generated by Claude-3.7-Sonnet

utils/srt_parser.go

@@ -0,0 +1,107 @@
+// Package utils 提供工具函数
+package utils
+
+import (
+	"fmt"
+	"regexp"
+	"strconv"
+	"strings"
+	"time"
+
+	"gitea.malaihome.work/ans/podcast_search/model"
+)
+
+// SrtParser 用于解析SRT格式的转录文件
+type SrtParser struct {
+	// SRT格式匹配正则表达式
+	srtMatcher *regexp.Regexp
+	// 换行符匹配正则表达式
+	newlineMatcher *regexp.Regexp
+}
+
+// NewSrtParser 创建新的SRT解析器
+func NewSrtParser() *SrtParser {
+	return &SrtParser{
+		srtMatcher: regexp.MustCompile(
+			`(\d*)\s?((\d{2}):(\d{2}):(\d{2})[,|.](\d{3})) +--> +((\d{2}):(\d{2}):(\d{2})[,|.](\d{3})).*[\r\n]+\s*([^\r\n]+(\n|\r|\n)(?:.*))`,
+		),
+		newlineMatcher: regexp.MustCompile(`[\r\n]+`),
+	}
+}
+
+// Parse 解析SRT格式的转录文件
+func (p *SrtParser) Parse(srtData string) (*model.Transcript, error) {
+	matches := p.srtMatcher.FindAllStringSubmatch(srtData, -1)
+	subtitles := make([]model.Subtitle, 0)
+	i := 0
+
+	for _, match := range matches {
+		i++
+
+		// 索引可能存在也可能不存在
+		index := i
+		if match[1] != "" {
+			parsedIndex, err := strconv.Atoi(match[1])
+			if err == nil {
+				index = parsedIndex
+			}
+		}
+
+		// 解析开始时间
+		startTimeHours, _ := strconv.Atoi(match[3])
+		startTimeMinutes, _ := strconv.Atoi(match[4])
+		startTimeSeconds, _ := strconv.Atoi(match[5])
+		startTimeMilliseconds, _ := strconv.Atoi(match[6])
+
+		// 解析结束时间
+		endTimeHours, _ := strconv.Atoi(match[8])
+		endTimeMinutes, _ := strconv.Atoi(match[9])
+		endTimeSeconds, _ := strconv.Atoi(match[10])
+		endTimeMilliseconds, _ := strconv.Atoi(match[11])
+
+		// 计算开始和结束时间的Duration
+		startDuration := time.Duration(startTimeHours)*time.Hour +
+			time.Duration(startTimeMinutes)*time.Minute +
+			time.Duration(startTimeSeconds)*time.Second +
+			time.Duration(startTimeMilliseconds)*time.Millisecond
+
+		endDuration := time.Duration(endTimeHours)*time.Hour +
+			time.Duration(endTimeMinutes)*time.Minute +
+			time.Duration(endTimeSeconds)*time.Second +
+			time.Duration(endTimeMilliseconds)*time.Millisecond
+
+		// 解析文本内容
+		textLines := p.newlineMatcher.Split(match[12], -1)
+		text := ""
+
+		// 处理文本行
+		for _, line := range textLines {
+			if text == "" {
+				text = line
+			} else if strings.HasSuffix(text, " ") || strings.HasPrefix(line, " ") {
+				text += line
+			} else {
+				text += " " + line
+			}
+		}
+
+		// 创建字幕
+		subtitle := model.Subtitle{
+			Index:   index,
+			Start:   startDuration,
+			End:     endDuration,
+			Data:    strings.TrimSpace(text),
+			Speaker: "", // SRT格式通常不包含发言人信息
+		}
+
+		subtitles = append(subtitles, subtitle)
+	}
+
+	if len(subtitles) == 0 {
+		return nil, fmt.Errorf("解析SRT失败: 未找到有效的字幕")
+	}
+
+	return &model.Transcript{
+		Subtitles: subtitles,
+	}, nil
+}