Honor Go settings written with go env -w (#344)

crates/socket-patch-cli/CLI_CONTRACT.md

@@ -1143,7 +1143,7 @@ Env-only knobs used for hosted upstream restoration and JVM metadata verificatio
 | Env var | Default | Notes |
 |---|---|---|
 | `SOCKET_NPM_REGISTRY` | `https://registry.npmjs.org` | Base for npm version documents (`<base>/<name>/<version>`, a scoped name's `/` as `%2f`; `dist.tarball` / `integrity` / `shasum`) the npm-family and vlt upstream restore reads, unless the project names another registry (yarn berry `npmRegistryServer`, pnpm's lock-sibling `.npmrc` `registry` / `@scope:registry` or pnpm-workspace.yaml `registry` / `registries` as the pnpm major reads them, vlt's node registry), whose document is read first. |
-| `SOCKET_GOPROXY` | `https://proxy.golang.org` | Go module proxy used for upstream restoration, honoring `GOPROXY`, `GONOPROXY` and `GOPRIVATE`. |
+| `SOCKET_GOPROXY` | `https://proxy.golang.org` | Go module proxy used for upstream restoration, honoring `GOPROXY`, `GONOPROXY` and `GOPRIVATE`. Like every Go setting socket-patch reads (`GOMODCACHE`, `GOPATH`, `GOWORK`, `GOSUMDB`, `GONOSUMDB`), each resolves as go resolves it: the environment, then the `go env -w` file (`$GOENV`, default `os.UserConfigDir()/go/env`; `GOENV=off` disables it), then go's default. |
 | `SOCKET_MAVEN_REGISTRY` | `https://repo1.maven.org/maven2` | maven2 base for the fallback upstream-pom download. |
 | `SOCKET_CRATES_INDEX` | `https://index.crates.io` | v5.0 upstream restore: the crates.io sparse index whose `checksum` a restored `Cargo.lock` entry gets back. |
 | `SOCKET_GOSUMDB_URL` | `https://sum.golang.org` | v5.0 upstream restore: the checksum database the restored go.sum lines are checked against. Without it, `GOSUMDB=off` or a module matching `GONOSUMDB` (default `GOPRIVATE`) skips the database and the hashes come from the module proxy's bytes alone (`SOCKET_GOPROXY` above). |

crates/socket-patch-cli/tests/e2e_golang.rs

@@ -282,6 +282,83 @@ async fn scan_discovers_go_modules() {
     );
 }
 
+/// #344: a module cache configured with `go env -w GOMODCACHE=…` (the
+/// `$GOENV` file, not the environment) is the one scanned, exactly as
+/// `go env GOMODCACHE` and `go build` resolve it.
+#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
+async fn scan_finds_the_cache_named_by_the_go_env_file() {
+    let server = MockServer::start().await;
+    mount_batch(&server).await;
+    let api_url = server.uri();
+
+    let dir = tempfile::tempdir().unwrap();
+    let cache_dir = dir.path().join("configured-cache");
+    let gin_dir = cache_dir.join("github.com/gin-gonic/gin@v1.9.1");
+    std::fs::create_dir_all(&gin_dir).unwrap();
+    std::fs::write(
+        gin_dir.join("go.mod"),
+        "module github.com/gin-gonic/gin\n\ngo 1.21\n",
+    )
+    .unwrap();
+    let project = dir.path().join("project");
+    std::fs::create_dir_all(&project).unwrap();
+    std::fs::write(
+        project.join("go.mod"),
+        "module example.com/myproject\n\ngo 1.21\n",
+    )
+    .unwrap();
+    // What `go env -w GOMODCACHE=<cache>` writes.
+    let goenv = dir.path().join("goenv");
+    std::fs::write(&goenv, format!("GOMODCACHE={}\n", cache_dir.display())).unwrap();
+    // HOME points at an empty dir so the `$HOME/go/pkg/mod` default finds
+    // nothing.
+    let home = dir.path().join("home");
+    std::fs::create_dir_all(&home).unwrap();
+
+    let (project2, api) = (project.clone(), api_url.clone());
+    let output = tokio::task::spawn_blocking(move || {
+        common::hermetic::command(&binary())
+            .args(["scan", "--json", "--cwd", project2.to_str().unwrap()])
+            .current_dir(&project2)
+            .env("GOENV", &goenv)
+            .env("HOME", &home)
+            .env("USERPROFILE", &home)
+            .env("SOCKET_API_URL", &api)
+            .env("SOCKET_API_TOKEN", "sktsec_dummy_e2e_golang_token_api")
+            .env("SOCKET_ORG_SLUG", ORG)
+            .env_remove("GOMODCACHE")
+            .env_remove("GOPATH")
+            .env_remove("SOCKET_ECOSYSTEMS")
+            .env_remove("SOCKET_GLOBAL")
+            .env_remove("SOCKET_GLOBAL_PREFIX")
+            .env_remove("SOCKET_JSON")
+            .env_remove("SOCKET_SILENT")
+            .env_remove("SOCKET_OFFLINE")
+            .env_remove("SOCKET_PROXY_URL")
+            .env_remove("SOCKET_BATCH_SIZE")
+            .output()
+            .expect("Failed to run socket-patch binary")
+    })
+    .await
+    .unwrap();
+    let stdout = String::from_utf8_lossy(&output.stdout);
+    let json: serde_json::Value = serde_json::from_str(&stdout).unwrap_or_else(|e| {
+        panic!(
+            "scan --json must emit JSON ({e}):\n{stdout}\n{}",
+            String::from_utf8_lossy(&output.stderr)
+        )
+    });
+    assert_eq!(
+        json["scannedPackages"], 1,
+        "the go env -w GOMODCACHE cache must be crawled; got:\n{json:#}"
+    );
+    let purls = batched_purls(&server).await;
+    assert_eq!(
+        purls,
+        BTreeSet::from(["pkg:golang/github.com/gin-gonic/gin@v1.9.1".to_string()])
+    );
+}
+
 /// Verify `socket-patch scan` discovers AND case-decodes Go modules.
 ///
 /// Go's module cache stores uppercase letters as `!`+lowercase, so

crates/socket-patch-core/src/crawlers/go_crawler.rs

@@ -175,14 +175,14 @@ impl GoCrawler {
     // ------------------------------------------------------------------
 
     /// Get `GOMODCACHE`, falling back to `$GOPATH/pkg/mod/` or `$HOME/go/pkg/mod/`.
+    /// Each setting resolves as go resolves it: the environment, then the
+    /// `go env -w` file ([`crate::utils::go_env`]), so a cache configured
+    /// with `go env -w GOMODCACHE=…` is the one crawled (#344).
     fn get_gomodcache() -> Option<PathBuf> {
-        if let Ok(cache) = std::env::var("GOMODCACHE") {
-            let p = PathBuf::from(cache);
-            if !p.as_os_str().is_empty() {
-                return Some(p);
-            }
+        if let Some(cache) = crate::utils::go_env::go_env("GOMODCACHE") {
+            return Some(PathBuf::from(cache));
         }
-        if let Ok(gopath) = std::env::var("GOPATH") {
+        if let Some(gopath) = crate::utils::go_env::go_env("GOPATH") {
             // GOPATH may list several directories separated by the OS path
             // separator (`:` on Unix, `;` on Windows). Go uses the FIRST
             // entry for the module cache, so split rather than treating the
@@ -314,12 +314,13 @@ fn go_sum_scope(cwd: &Path) -> Option<Vec<(String, String)>> {
 
 /// Whether the go command would build `cwd` in workspace mode: `GOWORK`
 /// names a file, or (`GOWORK` unset or empty) a `go.work` exists in `cwd`
-/// or an ancestor. `GOWORK=off` disables workspaces.
+/// or an ancestor. `GOWORK=off` disables workspaces. `GOWORK` resolves
+/// from the environment, then the `go env -w` file.
 fn workspace_in_effect(cwd: &Path) -> bool {
-    match std::env::var_os("GOWORK") {
+    match crate::utils::go_env::go_env("GOWORK") {
         Some(v) if v == "off" => false,
-        Some(v) if !v.is_empty() => true,
-        _ => go_work_at_or_above(cwd, &std::env::current_dir().unwrap_or_default()),
+        Some(_) => true,
+        None => go_work_at_or_above(cwd, &std::env::current_dir().unwrap_or_default()),
     }
 }
 
@@ -1565,6 +1566,8 @@ mod tests {
         // own project. Twin of `m2_repo_path`'s / `nuget_home`'s guards.
         let gopath_a = tempfile::tempdir().unwrap();
         let home_b = tempfile::tempdir().unwrap();
+        // No `go env -w` file: the developer's own must not leak in.
+        let _goenv = EnvGuard::set("GOENV", "off");
 
         // Case A: empty GOMODCACHE falls through to GOPATH, and the empty
         // FIRST GOPATH entry is skipped in favor of the next non-empty one
@@ -1602,6 +1605,33 @@ mod tests {
         );
     }
 
+    /// #344: GOMODCACHE / GOPATH written with `go env -w` (the `$GOENV`
+    /// file) locate the cache go itself uses; the environment still wins.
+    #[test]
+    #[serial_test::serial]
+    fn test_get_gomodcache_reads_the_go_env_file() {
+        let dir = tempfile::tempdir().unwrap();
+        let file = dir.path().join("env");
+        let cache = dir.path().join("custom-cache");
+        let gopath = dir.path().join("gopath");
+        let _gomodcache = EnvGuard::unset("GOMODCACHE");
+        let _gopath = EnvGuard::unset("GOPATH");
+        let _goenv = EnvGuard::set("GOENV", file.to_str().unwrap());
+
+        std::fs::write(&file, format!("GOMODCACHE={}\n", cache.display())).unwrap();
+        assert_eq!(GoCrawler::get_gomodcache(), Some(cache.clone()));
+
+        std::fs::write(&file, format!("GOPATH={}\n", gopath.display())).unwrap();
+        assert_eq!(
+            GoCrawler::get_gomodcache(),
+            Some(gopath.join("pkg").join("mod"))
+        );
+
+        let env_cache = dir.path().join("env-cache");
+        let _env = EnvGuard::set("GOMODCACHE", env_cache.to_str().unwrap());
+        assert_eq!(GoCrawler::get_gomodcache(), Some(env_cache));
+    }
+
     #[tokio::test]
     async fn test_parse_versioned_dir_second_visit_same_purl_returns_none() {
         // The `seen` dedup contract: crawl_all threads one HashSet through

crates/socket-patch-core/src/patch/redirect/upstream/client.rs

@@ -691,15 +691,17 @@ pub(crate) const DEFAULT_GOSUMDB: &str = "https://sum.golang.org";
 /// The checksum database go would consult for `module`, or `None` when go
 /// would not (`GOSUMDB=off`, or the module matches `GONOSUMDB` /
 /// `GOPRIVATE`) — the hashes are then computed from the module proxy's
-/// bytes instead. An explicit `SOCKET_GOSUMDB_URL` always wins.
+/// bytes instead. An explicit `SOCKET_GOSUMDB_URL` always wins. The Go
+/// settings resolve from the environment, then the `go env -w` file
+/// ([`crate::utils::go_env`]).
 fn gosumdb_base(module: &str) -> Option<String> {
     if let Ok(v) = std::env::var("SOCKET_GOSUMDB_URL") {
         let v = v.trim().trim_end_matches('/').to_string();
         if !v.is_empty() {
             return Some(v);
         }
     }
-    let nonempty = |key: &str| std::env::var(key).ok().filter(|v| !v.trim().is_empty());
+    let nonempty = |key: &str| crate::utils::go_env::go_env(key).filter(|v| !v.trim().is_empty());
     if nonempty("GOSUMDB").is_some_and(|v| v.trim() == "off") {
         return None;
     }

crates/socket-patch-core/src/utils/go_env.rs

@@ -0,0 +1,183 @@
+//! Go configuration lookup the way the `go` command resolves it (#344).
+//!
+//! go reads each setting from the process environment first, then from the
+//! per-user Go environment file that `go env -w` writes, then falls back to
+//! its built-in default. The Go docs tell users to configure `GOPRIVATE`,
+//! `GOPROXY` and `GOMODCACHE` with `go env -w`, so consulting only the
+//! environment silently ignores those settings: a private module path is
+//! requested from `proxy.golang.org`, a corporate mirror is bypassed, and a
+//! relocated module cache is not found.
+//!
+//! The file is `$GOENV` when set (`GOENV=off` disables it), else
+//! `os.UserConfigDir()/go/env`: `$XDG_CONFIG_HOME/go/env` (or
+//! `~/.config/go/env`) on Linux and other Unix, `~/Library/Application
+//! Support/go/env` on macOS, `%AppData%\go\env` on Windows.
+//!
+//! Out of scope: `$GOROOT/go.env` (Go 1.21+), the toolchain's own lowest
+//! layer. Upstream Go ships it with the same values as the built-in
+//! defaults, and locating it would mean running `go`.
+
+use std::path::PathBuf;
+
+/// The effective value of Go setting `key`: a non-empty environment variable
+/// wins, else the Go environment file's non-empty value, else `None` (the
+/// caller applies go's default). An empty environment variable falls
+/// through to the file, as go's own `cfg.Getenv` does.
+pub(crate) fn go_env(key: &str) -> Option<String> {
+    go_env_with(key, |k| std::env::var(k).ok())
+}
+
+/// [`go_env`] over an injected environment lookup.
+pub(crate) fn go_env_with(key: &str, env: impl Fn(&str) -> Option<String>) -> Option<String> {
+    if let Some(v) = env(key).filter(|v| !v.is_empty()) {
+        return Some(v);
+    }
+    let file = go_env_file(&env)?;
+    let text = crate::utils::fs::read_regular_to_string_sync(&file).ok()?;
+    lookup_in_env_file(&text, key).filter(|v| !v.is_empty())
+}
+
+/// The Go environment file go would read, or `None` when `GOENV=off` or no
+/// user config directory resolves.
+fn go_env_file(env: &impl Fn(&str) -> Option<String>) -> Option<PathBuf> {
+    match env("GOENV").filter(|v| !v.is_empty()) {
+        Some(v) if v == "off" => None,
+        Some(v) => Some(PathBuf::from(v)),
+        None => Some(user_config_dir(env)?.join("go").join("env")),
+    }
+}
+
+/// Go's `os.UserConfigDir`.
+fn user_config_dir(env: &impl Fn(&str) -> Option<String>) -> Option<PathBuf> {
+    let absolute = |v: Option<String>| {
+        v.filter(|v| !v.is_empty())
+            .map(PathBuf::from)
+            .filter(|p| p.is_absolute())
+    };
+    if cfg!(windows) {
+        return absolute(env("AppData").or_else(|| env("APPDATA")));
+    }
+    let home = || absolute(env("HOME"));
+    if cfg!(target_os = "macos") {
+        return Some(home()?.join("Library").join("Application Support"));
+    }
+    match env("XDG_CONFIG_HOME").filter(|v| !v.is_empty()) {
+        // go refuses a relative XDG_CONFIG_HOME rather than falling back.
+        Some(xdg) => absolute(Some(xdg)),
+        None => Some(home()?.join(".config")),
+    }
+}
+
+/// `key`'s value in a Go environment file (`cmd/go/internal/cfg`
+/// `readEnvFile`): one `KEY=VALUE` per line, the value taken verbatim after
+/// the first `=`; a line not starting with an uppercase letter (blank, a
+/// `#` comment) or without `=` is skipped. A later line for the same key
+/// wins.
+fn lookup_in_env_file(text: &str, key: &str) -> Option<String> {
+    text.lines()
+        .map(|line| line.strip_suffix('\r').unwrap_or(line))
+        .filter(|line| line.starts_with(|c: char| c.is_ascii_uppercase()))
+        .filter_map(|line| line.split_once('='))
+        .rfind(|(k, _)| *k == key)
+        .map(|(_, v)| v.to_string())
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::collections::HashMap;
+
+    fn env_of(pairs: &[(&str, &str)]) -> impl Fn(&str) -> Option<String> {
+        let map: HashMap<String, String> = pairs
+            .iter()
+            .map(|(k, v)| (k.to_string(), v.to_string()))
+            .collect();
+        move |k: &str| map.get(k).cloned()
+    }
+
+    #[test]
+    fn parses_go_env_w_output() {
+        let text = "# comment\nGOPROXY=http://127.0.0.1:18702\r\nGOPRIVATE=example.com,*.corp\n\
+                    lower=x\nNOEQUALS\nGOFLAGS=-mod=mod -tags=a=b\nGOPROXY=https://mirror\n";
+        assert_eq!(
+            lookup_in_env_file(text, "GOPROXY").as_deref(),
+            Some("https://mirror")
+        );
+        assert_eq!(
+            lookup_in_env_file(text, "GOPRIVATE").as_deref(),
+            Some("example.com,*.corp")
+        );
+        assert_eq!(
+            lookup_in_env_file(text, "GOFLAGS").as_deref(),
+            Some("-mod=mod -tags=a=b")
+        );
+        assert_eq!(lookup_in_env_file(text, "lower"), None);
+        assert_eq!(lookup_in_env_file(text, "NOEQUALS"), None);
+        assert_eq!(lookup_in_env_file(text, "GONOPROXY"), None);
+    }
+
+    /// env (non-empty) → `$GOENV` file → `None`; an empty env var falls
+    /// through to the file and `GOENV=off` disables it.
+    #[test]
+    fn env_then_goenv_file_then_default() {
+        let dir = tempfile::tempdir().unwrap();
+        let file = dir.path().join("env");
+        std::fs::write(&file, "GOPRIVATE=example.com\nGOMODCACHE=/cache\n").unwrap();
+        let goenv = file.to_str().unwrap();
+
+        let env = env_of(&[("GOENV", goenv)]);
+        assert_eq!(
+            go_env_with("GOPRIVATE", &env).as_deref(),
+            Some("example.com")
+        );
+        assert_eq!(go_env_with("GOPROXY", &env), None);
+
+        let env = env_of(&[("GOENV", goenv), ("GOPRIVATE", "env.example")]);
+        assert_eq!(
+            go_env_with("GOPRIVATE", &env).as_deref(),
+            Some("env.example")
+        );
+        let env = env_of(&[("GOENV", goenv), ("GOPRIVATE", "")]);
+        assert_eq!(
+            go_env_with("GOPRIVATE", &env).as_deref(),
+            Some("example.com")
+        );
+
+        let env = env_of(&[("GOENV", "off"), ("HOME", "/nonexistent")]);
+        assert_eq!(go_env_with("GOPRIVATE", &env), None);
+        let env = env_of(&[("GOENV", dir.path().join("missing").to_str().unwrap())]);
+        assert_eq!(go_env_with("GOPRIVATE", &env), None);
+    }
+
+    /// Without `GOENV` the file is `os.UserConfigDir()/go/env`.
+    #[test]
+    fn default_file_is_under_the_user_config_dir() {
+        let dir = tempfile::tempdir().unwrap();
+        let base = dir.path();
+        let config = if cfg!(windows) {
+            base.join("roaming")
+        } else if cfg!(target_os = "macos") {
+            base.join("Library").join("Application Support")
+        } else {
+            base.join("xdg")
+        };
+        std::fs::create_dir_all(config.join("go")).unwrap();
+        std::fs::write(config.join("go").join("env"), "GOPROXY=https://mirror\n").unwrap();
+        let env = env_of(&[
+            ("HOME", base.to_str().unwrap()),
+            ("XDG_CONFIG_HOME", base.join("xdg").to_str().unwrap()),
+            ("AppData", base.join("roaming").to_str().unwrap()),
+        ]);
+        assert_eq!(
+            go_env_with("GOPROXY", &env).as_deref(),
+            Some("https://mirror")
+        );
+        // A relative XDG_CONFIG_HOME / HOME resolves nothing.
+        let env = env_of(&[
+            ("HOME", "rel"),
+            ("XDG_CONFIG_HOME", "rel"),
+            ("AppData", "rel"),
+        ]);
+        assert_eq!(go_env_with("GOPROXY", &env), None);
+    }
+}

crates/socket-patch-core/src/utils/mod.rs

@@ -7,6 +7,7 @@ pub(crate) mod durability;
 pub mod env_compat;
 pub mod failpoint;
 pub mod fs;
+pub(crate) mod go_env;
 pub mod group_commit;
 pub(crate) mod http;
 pub(crate) mod line_endings;