Files
Calcium-Ion 128089fa7d feat(i18n): translate web console messages in the frontend and AI-client errors in the backend (#7712)
* feat(i18n): translate web console messages in the frontend and AI-client errors in the backend

Chinese text was hard-coded in API messages, stored log content, emails and
server logs, so it reached users of every interface language.

Text is now translated where its reader is:

- Web console APIs return the English source text as message_key with
  message_params, and the web console translates it in all seven languages.
  message keeps the English text for other API callers.
- Errors returned to AI clients, emails and notifications are translated by
  the backend (en, zh-CN, zh-TW) in the language saved on the account, then
  the language of the request's Accept-Language header.
- A reader who states no language still gets the original Chinese text of the
  messages that were Chinese before, word for word. Gateways that use another
  new-api as their upstream match that text, for example to disable a channel
  whose account ran out of quota.
- Usage, top-up and system log content is stored as content_parts (key and
  params) and rendered in the viewer's language. Task fail reasons and refund
  reasons written by the gateway are English keys. Rows written earlier are
  shown as before.
- Server logs are English.

Also:

- The saved interface language zhTW, and zh-HK, zh-MO and zh-Hant, now select
  Traditional Chinese in the backend instead of Simplified Chinese.
- The built-in group descriptions are translated; descriptions an
  administrator saved are shown as written.
- .agents/rules/i18n.md records the conventions. Tests in i18n/ fail on a
  non-English string literal, on a web console message without translations
  and on a locale key without a constant.

* feat(i18n): print server logs in Chinese when LOG_LANGUAGE asks for it

Server log lines are English by default. With LOG_LANGUAGE=zh-CN, the lines
that were Chinese before print their original Chinese text.

common.LogText formats a log line from its English format, which is also the
key of its Chinese text in common/log_text.zh-CN.json. A line without an entry
prints in English. go vet checks the arguments of every call, and
TestServerLogTexts checks that each entry still matches a call and has the
same verbs as its key.

* docs: list LOG_LANGUAGE in the docker compose example

* feat(i18n): translate every server log line and add DEFAULT_LANGUAGE

Every line the gateway prints goes through common.LogText. With
DEFAULT_LANGUAGE=zh-CN, the lines that have a Chinese text in
common/log_text.zh-CN.json print it and the others print English.

DEFAULT_LANGUAGE replaces LOG_LANGUAGE. It also sets the language of
backend messages, emails and notifications for a reader who states no
language: en, zh-CN or zh-TW, English when unset. The list of messages
that stayed Chinese for such readers is removed.

Review fixes:
- Channel test logs store the token name "Model test"; the console shows
  it in the interface language.
- Task fail reasons written by the gateway are fixed sentences the console
  translates. The HTTP status and the upstream detail go to the server log
  when a polled task is failed.
- The auto-disable reason of a slow channel is written in DEFAULT_LANGUAGE.

Also:
- Drop the repository-scanning lint tests, the i18n rules file and its
  read gate. AGENTS.md keeps the conventions in four lines.
- Payment error toasts keep their previous text and only add the
  translated message key.
- Remove guards and helpers that no caller needs.

* fix(i18n): address review on language lookup, notices and error responses

- Relay requests no longer load the user a second time to find the
  language: the settings TokenAuth put into the request are used as they
  are, also when they hold no language.
- Upstream model update notices are rendered in each recipient's saved
  language, DEFAULT_LANGUAGE when none.
- /api/usage/token and /api/log/token answer with backend-translated
  messages again, with the keys and status codes used before this PR.
  Their callers are scripts, not the web console.
- Webhook notifications fill {{value}} placeholders like the other
  senders. Running the rendered text through Sprintf appended
  %!(EXTRA ...) and broke %2F in links.
- RootUserLanguage reads the setting of the root user it already loaded.
- zh_TW and zh_HK are read as Traditional Chinese.
- Add common.ApiErrorStatus for error responses under another HTTP status
  and use it where those responses were built by hand.
- Remove middleware.GetLanguage, which had no callers.
2026-10-09 21:39:51 +08:00

350 lines
9.7 KiB
Go

package common
import (
"context"
"encoding/binary"
"fmt"
"io"
"strings"
"github.com/abema/go-mp4"
"github.com/go-audio/aiff"
"github.com/go-audio/wav"
"github.com/jfreymuth/oggvorbis"
"github.com/mewkiz/flac"
"github.com/pkg/errors"
"github.com/tcolgate/mp3"
"github.com/yapingcat/gomedia/go-codec"
)
// GetAudioDuration 使用纯 Go 库获取音频文件的时长(秒)。
// 它不再依赖外部的 ffmpeg 或 ffprobe 程序。
func GetAudioDuration(ctx context.Context, f io.ReadSeeker, ext string) (duration float64, err error) {
SysLog(LogText("GetAudioDuration: ext=%s", ext))
ext = strings.ToLower(ext)
// 根据文件扩展名选择解析器
switch ext {
case ".mp3":
duration, err = getMP3Duration(f)
case ".wav":
duration, err = getWAVDuration(f)
case ".flac":
duration, err = getFLACDuration(f)
case ".m4a", ".mp4":
duration, err = getM4ADuration(f)
case ".ogg", ".oga", ".opus":
duration, err = getOGGDuration(f)
if err != nil {
duration, err = getOpusDuration(f)
}
case ".aiff", ".aif", ".aifc":
duration, err = getAIFFDuration(f)
case ".webm":
duration, err = getWebMDuration(f)
case ".aac":
duration, err = getAACDuration(f)
default:
return 0, fmt.Errorf("unsupported audio format: %s", ext)
}
SysLog(LogText("GetAudioDuration: duration=%f", duration))
return duration, err
}
// getMP3Duration 解析 MP3 文件以获取时长。
// 注意:对于 VBR (Variable Bitrate) MP3,这个估算可能不完全精确,但通常足够好。
// FFmpeg 在这种情况下会扫描整个文件来获得精确值,但这里的库提供了快速估算。
func getMP3Duration(r io.Reader) (float64, error) {
d := mp3.NewDecoder(r)
var f mp3.Frame
skipped := 0
duration := 0.0
for {
if err := d.Decode(&f, &skipped); err != nil {
if err == io.EOF {
break
}
return 0, errors.Wrap(err, "failed to decode mp3 frame")
}
duration += f.Duration().Seconds()
}
return duration, nil
}
// getWAVDuration 解析 WAV 文件头以获取时长。
func getWAVDuration(r io.ReadSeeker) (float64, error) {
// 1. 强制复位指针
r.Seek(0, io.SeekStart)
dec := wav.NewDecoder(r)
// IsValidFile 会读取 fmt 块
if !dec.IsValidFile() {
return 0, errors.New("invalid wav file")
}
// 尝试寻找 data 块
if err := dec.FwdToPCM(); err != nil {
return 0, errors.Wrap(err, "failed to find PCM data chunk")
}
pcmSize := int64(dec.PCMSize)
// 如果读出来的 Size 是 0,尝试用文件大小反推
if pcmSize == 0 {
// 获取文件总大小
currentPos, _ := r.Seek(0, io.SeekCurrent) // 当前通常在 data chunk header 之后
endPos, _ := r.Seek(0, io.SeekEnd)
fileSize := endPos
// 恢复位置(虽然如果不继续读也没关系)
r.Seek(currentPos, io.SeekStart)
// 数据区大小 ≈ 文件总大小 - 当前指针位置(即Header大小)
// 注意:FwdToPCM 成功后,CurrentPos 应该刚好指向 Data 区数据的开始
// 或者是 Data Chunk ID + Size 之后。
// WAV Header 一般 44 字节。
if fileSize > 44 {
// 如果 FwdToPCM 成功,Reader 应该位于 data 块的数据起始处
// 所以剩余的所有字节理论上都是音频数据
pcmSize = fileSize - currentPos
// 简单的兜底:如果算出来还是负数或0,强制按文件大小-44计算
if pcmSize <= 0 {
pcmSize = fileSize - 44
}
}
}
numChans := int64(dec.NumChans)
bitDepth := int64(dec.BitDepth)
sampleRate := float64(dec.SampleRate)
if sampleRate == 0 || numChans == 0 || bitDepth == 0 {
return 0, errors.New("invalid wav header metadata")
}
bytesPerFrame := numChans * (bitDepth / 8)
if bytesPerFrame == 0 {
return 0, errors.New("invalid byte depth calculation")
}
totalFrames := pcmSize / bytesPerFrame
durationSeconds := float64(totalFrames) / sampleRate
return durationSeconds, nil
}
// getFLACDuration 解析 FLAC 文件的 STREAMINFO 块。
func getFLACDuration(r io.Reader) (float64, error) {
stream, err := flac.Parse(r)
if err != nil {
return 0, errors.Wrap(err, "failed to parse flac stream")
}
defer stream.Close()
// 时长 = 总采样数 / 采样率
duration := float64(stream.Info.NSamples) / float64(stream.Info.SampleRate)
return duration, nil
}
// getM4ADuration 解析 M4A/MP4 文件的 'mvhd' box。
func getM4ADuration(r io.ReadSeeker) (float64, error) {
// go-mp4 库需要 ReadSeeker 接口
info, err := mp4.Probe(r)
if err != nil {
return 0, errors.Wrap(err, "failed to probe m4a/mp4 file")
}
// 时长 = Duration / Timescale
return float64(info.Duration) / float64(info.Timescale), nil
}
// getOGGDuration 解析 OGG/Vorbis 文件以获取时长。
func getOGGDuration(r io.ReadSeeker) (float64, error) {
// 重置 reader 到开头
if _, err := r.Seek(0, io.SeekStart); err != nil {
return 0, errors.Wrap(err, "failed to seek ogg file")
}
reader, err := oggvorbis.NewReader(r)
if err != nil {
return 0, errors.Wrap(err, "failed to create ogg vorbis reader")
}
// 计算时长 = 总采样数 / 采样率
// 需要读取整个文件来获取总采样数
channels := reader.Channels()
sampleRate := reader.SampleRate()
// 估算方法:读取到文件结尾
var totalSamples int64
buf := make([]float32, 4096*channels)
for {
n, err := reader.Read(buf)
if err == io.EOF {
break
}
if err != nil {
return 0, errors.Wrap(err, "failed to read ogg samples")
}
totalSamples += int64(n / channels)
}
duration := float64(totalSamples) / float64(sampleRate)
return duration, nil
}
// getOpusDuration 解析 Opus 文件(在 OGG 容器中)以获取时长。
func getOpusDuration(r io.ReadSeeker) (float64, error) {
// Opus 通常封装在 OGG 容器中
// 我们需要解析 OGG 页面来获取时长信息
if _, err := r.Seek(0, io.SeekStart); err != nil {
return 0, errors.Wrap(err, "failed to seek opus file")
}
// 读取 OGG 页面头部
var totalGranulePos int64
buf := make([]byte, 27) // OGG 页面头部最小大小
for {
n, err := r.Read(buf)
if err == io.EOF {
break
}
if err != nil {
return 0, errors.Wrap(err, "failed to read opus/ogg page")
}
if n < 27 {
break
}
// 检查 OGG 页面标识 "OggS"
if string(buf[0:4]) != "OggS" {
// 跳过一些字节继续寻找
if _, err := r.Seek(-26, io.SeekCurrent); err != nil {
break
}
continue
}
// 读取 granule position (字节 6-13, 小端序)
granulePos := int64(binary.LittleEndian.Uint64(buf[6:14]))
if granulePos > totalGranulePos {
totalGranulePos = granulePos
}
// 读取段表大小
numSegments := int(buf[26])
segmentTable := make([]byte, numSegments)
if _, err := io.ReadFull(r, segmentTable); err != nil {
break
}
// 计算页面数据大小并跳过
var pageSize int
for _, segSize := range segmentTable {
pageSize += int(segSize)
}
if _, err := r.Seek(int64(pageSize), io.SeekCurrent); err != nil {
break
}
}
// Opus 的采样率固定为 48000 Hz
duration := float64(totalGranulePos) / 48000.0
return duration, nil
}
// getAIFFDuration 解析 AIFF 文件头以获取时长。
func getAIFFDuration(r io.ReadSeeker) (float64, error) {
if _, err := r.Seek(0, io.SeekStart); err != nil {
return 0, errors.Wrap(err, "failed to seek aiff file")
}
dec := aiff.NewDecoder(r)
if !dec.IsValidFile() {
return 0, errors.New("invalid aiff file")
}
d, err := dec.Duration()
if err != nil {
return 0, errors.Wrap(err, "failed to get aiff duration")
}
return d.Seconds(), nil
}
// getWebMDuration 解析 WebM 文件以获取时长。
// WebM 使用 Matroska 容器格式
func getWebMDuration(r io.ReadSeeker) (float64, error) {
if _, err := r.Seek(0, io.SeekStart); err != nil {
return 0, errors.Wrap(err, "failed to seek webm file")
}
// WebM/Matroska 文件的解析比较复杂
// 这里提供一个简化的实现,读取 EBML 头部
// 对于完整的 WebM 解析,可能需要使用专门的库
// 简单实现:查找 Duration 元素
// WebM Duration 的 Element ID 是 0x4489
// 这是一个简化版本,可能不适用于所有 WebM 文件
buf := make([]byte, 8192)
n, err := r.Read(buf)
if err != nil && err != io.EOF {
return 0, errors.Wrap(err, "failed to read webm file")
}
// 尝试查找 Duration 元素(这是一个简化的方法)
// 实际的 WebM 解析需要完整的 EBML 解析器
// 这里返回错误,建议使用专门的库
if n > 0 {
// 检查 EBML 标识
if len(buf) >= 4 && binary.BigEndian.Uint32(buf[0:4]) == 0x1A45DFA3 {
// 这是一个有效的 EBML 文件
// 但完整解析需要更复杂的逻辑
return 0, errors.New("webm duration parsing requires full EBML parser (consider using ffprobe for webm files)")
}
}
return 0, errors.New("failed to parse webm file")
}
// getAACDuration 解析 AAC (ADTS格式) 文件以获取时长。
// 使用 gomedia 库来解析 AAC ADTS 帧
func getAACDuration(r io.ReadSeeker) (float64, error) {
if _, err := r.Seek(0, io.SeekStart); err != nil {
return 0, errors.Wrap(err, "failed to seek aac file")
}
// 读取整个文件内容
data, err := io.ReadAll(r)
if err != nil {
return 0, errors.Wrap(err, "failed to read aac file")
}
var totalFrames int64
var sampleRate int
// 使用 gomedia 的 SplitAACFrame 函数来分割 AAC 帧
codec.SplitAACFrame(data, func(aac []byte) {
// 解析 ADTS 头部以获取采样率信息
if len(aac) >= 7 {
// 使用 ConvertADTSToASC 来获取音频配置信息
asc, err := codec.ConvertADTSToASC(aac)
if err == nil && sampleRate == 0 {
sampleRate = codec.AACSampleIdxToSample(int(asc.Sample_freq_index))
}
totalFrames++
}
})
if sampleRate == 0 || totalFrames == 0 {
return 0, errors.New("no valid aac frames found")
}
// 每个 AAC ADTS 帧包含 1024 个采样
totalSamples := totalFrames * 1024
duration := float64(totalSamples) / float64(sampleRate)
return duration, nil
}