From bc077db0ee4f3730183b18cde74ddd7581c4cb47 Mon Sep 17 00:00:00 2001 From: Kunal Karmakar Date: Sun, 19 Apr 2026 06:26:52 +0000 Subject: [PATCH] Deduplicate ParseDataAudioURL function --- pkg/providers/common/common.go | 68 +------------------ pkg/providers/common/common_test.go | 31 +++++++++ .../responses_common.go | 22 +----- .../responses_common_test.go | 36 ---------- 4 files changed, 36 insertions(+), 121 deletions(-) diff --git a/pkg/providers/common/common.go b/pkg/providers/common/common.go index afc6877b1..5e03bc0c2 100644 --- a/pkg/providers/common/common.go +++ b/pkg/providers/common/common.go @@ -127,7 +127,7 @@ func SerializeMessages(messages []Message) []any { continue } - if format, data, ok := parseDataAudioURL(mediaURL); ok { + if format, data, ok := ParseDataAudioURL(mediaURL); ok { parts = append(parts, map[string]any{ "type": "input_audio", "input_audio": map[string]any{ @@ -205,7 +205,8 @@ func serializeToolCalls(toolCalls []ToolCall) []openaiToolCall { return out } -func parseDataAudioURL(mediaURL string) (format, data string, ok bool) { +// ParseDataAudioURL extracts the format and base64 data from a data:audio/... URL. +func ParseDataAudioURL(mediaURL string) (format, data string, ok bool) { if !strings.HasPrefix(mediaURL, "data:audio/") { return "", "", false } @@ -478,69 +479,6 @@ func AsInt(v any) (int, bool) { } } -// ExtractProtocol extracts the effective protocol and model identifier from a -// model configuration. -// -// The explicit Provider field takes precedence. When Provider is empty, the -// protocol is inferred from Model. Plain model names default to "openai". -// Provider-prefixed models strip the first slash-separated segment from the -// returned model ID. -// -// The returned protocol is normalized to the provider's canonical spelling. -// Examples: -// - Model "openai/gpt-4o" -> ("openai", "gpt-4o") -// - Model "nvidia/z-ai/glm-5.1" -> ("nvidia", "z-ai/glm-5.1") -// - Provider "nvidia", Model "z-ai/glm-5.1" -> ("nvidia", "z-ai/glm-5.1") -// - Provider "openai", Model "openai/gpt-4o" -> ("openai", "openai/gpt-4o") -// - Model "gpt-4o" -> ("openai", "gpt-4o") -func ExtractProtocol(model string) (protocol, modelID string) { - if cfg == nil { - return "", "" - } - - model := strings.TrimSpace(cfg.Model) - if provider := strings.TrimSpace(cfg.Provider); provider != "" { - return NormalizeProvider(provider), model - } - if model == "" { - return "", "" - } - - protocol, rest, found := strings.Cut(model, "/") - if !found { - return "openai", model - } - protocol = strings.TrimSpace(protocol) - if protocol == "" { - return "", strings.TrimSpace(rest) - } - return NormalizeProvider(protocol), strings.TrimSpace(rest) -} - -// NormalizeAnthropicBaseURL ensures the Anthropic base URL is properly formatted. -// It removes a trailing /v1 suffix if present (to avoid duplication), then -// re-appends /v1 when appendV1Suffix is true. An empty apiBase falls back to -// defaultBaseURL. -func NormalizeAnthropicBaseURL(apiBase, defaultBaseURL string, appendV1Suffix bool) string { - base := strings.TrimSpace(apiBase) - if base == "" { - return defaultBaseURL - } - - base = strings.TrimRight(base, "/") - if before, ok := strings.CutSuffix(base, "/v1"); ok { - base = before - } - if base == "" { - return defaultBaseURL - } - - if appendV1Suffix { - return base + "/v1" - } - return base -} - // AsFloat converts various numeric types to float64. func AsFloat(v any) (float64, bool) { switch val := v.(type) { diff --git a/pkg/providers/common/common_test.go b/pkg/providers/common/common_test.go index 71c1bd1d1..1f9a9b827 100644 --- a/pkg/providers/common/common_test.go +++ b/pkg/providers/common/common_test.go @@ -691,6 +691,37 @@ func TestExtractProtocol(t *testing.T) { } } +// --- ParseDataAudioURL tests --- + +func TestParseDataAudioURL(t *testing.T) { + tests := []struct { + name string + mediaURL string + wantFormat string + wantData string + wantOK bool + }{ + {"valid mp3", "data:audio/mp3;base64,SGVsbG8=", "mp3", "SGVsbG8=", true}, + {"valid wav", "data:audio/wav;base64,AAAA", "wav", "AAAA", true}, + {"not audio", "data:image/png;base64,abc", "", "", false}, + {"no comma", "data:audio/mp3;base64", "", "", false}, + {"empty data", "data:audio/mp3;base64,", "", "", false}, + {"empty string", "", "", "", false}, + } + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + format, data, ok := ParseDataAudioURL(tt.mediaURL) + if ok != tt.wantOK || format != tt.wantFormat || data != tt.wantData { + t.Errorf( + "ParseDataAudioURL(%q) = (%q, %q, %v), want (%q, %q, %v)", + tt.mediaURL, format, data, ok, + tt.wantFormat, tt.wantData, tt.wantOK, + ) + } + }) + } +} + // --- NormalizeAnthropicBaseURL tests --- func TestNormalizeAnthropicBaseURL(t *testing.T) { diff --git a/pkg/providers/openai_responses_common/responses_common.go b/pkg/providers/openai_responses_common/responses_common.go index 839471f69..17b731ed4 100644 --- a/pkg/providers/openai_responses_common/responses_common.go +++ b/pkg/providers/openai_responses_common/responses_common.go @@ -10,6 +10,7 @@ import ( "github.com/openai/openai-go/v3" "github.com/openai/openai-go/v3/responses" + "github.com/sipeed/picoclaw/pkg/providers/common" "github.com/sipeed/picoclaw/pkg/providers/protocoltypes" ) @@ -118,7 +119,7 @@ func BuildMultipartContent(text string, media []string) responses.ResponseInputM }, }) } else if strings.HasPrefix(mediaURL, "data:audio/") { - if format, data, ok := ParseDataAudioURL(mediaURL); ok { + if format, data, ok := common.ParseDataAudioURL(mediaURL); ok { parts = append(parts, responses.ResponseInputContentUnionParam{ OfInputFile: &responses.ResponseInputFileParam{ FileData: openai.Opt(data), @@ -132,25 +133,6 @@ func BuildMultipartContent(text string, media []string) responses.ResponseInputM return parts } -// ParseDataAudioURL extracts the format and base64 data from a data:audio/... URL. -func ParseDataAudioURL(mediaURL string) (format, data string, ok bool) { - if !strings.HasPrefix(mediaURL, "data:audio/") { - return "", "", false - } - payload := strings.TrimPrefix(mediaURL, "data:audio/") - meta, data, found := strings.Cut(payload, ",") - if !found { - return "", "", false - } - format, _, _ = strings.Cut(meta, ";") - format = strings.TrimSpace(format) - data = strings.TrimSpace(data) - if format == "" || data == "" { - return "", "", false - } - return format, data, true -} - // ResolveToolCall extracts the function name and JSON arguments string from a ToolCall. // Returns ok=false if the tool call has no name or if arguments fail to marshal. func ResolveToolCall(tc protocoltypes.ToolCall) (name string, arguments string, ok bool) { diff --git a/pkg/providers/openai_responses_common/responses_common_test.go b/pkg/providers/openai_responses_common/responses_common_test.go index 0d41190b1..ace91edf0 100644 --- a/pkg/providers/openai_responses_common/responses_common_test.go +++ b/pkg/providers/openai_responses_common/responses_common_test.go @@ -506,42 +506,6 @@ func TestParseResponseBody_CanceledStatus(t *testing.T) { } } -// --- ParseDataAudioURL tests --- - -func TestParseDataAudioURL_Valid(t *testing.T) { - format, data, ok := ParseDataAudioURL("data:audio/mp3;base64,SGVsbG8=") - if !ok { - t.Fatal("expected ok=true") - } - if format != "mp3" { - t.Errorf("format = %q, want %q", format, "mp3") - } - if data != "SGVsbG8=" { - t.Errorf("data = %q, want %q", data, "SGVsbG8=") - } -} - -func TestParseDataAudioURL_NotAudio(t *testing.T) { - _, _, ok := ParseDataAudioURL("data:image/png;base64,abc") - if ok { - t.Error("expected ok=false for non-audio URL") - } -} - -func TestParseDataAudioURL_MalformedNoComma(t *testing.T) { - _, _, ok := ParseDataAudioURL("data:audio/mp3;base64") - if ok { - t.Error("expected ok=false for malformed URL") - } -} - -func TestParseDataAudioURL_EmptyData(t *testing.T) { - _, _, ok := ParseDataAudioURL("data:audio/mp3;base64,") - if ok { - t.Error("expected ok=false for empty data") - } -} - // --- BuildMultipartContent tests --- func TestBuildMultipartContent_TextOnly(t *testing.T) {