config,server: add upstream.ignorePaths (#869)

Add upstream.ignorePaths config to prevent model swaps for static-asset
requests made through the /upstream/<model>/<path> passthrough endpoint.

- add UpstreamConfig with compiled *regexp.Regexp slice; invalid regex
returns an error at load time
- apply a default pattern matching common static-asset suffixes
(.js/.json/.css/.png/.gif/.jpg/.jpeg/.ico/.txt) when unset
- in handleUpstream, return 409 Conflict when a path matches and the
local model is not already loaded; peer and already-loaded models fall
through to normal dispatch
- update config-schema.json and config.example.yaml

Updates discussion: #868

internal/config/config.go

@@ -163,6 +163,9 @@ type Config struct {
 
 	// support remote peers, see issue #433, #296
 	Peers PeerDictionaryConfig `yaml:"peers"`
+
+	// upstream controls behaviour of the /upstream passthrough endpoint
+	Upstream UpstreamConfig `yaml:"upstream"`
 }
 
 // RoutingConfig is the canonical, normalized routing/scheduling configuration.
@@ -270,6 +273,12 @@ func LoadConfigFromReader(r io.Reader) (Config, error) {
 		return Config{}, fmt.Errorf("globalTTL must be >= 0")
 	}
 
+	// Apply default for upstream.ignorePaths when not specified. The default
+	// matches common static-asset suffixes so they do not trigger a swap.
+	if len(config.Upstream.IgnorePaths) == 0 {
+		config.Upstream.IgnorePaths = DefaultUpstreamIgnorePaths()
+	}
+
 	switch config.LogToStdout {
 	case LogToStdoutProxy, LogToStdoutUpstream, LogToStdoutBoth, LogToStdoutNone:
 	default:

internal/config/config_posix_test.go

@@ -266,6 +266,9 @@ groups:
 			"mthree":    "model3",
 		},
 		Groups: expectedGroups,
+		Upstream: UpstreamConfig{
+			IgnorePaths: DefaultUpstreamIgnorePaths(),
+		},
 		Routing: RoutingConfig{
 			Router: RouterConfig{
 				Use: "group",

internal/config/config_windows_test.go

@@ -255,6 +255,9 @@ groups:
 			"mthree":    "model3",
 		},
 		Groups: expectedGroups,
+		Upstream: UpstreamConfig{
+			IgnorePaths: DefaultUpstreamIgnorePaths(),
+		},
 		Routing: RoutingConfig{
 			Router: RouterConfig{
 				Use: "group",

internal/config/upstream.go

@@ -0,0 +1,55 @@
+package config
+
+import (
+	"fmt"
+	"regexp"
+
+	"gopkg.in/yaml.v3"
+)
+
+// DefaultUpstreamIgnorePathsPattern is the default regular expression applied
+// to upstream.ignorePaths when the section is empty or absent from the config.
+// It matches common static-asset suffixes so requests for .js/.css/.png/etc.
+// files do not trigger a model swap.
+const DefaultUpstreamIgnorePathsPattern = `.*\.(js|json|css|png|gif|jpg|jpeg|ico|txt)$`
+
+// DefaultUpstreamIgnorePaths returns the default compiled ignore paths used
+// when upstream.ignorePaths is not specified in the config. The returned slice
+// is fresh so callers may mutate it without affecting other configs.
+func DefaultUpstreamIgnorePaths() []*regexp.Regexp {
+	return []*regexp.Regexp{regexp.MustCompile(DefaultUpstreamIgnorePathsPattern)}
+}
+
+// UpstreamConfig controls behaviour of the /upstream passthrough endpoint.
+type UpstreamConfig struct {
+	// IgnorePaths is a slice of compiled regular expressions. Any request to
+	// /upstream/<model>/<path> whose remaining path matches any of these
+	// expressions will be ignored and not trigger a swap. When the config
+	// does not specify any patterns, DefaultUpstreamIgnorePaths is applied.
+	IgnorePaths []*regexp.Regexp `yaml:"-"`
+}
+
+// rawUpstreamConfig is the intermediate form used to unmarshal the YAML into
+// plain strings, which are then compiled into *regexp.Regexp.
+type rawUpstreamConfig struct {
+	IgnorePaths []string `yaml:"ignorePaths"`
+}
+
+// UnmarshalYAML compiles each ignorePaths entry into a *regexp.Regexp. If any
+// entry fails to compile, an error is returned.
+func (u *UpstreamConfig) UnmarshalYAML(value *yaml.Node) error {
+	var raw rawUpstreamConfig
+	if err := value.Decode(&raw); err != nil {
+		return err
+	}
+	patterns := make([]*regexp.Regexp, 0, len(raw.IgnorePaths))
+	for _, p := range raw.IgnorePaths {
+		re, err := regexp.Compile(p)
+		if err != nil {
+			return fmt.Errorf("upstream.ignorePaths: invalid regular expression %q: %w", p, err)
+		}
+		patterns = append(patterns, re)
+	}
+	u.IgnorePaths = patterns
+	return nil
+}

internal/config/upstream_test.go

@@ -0,0 +1,88 @@
+package config
+
+import (
+	"regexp"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+const upstreamConfigHeader = `
+models:
+  model1:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8080"
+`
+
+func TestConfig_UpstreamIgnorePaths_DefaultWhenAbsent(t *testing.T) {
+	// When upstream is not specified at all, the default pattern is applied.
+	content := upstreamConfigHeader
+	cfg, err := LoadConfigFromReader(strings.NewReader(content))
+	require.NoError(t, err)
+	require.Len(t, cfg.Upstream.IgnorePaths, 1)
+
+	def := cfg.Upstream.IgnorePaths[0]
+	assert.IsType(t, &regexp.Regexp{}, def)
+	assert.Equal(t, DefaultUpstreamIgnorePathsPattern, def.String())
+
+	// The default matches common static-asset suffixes.
+	assert.True(t, def.MatchString("/foo.js"))
+	assert.True(t, def.MatchString("/bar/baz.json"))
+	assert.True(t, def.MatchString("/static/img.png"))
+	assert.True(t, def.MatchString("/notes.txt"))
+	assert.True(t, def.MatchString("/favicon.ico"))
+	// And does not match inference API paths.
+	assert.False(t, def.MatchString("/v1/chat/completions"))
+	assert.False(t, def.MatchString("/v1/models"))
+	assert.False(t, def.MatchString("/health"))
+}
+
+func TestConfig_UpstreamIgnorePaths_DefaultWhenSectionEmpty(t *testing.T) {
+	// When upstream is present but ignorePaths is omitted, the default is still
+	// applied.
+	content := `upstream: {}` + "\n" + upstreamConfigHeader
+	cfg, err := LoadConfigFromReader(strings.NewReader(content))
+	require.NoError(t, err)
+	require.Len(t, cfg.Upstream.IgnorePaths, 1)
+	assert.Equal(t, DefaultUpstreamIgnorePathsPattern, cfg.Upstream.IgnorePaths[0].String())
+}
+
+func TestConfig_UpstreamIgnorePaths_Compiles(t *testing.T) {
+	content := `
+upstream:
+  ignorePaths:
+    - ".*\\.(js|json|css|png|gif|jpg|jpeg|txt)$"
+    - "^/static/.*"
+` + upstreamConfigHeader
+
+	cfg, err := LoadConfigFromReader(strings.NewReader(content))
+	require.NoError(t, err)
+	require.Len(t, cfg.Upstream.IgnorePaths, 2)
+
+	// Verify the patterns are compiled into *regexp.Regexp and match as expected.
+	assert.True(t, cfg.Upstream.IgnorePaths[0].MatchString("/foo.js"))
+	assert.True(t, cfg.Upstream.IgnorePaths[0].MatchString("/bar/baz.json"))
+	assert.False(t, cfg.Upstream.IgnorePaths[0].MatchString("/v1/chat/completions"))
+	assert.True(t, cfg.Upstream.IgnorePaths[1].MatchString("/static/foo.png"))
+	assert.False(t, cfg.Upstream.IgnorePaths[1].MatchString("/v1/chat/completions"))
+
+	// Confirm the type is *regexp.Regexp to satisfy the API contract.
+	for _, re := range cfg.Upstream.IgnorePaths {
+		assert.IsType(t, &regexp.Regexp{}, re)
+	}
+}
+
+func TestConfig_UpstreamIgnorePaths_InvalidRegexReturnsError(t *testing.T) {
+	content := `
+upstream:
+  ignorePaths:
+    - "[invalid("
+` + upstreamConfigHeader
+
+	_, err := LoadConfigFromReader(strings.NewReader(content))
+	require.Error(t, err)
+	assert.Contains(t, err.Error(), "upstream.ignorePaths")
+	assert.Contains(t, err.Error(), "invalid regular expression")
+}