>,

objdiff-cli/src/cmd/report.rs

@@ -94,7 +94,7 @@ fn generate(args: GenerateArgs) -> Result<()> {
     };
     info!(
         "Generating report for {} units (using {} threads)",
-        project.objects.len(),
+        project.units().len(),
         if args.deduplicate { 1 } else { rayon::current_num_threads() }
     );
 
@@ -103,7 +103,7 @@ fn generate(args: GenerateArgs) -> Result<()> {
     let mut existing_functions: HashSet<String> = HashSet::new();
     if args.deduplicate {
         // If deduplicating, we need to run single-threaded
-        for object in &mut project.objects {
+        for object in project.units.as_deref_mut().unwrap_or_default() {
             if let Some(unit) = report_object(
                 object,
                 project_dir,
@@ -116,7 +116,9 @@ fn generate(args: GenerateArgs) -> Result<()> {
         }
     } else {
         let vec = project
-            .objects
+            .units
+            .as_deref_mut()
+            .unwrap_or_default()
             .par_iter_mut()
             .map(|object| {
                 report_object(
@@ -132,7 +134,7 @@ fn generate(args: GenerateArgs) -> Result<()> {
     }
     let measures = units.iter().flat_map(|u| u.measures.into_iter()).collect();
     let mut categories = Vec::new();
-    for category in &project.progress_categories {
+    for category in project.progress_categories() {
         categories.push(ReportCategory {
             id: category.id.clone(),
             name: category.name.clone(),
@@ -199,7 +201,7 @@ fn report_object(
             .unwrap_or_default(),
         auto_generated: object.metadata.as_ref().and_then(|m| m.auto_generated),
     };
-    let mut measures = Measures::default();
+    let mut measures = Measures { total_units: 1, ..Default::default() };
     let mut sections = vec![];
     let mut functions = vec![];
 
@@ -237,7 +239,7 @@ fn report_object(
         }
 
         for (symbol, symbol_diff) in section.symbols.iter().zip(&section_diff.symbols) {
-            if symbol.size == 0 {
+            if symbol.size == 0 || symbol.flags.0.contains(ObjSymbolFlags::Hidden) {
                 continue;
             }
             if let Some(existing_functions) = &mut existing_functions {
@@ -280,6 +282,7 @@ fn report_object(
     if metadata.complete.unwrap_or(false) {
         measures.complete_code = measures.total_code;
         measures.complete_data = measures.total_data;
+        measures.complete_units = 1;
     }
     measures.calc_fuzzy_match_percent();
     measures.calc_matched_percent();

objdiff-cli/src/main.rs

@@ -2,6 +2,12 @@ mod argp_version;
 mod cmd;
 mod util;
 
+// musl's allocator is very slow, so use mimalloc when targeting musl.
+// Otherwise, use the system allocator to avoid extra code size.
+#[cfg(target_env = "musl")]
+#[global_allocator]
+static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
+
 use std::{env, ffi::OsStr, fmt::Display, path::PathBuf, str::FromStr};
 
 use anyhow::{Error, Result};

objdiff-core/Cargo.toml

@@ -1,23 +1,24 @@
 [package]
 name = "objdiff-core"
-version = "2.0.0-beta.6"
-edition = "2021"
-rust-version = "1.70"
-authors = ["Luke Street <luke@street.dev>"]
-license = "MIT OR Apache-2.0"
-repository = "https://github.com/encounter/objdiff"
-readme = "../README.md"
+version.workspace = true
+edition.workspace = true
+rust-version.workspace = true
+authors.workspace = true
+license.workspace = true
+repository.workspace = true
+readme = "README.md"
 description = """
 A local diffing tool for decompilation projects.
 """
+documentation = "https://docs.rs/objdiff-core"
 
 [lib]
 crate-type = ["cdylib", "rlib"]
 
 [features]
 all = ["config", "dwarf", "mips", "ppc", "x86", "arm", "bindings"]
-any-arch = [] # Implicit, used to check if any arch is enabled
-config = ["globset", "semver", "serde_json", "serde_yaml"]
+any-arch = ["bimap"] # Implicit, used to check if any arch is enabled
+config = ["bimap", "globset", "semver", "serde_json", "serde_yaml"]
 dwarf = ["gimli"]
 mips = ["any-arch", "rabbitizer"]
 ppc = ["any-arch", "cwdemangle", "cwextab", "ppc750cl"]
@@ -26,51 +27,55 @@ arm = ["any-arch", "cpp_demangle", "unarm", "arm-attr"]
 bindings = ["serde_json", "prost", "pbjson"]
 wasm = ["bindings", "console_error_panic_hook", "console_log"]
 
+[package.metadata.docs.rs]
+features = ["all"]
+
 [dependencies]
-anyhow = "1.0.82"
-byteorder = "1.5.0"
-filetime = "0.2.23"
-flagset = "0.4.5"
-log = "0.4.21"
-memmap2 = "0.9.4"
-num-traits = "0.2.18"
-object = { version = "0.36.0", features = ["read_core", "std", "elf", "pe"], default-features = false }
-pbjson = { version = "0.7.0", optional = true }
-prost = { version = "0.13.1", optional = true }
-serde = { version = "1", features = ["derive"] }
-similar = { version = "2.5.0", default-features = false }
-strum = { version = "0.26.2", features = ["derive"] }
-wasm-bindgen = "0.2.93"
-tsify-next = { version = "0.5.4", default-features = false, features = ["js"] }
-console_log = { version = "1.0.0", optional = true }
-console_error_panic_hook = { version = "0.1.7", optional = true }
+anyhow = "1.0"
+bimap = { version = "0.6", features = ["serde"], optional = true }
+byteorder = "1.5"
+filetime = "0.2"
+flagset = "0.4"
+log = "0.4"
+memmap2 = "0.9"
+num-traits = "0.2"
+object = { version = "0.36", features = ["read_core", "std", "elf", "pe"], default-features = false }
+pbjson = { version = "0.7", optional = true }
+prost = { version = "0.13", optional = true }
+serde = { version = "1.0", features = ["derive"] }
+similar = { version = "2.6", default-features = false }
+strum = { version = "0.26", features = ["derive"] }
+wasm-bindgen = "0.2"
+tsify-next = { version = "0.5", default-features = false, features = ["js"] }
+console_log = { version = "1.0", optional = true }
+console_error_panic_hook = { version = "0.1", optional = true }
 
 # config
-globset = { version = "0.4.14", features = ["serde1"], optional = true }
-semver = { version = "1.0.22", optional = true }
-serde_json = { version = "1.0.116", optional = true }
-serde_yaml = { version = "0.9.34", optional = true }
+globset = { version = "0.4", features = ["serde1"], optional = true }
+semver = { version = "1.0", optional = true }
+serde_json = { version = "1.0", optional = true }
+serde_yaml = { version = "0.9", optional = true }
 
 # dwarf
-gimli = { version = "0.29.0", default-features = false, features = ["read-all"], optional = true }
+gimli = { version = "0.31", default-features = false, features = ["read-all"], optional = true }
 
 # ppc
-cwdemangle = { version = "1.0.0", optional = true }
-cwextab = { version = "0.2.3", optional = true }
-ppc750cl = { git = "https://github.com/encounter/ppc750cl", rev = "6cbd7d888c7082c2c860f66cbb9848d633f753ed", optional = true }
+cwdemangle = { version = "1.0", optional = true }
+cwextab = { version = "1.0.2", optional = true }
+ppc750cl = { version = "0.3", optional = true }
 
 # mips
-rabbitizer = { version = "1.11.0", optional = true }
+rabbitizer = { version = "1.12", optional = true }
 
 # x86
-cpp_demangle = { version = "0.4.3", optional = true }
-iced-x86 = { version = "1.21.0", default-features = false, features = ["std", "decoder", "intel", "gas", "masm", "nasm", "exhaustive_enums"], optional = true }
-msvc-demangler = { version = "0.10.0", optional = true }
+cpp_demangle = { version = "0.4", optional = true }
+iced-x86 = { version = "1.21", default-features = false, features = ["std", "decoder", "intel", "gas", "masm", "nasm", "exhaustive_enums"], optional = true }
+msvc-demangler = { version = "0.10", optional = true }
 
 # arm
-unarm = { version = "1.5.0", optional = true }
-arm-attr = { version = "0.1.1", optional = true }
+unarm = { version = "1.6", optional = true }
+arm-attr = { version = "0.1", optional = true }
 
 [build-dependencies]
-prost-build = "0.13.1"
-pbjson-build = "0.7.0"
+prost-build = "0.13"
+pbjson-build = "0.7"

objdiff-core/README.md

@@ -0,0 +1,14 @@
+# objdiff-core
+
+objdiff-core contains the core functionality of [objdiff](https://github.com/encounter/objdiff), a tool for comparing object files in decompilation projects. See the main repository for more information.
+
+## Crate feature flags
+
+- **`all`**: Enables all main features.
+- **`config`**: Enables objdiff configuration file support.
+- **`dwarf`**: Enables extraction of line number information from DWARF debug sections.
+- **`mips`**: Enables the MIPS backend powered by [rabbitizer](https://github.com/Decompollaborate/rabbitizer). (Note: C library with Rust bindings)
+- **`ppc`**: Enables the PowerPC backend powered by [ppc750cl](https://github.com/encounter/ppc750cl).
+- **`x86`**: Enables the x86 backend powered by [iced-x86](https://crates.io/crates/iced-x86).
+- **`arm`**: Enables the ARM backend powered by [unarm](https://github.com/AetiasHax/unarm).
+- **`bindings`**: Enables serialization and deserialization of objdiff data structures.

objdiff-core/protos/report.proto

@@ -32,6 +32,10 @@ message Measures {
   uint64 complete_data = 13;
   // Completed (or "linked") data percent
   float complete_data_percent = 14;
+  // Total number of units
+  uint32 total_units = 15;
+  // Completed (or "linked") units
+  uint32 complete_units = 16;
 }
 
 // Project progress report

objdiff-core/src/arch/mod.rs

@@ -1,11 +1,13 @@
-use std::{borrow::Cow, collections::BTreeMap};
+use std::{borrow::Cow, collections::BTreeMap, ffi::CStr};
 
 use anyhow::{bail, Result};
+use byteorder::ByteOrder;
 use object::{Architecture, File, Object, ObjectSymbol, Relocation, RelocationFlags, Symbol};
 
 use crate::{
     diff::DiffObjConfig,
     obj::{ObjIns, ObjReloc, ObjSection},
+    util::ReallySigned,
 };
 
 #[cfg(feature = "arm")]
@@ -17,6 +19,97 @@ pub mod ppc;
 #[cfg(feature = "x86")]
 pub mod x86;
 
+/// Represents the type of data associated with an instruction
+pub enum DataType {
+    Int8,
+    Int16,
+    Int32,
+    Int64,
+    Int128,
+    Float,
+    Double,
+    Bytes,
+    String,
+}
+
+impl DataType {
+    pub fn display_bytes<Endian: ByteOrder>(&self, bytes: &[u8]) -> Option<String> {
+        if self.required_len().is_some_and(|l| bytes.len() < l) {
+            return None;
+        }
+
+        match self {
+            DataType::Int8 => {
+                let i = i8::from_ne_bytes(bytes.try_into().unwrap());
+                if i < 0 {
+                    format!("Int8: {:#x} ({:#x})", i, ReallySigned(i))
+                } else {
+                    format!("Int8: {:#x}", i)
+                }
+            }
+            DataType::Int16 => {
+                let i = Endian::read_i16(bytes);
+                if i < 0 {
+                    format!("Int16: {:#x} ({:#x})", i, ReallySigned(i))
+                } else {
+                    format!("Int16: {:#x}", i)
+                }
+            }
+            DataType::Int32 => {
+                let i = Endian::read_i32(bytes);
+                if i < 0 {
+                    format!("Int32: {:#x} ({:#x})", i, ReallySigned(i))
+                } else {
+                    format!("Int32: {:#x}", i)
+                }
+            }
+            DataType::Int64 => {
+                let i = Endian::read_i64(bytes);
+                if i < 0 {
+                    format!("Int64: {:#x} ({:#x})", i, ReallySigned(i))
+                } else {
+                    format!("Int64: {:#x}", i)
+                }
+            }
+            DataType::Int128 => {
+                let i = Endian::read_i128(bytes);
+                if i < 0 {
+                    format!("Int128: {:#x} ({:#x})", i, ReallySigned(i))
+                } else {
+                    format!("Int128: {:#x}", i)
+                }
+            }
+            DataType::Float => {
+                format!("Float: {}", Endian::read_f32(bytes))
+            }
+            DataType::Double => {
+                format!("Double: {}", Endian::read_f64(bytes))
+            }
+            DataType::Bytes => {
+                format!("Bytes: {:#?}", bytes)
+            }
+            DataType::String => {
+                format!("String: {:?}", CStr::from_bytes_until_nul(bytes).ok()?)
+            }
+        }
+        .into()
+    }
+
+    fn required_len(&self) -> Option<usize> {
+        match self {
+            DataType::Int8 => Some(1),
+            DataType::Int16 => Some(2),
+            DataType::Int32 => Some(4),
+            DataType::Int64 => Some(8),
+            DataType::Int128 => Some(16),
+            DataType::Float => Some(4),
+            DataType::Double => Some(8),
+            DataType::Bytes => None,
+            DataType::String => None,
+        }
+    }
+}
+
 pub trait ObjArch: Send + Sync {
     fn process_code(
         &self,
@@ -41,6 +134,16 @@ pub trait ObjArch: Send + Sync {
     fn display_reloc(&self, flags: RelocationFlags) -> Cow<'static, str>;
 
     fn symbol_address(&self, symbol: &Symbol) -> u64 { symbol.address() }
+
+    fn guess_data_type(&self, _instruction: &ObjIns) -> Option<DataType> { None }
+
+    fn display_data_type(&self, _ty: DataType, bytes: &[u8]) -> Option<String> {
+        Some(format!("Bytes: {:#x?}", bytes))
+    }
+
+    // Downcast methods
+    #[cfg(feature = "ppc")]
+    fn ppc(&self) -> Option<&ppc::ObjArchPpc> { None }
 }
 
 pub struct ProcessCodeResult {
@@ -48,7 +151,7 @@ pub struct ProcessCodeResult {
     pub insts: Vec<ObjIns>,
 }
 
-pub fn new_arch(object: &object::File) -> Result<Box<dyn ObjArch>> {
+pub fn new_arch(object: &File) -> Result<Box<dyn ObjArch>> {
     Ok(match object.architecture() {
         #[cfg(feature = "ppc")]
         Architecture::PowerPc => Box::new(ppc::ObjArchPpc::new(object)?),

objdiff-core/src/arch/ppc.rs

@@ -1,13 +1,18 @@
 use std::{borrow::Cow, collections::BTreeMap};
 
-use anyhow::{bail, Result};
-use object::{elf, File, Relocation, RelocationFlags};
-use ppc750cl::{Argument, InsIter, GPR};
+use anyhow::{bail, ensure, Result};
+use byteorder::BigEndian;
+use cwextab::{decode_extab, ExceptionTableData};
+use object::{
+    elf, File, Object, ObjectSection, ObjectSymbol, Relocation, RelocationFlags, RelocationTarget,
+    Symbol, SymbolKind,
+};
+use ppc750cl::{Argument, InsIter, Opcode, GPR};
 
 use crate::{
-    arch::{ObjArch, ProcessCodeResult},
+    arch::{DataType, ObjArch, ProcessCodeResult},
     diff::DiffObjConfig,
-    obj::{ObjIns, ObjInsArg, ObjInsArgValue, ObjReloc, ObjSection},
+    obj::{ObjIns, ObjInsArg, ObjInsArgValue, ObjReloc, ObjSection, ObjSymbol},
 };
 
 // Relative relocation, can be Simm, Offset or BranchDest
@@ -22,10 +27,13 @@ fn is_rel_abs_arg(arg: &Argument) -> bool {
 
 fn is_offset_arg(arg: &Argument) -> bool { matches!(arg, Argument::Offset(_)) }
 
-pub struct ObjArchPpc {}
+pub struct ObjArchPpc {
+    /// Exception info
+    pub extab: Option<BTreeMap<usize, ExceptionInfo>>,
+}
 
 impl ObjArchPpc {
-    pub fn new(_file: &File) -> Result<Self> { Ok(Self {}) }
+    pub fn new(file: &File) -> Result<Self> { Ok(Self { extab: decode_exception_info(file)? }) }
 }
 
 impl ObjArch for ObjArchPpc {
@@ -178,6 +186,42 @@ impl ObjArch for ObjArchPpc {
             _ => Cow::Owned(format!("<{flags:?}>")),
         }
     }
+
+    fn guess_data_type(&self, instruction: &ObjIns) -> Option<super::DataType> {
+        // Always shows the first string of the table. Not ideal, but it's really hard to find
+        // the actual string being referenced.
+        if instruction.reloc.as_ref().is_some_and(|r| r.target.name.starts_with("@stringBase")) {
+            return Some(DataType::String);
+        }
+
+        match Opcode::from(instruction.op as u8) {
+            Opcode::Lbz | Opcode::Lbzu | Opcode::Lbzux | Opcode::Lbzx => Some(DataType::Int8),
+            Opcode::Lhz | Opcode::Lhzu | Opcode::Lhzux | Opcode::Lhzx => Some(DataType::Int16),
+            Opcode::Lha | Opcode::Lhau | Opcode::Lhaux | Opcode::Lhax => Some(DataType::Int16),
+            Opcode::Lwz | Opcode::Lwzu | Opcode::Lwzux | Opcode::Lwzx => Some(DataType::Int32),
+            Opcode::Lfs | Opcode::Lfsu | Opcode::Lfsux | Opcode::Lfsx => Some(DataType::Float),
+            Opcode::Lfd | Opcode::Lfdu | Opcode::Lfdux | Opcode::Lfdx => Some(DataType::Double),
+
+            Opcode::Stb | Opcode::Stbu | Opcode::Stbux | Opcode::Stbx => Some(DataType::Int8),
+            Opcode::Sth | Opcode::Sthu | Opcode::Sthux | Opcode::Sthx => Some(DataType::Int16),
+            Opcode::Stw | Opcode::Stwu | Opcode::Stwux | Opcode::Stwx => Some(DataType::Int32),
+            Opcode::Stfs | Opcode::Stfsu | Opcode::Stfsux | Opcode::Stfsx => Some(DataType::Float),
+            Opcode::Stfd | Opcode::Stfdu | Opcode::Stfdux | Opcode::Stfdx => Some(DataType::Double),
+            _ => None,
+        }
+    }
+
+    fn display_data_type(&self, ty: DataType, bytes: &[u8]) -> Option<String> {
+        ty.display_bytes::<BigEndian>(bytes)
+    }
+
+    fn ppc(&self) -> Option<&ObjArchPpc> { Some(self) }
+}
+
+impl ObjArchPpc {
+    pub fn extab_for_symbol(&self, symbol: &ObjSymbol) -> Option<&ExceptionInfo> {
+        symbol.original_index.and_then(|i| self.extab.as_ref()?.get(&i))
+    }
 }
 
 fn push_reloc(args: &mut Vec<ObjInsArg>, reloc: &ObjReloc) -> Result<()> {
@@ -208,3 +252,132 @@ fn push_reloc(args: &mut Vec<ObjInsArg>, reloc: &ObjReloc) -> Result<()> {
     };
     Ok(())
 }
+
+#[derive(Debug, Clone)]
+pub struct ExtabSymbolRef {
+    pub original_index: usize,
+    pub name: String,
+    pub demangled_name: Option<String>,
+}
+
+#[derive(Debug, Clone)]
+pub struct ExceptionInfo {
+    pub eti_symbol: ExtabSymbolRef,
+    pub etb_symbol: ExtabSymbolRef,
+    pub data: ExceptionTableData,
+    pub dtors: Vec<ExtabSymbolRef>,
+}
+
+fn decode_exception_info(file: &File<'_>) -> Result<Option<BTreeMap<usize, ExceptionInfo>>> {
+    let Some(extab_section) = file.section_by_name("extab") else {
+        return Ok(None);
+    };
+    let Some(extabindex_section) = file.section_by_name("extabindex") else {
+        return Ok(None);
+    };
+
+    let mut result = BTreeMap::new();
+    let extab_relocations = extab_section.relocations().collect::<BTreeMap<u64, Relocation>>();
+    let extabindex_relocations =
+        extabindex_section.relocations().collect::<BTreeMap<u64, Relocation>>();
+
+    for extabindex in file.symbols().filter(|symbol| {
+        symbol.section_index() == Some(extabindex_section.index())
+            && symbol.kind() == SymbolKind::Data
+    }) {
+        if extabindex.size() != 12 {
+            log::warn!("Invalid extabindex entry size {}", extabindex.size());
+            continue;
+        }
+
+        // Each extabindex entry has two relocations:
+        // - 0x0: The function that the exception table is for
+        // - 0x8: The relevant entry in extab section
+        let Some(extab_func_reloc) = extabindex_relocations.get(&extabindex.address()) else {
+            log::warn!("Failed to find function relocation for extabindex entry");
+            continue;
+        };
+        let Some(extab_reloc) = extabindex_relocations.get(&(extabindex.address() + 8)) else {
+            log::warn!("Failed to find extab relocation for extabindex entry");
+            continue;
+        };
+
+        // Resolve the function and extab symbols
+        let Some(extab_func) = relocation_symbol(file, extab_func_reloc)? else {
+            log::warn!("Failed to find function symbol for extabindex entry");
+            continue;
+        };
+        let extab_func_name = extab_func.name()?;
+        let Some(extab) = relocation_symbol(file, extab_reloc)? else {
+            log::warn!("Failed to find extab symbol for extabindex entry");
+            continue;
+        };
+
+        let extab_start_addr = extab.address() - extab_section.address();
+        let extab_end_addr = extab_start_addr + extab.size();
+
+        // All relocations in the extab section are dtors
+        let mut dtors: Vec<ExtabSymbolRef> = vec![];
+        for (_, reloc) in extab_relocations.range(extab_start_addr..extab_end_addr) {
+            let Some(symbol) = relocation_symbol(file, reloc)? else {
+                log::warn!("Failed to find symbol for extab relocation");
+                continue;
+            };
+            dtors.push(make_symbol_ref(&symbol)?);
+        }
+
+        // Decode the extab data
+        let Some(extab_data) = extab_section.data_range(extab_start_addr, extab.size())? else {
+            log::warn!("Failed to get extab data for function {}", extab_func_name);
+            continue;
+        };
+        let data = match decode_extab(extab_data) {
+            Ok(decoded_data) => decoded_data,
+            Err(e) => {
+                log::warn!(
+                    "Exception table decoding failed for function {}, reason: {}",
+                    extab_func_name,
+                    e.to_string()
+                );
+                return Ok(None);
+            }
+        };
+
+        //Add the new entry to the list
+        result.insert(extab_func.index().0, ExceptionInfo {
+            eti_symbol: make_symbol_ref(&extabindex)?,
+            etb_symbol: make_symbol_ref(&extab)?,
+            data,
+            dtors,
+        });
+    }
+
+    Ok(Some(result))
+}
+
+fn relocation_symbol<'data, 'file>(
+    file: &'file File<'data>,
+    relocation: &Relocation,
+) -> Result<Option<Symbol<'data, 'file>>> {
+    let addend = relocation.addend();
+    match relocation.target() {
+        RelocationTarget::Symbol(idx) => {
+            ensure!(addend == 0, "Symbol relocations must have zero addend");
+            Ok(Some(file.symbol_by_index(idx)?))
+        }
+        RelocationTarget::Section(idx) => {
+            ensure!(addend >= 0, "Section relocations must have non-negative addend");
+            let addend = addend as u64;
+            Ok(file
+                .symbols()
+                .find(|symbol| symbol.section_index() == Some(idx) && symbol.address() == addend))
+        }
+        target => bail!("Unsupported relocation target: {target:?}"),
+    }
+}
+
+fn make_symbol_ref(symbol: &Symbol) -> Result<ExtabSymbolRef> {
+    let name = symbol.name()?.to_string();
+    let demangled_name = cwdemangle::demangle(&name, &cwdemangle::DemangleOptions::default());
+    Ok(ExtabSymbolRef { original_index: symbol.index().0, name, demangled_name })
+}

objdiff-core/src/bindings/report.rs

@@ -8,9 +8,10 @@ use serde_json::error::Category;
 include!(concat!(env!("OUT_DIR"), "/objdiff.report.rs"));
 include!(concat!(env!("OUT_DIR"), "/objdiff.report.serde.rs"));
 
-pub const REPORT_VERSION: u32 = 1;
+pub const REPORT_VERSION: u32 = 2;
 
 impl Report {
+    /// Attempts to parse the report as binary protobuf or JSON.
     pub fn parse(data: &[u8]) -> Result<Self> {
         if data.is_empty() {
             bail!(std::io::Error::from(std::io::ErrorKind::UnexpectedEof));
@@ -25,6 +26,7 @@ impl Report {
         Ok(report)
     }
 
+    /// Attempts to parse the report as JSON, migrating from the legacy report format if necessary.
     fn from_json(bytes: &[u8]) -> Result<Self, serde_json::Error> {
         match serde_json::from_slice::<Self>(bytes) {
             Ok(report) => Ok(report),
@@ -43,16 +45,23 @@ impl Report {
         }
     }
 
+    /// Migrates the report to the latest version.
+    /// Fails if the report version is newer than supported.
     pub fn migrate(&mut self) -> Result<()> {
         if self.version == 0 {
             self.migrate_v0()?;
         }
+        if self.version == 1 {
+            self.migrate_v1()?;
+        }
         if self.version != REPORT_VERSION {
             bail!("Unsupported report version: {}", self.version);
         }
         Ok(())
     }
 
+    /// Adds `complete_code`, `complete_data`, `complete_code_percent`, and `complete_data_percent`
+    /// to measures, and sets `progress_categories` in unit metadata.
     fn migrate_v0(&mut self) -> Result<()> {
         let Some(measures) = &mut self.measures else {
             bail!("Missing measures in report");
@@ -61,15 +70,16 @@ impl Report {
             let Some(unit_measures) = &mut unit.measures else {
                 bail!("Missing measures in report unit");
             };
-            let Some(metadata) = &mut unit.metadata else {
-                bail!("Missing metadata in report unit");
+            let mut complete = false;
+            if let Some(metadata) = &mut unit.metadata {
+                if metadata.module_name.is_some() || metadata.module_id.is_some() {
+                    metadata.progress_categories = vec!["modules".to_string()];
+                } else {
+                    metadata.progress_categories = vec!["dol".to_string()];
+                }
+                complete = metadata.complete.unwrap_or(false);
             };
-            if metadata.module_name.is_some() || metadata.module_id.is_some() {
-                metadata.progress_categories = vec!["modules".to_string()];
-            } else {
-                metadata.progress_categories = vec!["dol".to_string()];
-            }
-            if metadata.complete.unwrap_or(false) {
+            if complete {
                 unit_measures.complete_code = unit_measures.total_code;
                 unit_measures.complete_data = unit_measures.total_data;
                 unit_measures.complete_code_percent = 100.0;
@@ -84,10 +94,42 @@ impl Report {
             measures.complete_data += unit_measures.complete_data;
         }
         measures.calc_matched_percent();
+        self.calculate_progress_categories();
         self.version = 1;
         Ok(())
     }
 
+    /// Adds `total_units` and `complete_units` to measures.
+    fn migrate_v1(&mut self) -> Result<()> {
+        let Some(total_measures) = &mut self.measures else {
+            bail!("Missing measures in report");
+        };
+        for unit in &mut self.units {
+            let Some(measures) = &mut unit.measures else {
+                bail!("Missing measures in report unit");
+            };
+            let complete = unit.metadata.as_ref().and_then(|m| m.complete).unwrap_or(false) as u32;
+            let progress_categories =
+                unit.metadata.as_ref().map(|m| m.progress_categories.as_slice()).unwrap_or(&[]);
+            measures.total_units = 1;
+            measures.complete_units = complete;
+            total_measures.total_units += 1;
+            total_measures.complete_units += complete;
+            for id in progress_categories {
+                if let Some(category) = self.categories.iter_mut().find(|c| &c.id == id) {
+                    let Some(measures) = &mut category.measures else {
+                        bail!("Missing measures in category");
+                    };
+                    measures.total_units += 1;
+                    measures.complete_units += complete;
+                }
+            }
+        }
+        self.version = 2;
+        Ok(())
+    }
+
+    /// Calculate progress categories based on unit metadata.
     pub fn calculate_progress_categories(&mut self) {
         for unit in &self.units {
             let Some(metadata) = unit.metadata.as_ref() else {
@@ -117,6 +159,72 @@ impl Report {
             measures.calc_matched_percent();
         }
     }
+
+    /// Split the report into multiple reports based on progress categories.
+    /// Assumes progress categories are in the format `version`, `version.category`.
+    /// This is a hack for projects that generate all versions in a single report.
+    pub fn split(self) -> Vec<(String, Report)> {
+        let mut reports = Vec::new();
+        // Map units to Option to allow taking ownership
+        let mut units = self.units.into_iter().map(Some).collect::<Vec<_>>();
+        for category in &self.categories {
+            if category.id.contains(".") {
+                // Skip subcategories
+                continue;
+            }
+            fn is_sub_category(id: &str, parent: &str, sep: char) -> bool {
+                id.starts_with(parent)
+                    && id.get(parent.len()..).map_or(false, |s| s.starts_with(sep))
+            }
+            let mut sub_categories = self
+                .categories
+                .iter()
+                .filter(|c| is_sub_category(&c.id, &category.id, '.'))
+                .cloned()
+                .collect::<Vec<_>>();
+            // Remove category prefix
+            for sub_category in &mut sub_categories {
+                sub_category.id = sub_category.id[category.id.len() + 1..].to_string();
+            }
+            let mut sub_units = units
+                .iter_mut()
+                .filter_map(|opt| {
+                    let unit = opt.as_mut()?;
+                    let metadata = unit.metadata.as_ref()?;
+                    if metadata.progress_categories.contains(&category.id) {
+                        opt.take()
+                    } else {
+                        None
+                    }
+                })
+                .collect::<Vec<_>>();
+            for sub_unit in &mut sub_units {
+                // Remove leading version/ from unit name
+                if let Some(name) =
+                    sub_unit.name.strip_prefix(&category.id).and_then(|s| s.strip_prefix('/'))
+                {
+                    sub_unit.name = name.to_string();
+                }
+                // Filter progress categories
+                let Some(metadata) = sub_unit.metadata.as_mut() else {
+                    continue;
+                };
+                metadata.progress_categories = metadata
+                    .progress_categories
+                    .iter()
+                    .filter(|c| is_sub_category(c, &category.id, '.'))
+                    .map(|c| c[category.id.len() + 1..].to_string())
+                    .collect();
+            }
+            reports.push((category.id.clone(), Report {
+                measures: category.measures,
+                units: sub_units,
+                version: self.version,
+                categories: sub_categories,
+            }));
+        }
+        reports
+    }
 }
 
 impl Measures {
@@ -176,6 +284,8 @@ impl AddAssign for Measures {
         self.matched_functions += other.matched_functions;
         self.complete_code += other.complete_code;
         self.complete_data += other.complete_data;
+        self.total_units += other.total_units;
+        self.complete_units += other.complete_units;
     }
 }
 

objdiff-core/src/config/mod.rs

@@ -1,77 +1,100 @@
 use std::{
+    fs,
     fs::File,
-    io::Read,
+    io::{BufReader, BufWriter, Read},
     path::{Path, PathBuf},
 };
 
 use anyhow::{anyhow, Context, Result};
+use bimap::BiBTreeMap;
 use filetime::FileTime;
 use globset::{Glob, GlobSet, GlobSetBuilder};
 
-#[inline]
-fn bool_true() -> bool { true }
-
-#[derive(Default, Clone, serde::Deserialize)]
+#[derive(Default, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ProjectConfig {
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub min_version: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub custom_make: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub custom_args: Option<Vec<String>>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub target_dir: Option<PathBuf>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub base_dir: Option<PathBuf>,
-    #[serde(default = "bool_true")]
-    pub build_base: bool,
-    #[serde(default)]
-    pub build_target: bool,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub build_base: Option<bool>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub build_target: Option<bool>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub watch_patterns: Option<Vec<Glob>>,
-    #[serde(default, alias = "units")]
-    pub objects: Vec<ProjectObject>,
-    #[serde(default)]
-    pub progress_categories: Vec<ProjectProgressCategory>,
+    #[serde(default, alias = "objects", skip_serializing_if = "Option::is_none")]
+    pub units: Option<Vec<ProjectObject>>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub progress_categories: Option<Vec<ProjectProgressCategory>>,
 }
 
-#[derive(Default, Clone, serde::Deserialize)]
+impl ProjectConfig {
+    #[inline]
+    pub fn units(&self) -> &[ProjectObject] { self.units.as_deref().unwrap_or_default() }
+
+    #[inline]
+    pub fn units_mut(&mut self) -> &mut Vec<ProjectObject> {
+        self.units.get_or_insert_with(Vec::new)
+    }
+
+    #[inline]
+    pub fn progress_categories(&self) -> &[ProjectProgressCategory] {
+        self.progress_categories.as_deref().unwrap_or_default()
+    }
+
+    #[inline]
+    pub fn progress_categories_mut(&mut self) -> &mut Vec<ProjectProgressCategory> {
+        self.progress_categories.get_or_insert_with(Vec::new)
+    }
+}
+
+#[derive(Default, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ProjectObject {
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub name: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub path: Option<PathBuf>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub target_path: Option<PathBuf>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub base_path: Option<PathBuf>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     #[deprecated(note = "Use metadata.reverse_fn_order")]
     pub reverse_fn_order: Option<bool>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     #[deprecated(note = "Use metadata.complete")]
     pub complete: Option<bool>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub scratch: Option<ScratchConfig>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub metadata: Option<ProjectObjectMetadata>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub symbol_mappings: Option<SymbolMappings>,
 }
 
-#[derive(Default, Clone, serde::Deserialize)]
+pub type SymbolMappings = BiBTreeMap<String, String>;
+
+#[derive(Default, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ProjectObjectMetadata {
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub complete: Option<bool>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub reverse_fn_order: Option<bool>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub source_path: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub progress_categories: Option<Vec<String>>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub auto_generated: Option<bool>,
 }
 
-#[derive(Default, Clone, serde::Deserialize)]
+#[derive(Default, Clone, serde::Serialize, serde::Deserialize)]
 pub struct ProjectProgressCategory {
     #[serde(default)]
     pub id: String,
@@ -112,32 +135,36 @@ impl ProjectObject {
     }
 
     pub fn complete(&self) -> Option<bool> {
-        #[allow(deprecated)]
+        #[expect(deprecated)]
         self.metadata.as_ref().and_then(|m| m.complete).or(self.complete)
     }
 
     pub fn reverse_fn_order(&self) -> Option<bool> {
-        #[allow(deprecated)]
+        #[expect(deprecated)]
         self.metadata.as_ref().and_then(|m| m.reverse_fn_order).or(self.reverse_fn_order)
     }
 
     pub fn hidden(&self) -> bool {
         self.metadata.as_ref().and_then(|m| m.auto_generated).unwrap_or(false)
     }
+
+    pub fn source_path(&self) -> Option<&String> {
+        self.metadata.as_ref().and_then(|m| m.source_path.as_ref())
+    }
 }
 
 #[derive(Default, Clone, Eq, PartialEq, serde::Deserialize, serde::Serialize)]
 pub struct ScratchConfig {
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub platform: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub compiler: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub c_flags: Option<String>,
-    #[serde(default)]
+    #[serde(default, skip_serializing_if = "Option::is_none")]
     pub ctx_path: Option<PathBuf>,
-    #[serde(default)]
-    pub build_ctx: bool,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub build_ctx: Option<bool>,
 }
 
 pub const CONFIG_FILENAMES: [&str; 3] = ["objdiff.json", "objdiff.yml", "objdiff.yaml"];
@@ -150,13 +177,13 @@ pub const DEFAULT_WATCH_PATTERNS: &[&str] = &[
 #[derive(Clone, Eq, PartialEq)]
 pub struct ProjectConfigInfo {
     pub path: PathBuf,
-    pub timestamp: FileTime,
+    pub timestamp: Option<FileTime>,
 }
 
 pub fn try_project_config(dir: &Path) -> Option<(Result<ProjectConfig>, ProjectConfigInfo)> {
     for filename in CONFIG_FILENAMES.iter() {
         let config_path = dir.join(filename);
-        let Ok(mut file) = File::open(&config_path) else {
+        let Ok(file) = File::open(&config_path) else {
             continue;
         };
         let metadata = file.metadata();
@@ -165,9 +192,10 @@ pub fn try_project_config(dir: &Path) -> Option<(Result<ProjectConfig>, ProjectC
                 continue;
             }
             let ts = FileTime::from_last_modification_time(&metadata);
+            let mut reader = BufReader::new(file);
             let mut result = match filename.contains("json") {
-                true => read_json_config(&mut file),
-                false => read_yml_config(&mut file),
+                true => read_json_config(&mut reader),
+                false => read_yml_config(&mut reader),
             };
             if let Ok(config) = &result {
                 // Validate min_version if present
@@ -175,12 +203,41 @@ pub fn try_project_config(dir: &Path) -> Option<(Result<ProjectConfig>, ProjectC
                     result = Err(e);
                 }
             }
-            return Some((result, ProjectConfigInfo { path: config_path, timestamp: ts }));
+            return Some((result, ProjectConfigInfo { path: config_path, timestamp: Some(ts) }));
         }
     }
     None
 }
 
+pub fn save_project_config(
+    config: &ProjectConfig,
+    info: &ProjectConfigInfo,
+) -> Result<ProjectConfigInfo> {
+    if let Some(last_ts) = info.timestamp {
+        // Check if the file has changed since we last read it
+        if let Ok(metadata) = fs::metadata(&info.path) {
+            let ts = FileTime::from_last_modification_time(&metadata);
+            if ts != last_ts {
+                return Err(anyhow!("Config file has changed since last read"));
+            }
+        }
+    }
+    let mut writer =
+        BufWriter::new(File::create(&info.path).context("Failed to create config file")?);
+    let ext = info.path.extension().and_then(|ext| ext.to_str()).unwrap_or("json");
+    match ext {
+        "json" => serde_json::to_writer_pretty(&mut writer, config).context("Failed to write JSON"),
+        "yml" | "yaml" => {
+            serde_yaml::to_writer(&mut writer, config).context("Failed to write YAML")
+        }
+        _ => Err(anyhow!("Unknown config file extension: {ext}")),
+    }?;
+    let file = writer.into_inner().context("Failed to flush file")?;
+    let metadata = file.metadata().context("Failed to get file metadata")?;
+    let ts = FileTime::from_last_modification_time(&metadata);
+    Ok(ProjectConfigInfo { path: info.path.clone(), timestamp: Some(ts) })
+}
+
 fn validate_min_version(config: &ProjectConfig) -> Result<()> {
     let Some(min_version) = &config.min_version else { return Ok(()) };
     let version = semver::Version::parse(env!("CARGO_PKG_VERSION"))