// examples).
//
// The Dir, Target, Shlib, Root, ConflictDir, and Export file paths
// are all absolute paths.
//
// By default, the lists GoFiles, CgoFiles, and so on hold names of files in Dir
// (that is, paths relative to Dir, not absolute paths).
// The generated files added when using the -compiled and -test flags
// are absolute paths referring to cached copies of generated Go source files.

// import paths. We do this resolution lazily both to avoid file system work
// and because the eventual real load of the test imports (during 'go test')
// can produce better error messages if it starts with the original paths.
// The initial load of p loads all the non-test imports and rewrites
// the vendored paths, so nothing should ever call p.vendored(p.Imports).

		t.Logf("Missing expected secret paths:\n%s", strings.Join(sets.List[string](missingPaths), "\n"))
		t.Error("Missing expected secret paths. Verify VisitPodSecretNames() is correctly finding the missing paths, then correct expectedSecretPaths")
	}
	if extraPaths := secretPaths.Difference(expectedSecretPaths); len(extraPaths) > 0 {
		t.Logf("Extra secret paths:\n%s", strings.Join(sets.List[string](extraPaths), "\n"))

			},
		},
		{
			// Build the following graph, which contains a cross-signature loop
			// (A and C cross sign each other). Paths that include the A -> C -> A
			// (and vice versa) loop should be ignored, resulting in the paths:
			//   * Trust Anchor -> A -> B -> EE
			//   * Trust Anchor -> C -> B -> EE
			//   * Trust Anchor -> A -> C -> B -> EE
			//   * Trust Anchor -> C -> A -> B -> EE

	Dir          string   // directory containing package
	ImportPath   string   // canonical import path ("package path")
	GoFiles      []string // absolute paths to package source files
	NonGoFiles   []string // absolute paths to package non-Go files
	IgnoredFiles []string // absolute paths to ignored source files

	ModulePath    string            // module path (may be "" on module error)

					errs = appendErrors(errs, security.CheckEmptyValues("Ports", op.Ports))
					errs = appendErrors(errs, security.CheckEmptyValues("Methods", op.Methods))
					errs = appendErrors(errs, security.CheckEmptyValues("Paths", op.Paths))
					errs = appendErrors(errs, security.CheckEmptyValues("Hosts", op.Hosts))
					errs = appendErrors(errs, security.CheckEmptyValues("NotPorts", op.NotPorts))

//
// The special wildcard {$} matches only the end of the URL.
// For example, the pattern "/{$}" matches only the path "/",
// whereas the pattern "/" matches every path.
//
// For matching, both pattern paths and incoming request paths are unescaped segment by segment.
// So, for example, the path "/a%2Fb/100%25" is treated as having two segments, "a/b" and "100%".
// The pattern "/a%2fb/" matches it, but the pattern "/a/b/" does not.
//

		return
	}

	vars := mux.Vars(r)

	var args dsync.LockArgs
	var lockers []dsync.NetLocker
	for _, path := range strings.Split(vars["paths"], ",") {
		if path == "" {
			continue
		}
		args.Resources = append(args.Resources, path)
	}

	for _, lks := range z.serverPools[0].erasureLockers {
		lockers = append(lockers, lks...)
	}

// In particular, dict values should be linearized in sorted key order.
//
// The linearized index paths can be returned back to a structured
// representation (e.g. to emit C structs matching a signature) with a simple
// algorithm that recurses on each run of index paths with identical first
// elements.
class StructuredValueLinearizer {
 public:

					"jwks_uri": "{{.URL}}/.testing/keys"
			}`,
			fetchKeysFromRemote: true,
			want: &user.DefaultInfo{
				Name: "jane",
			},
		},
		{
			name: "discovery url and issuer url have paths",
			options: Options{
				JWTAuthenticator: apiserver.JWTAuthenticator{
					Issuer: apiserver.Issuer{
						URL:          "https://auth.example.com/a/b/foo",