Java正则表达式：精准查找双引号内所有指定单词的出现次数

String text = "variant "if and only if 5 divides by i without remainder, then it prints \"i + \" variant: \" + variant\"" variant";
String str_regex = "\".*(variant).*\""; // 错误的正则，会只匹配最后一个"variant"
Matcher str_matcher = Pattern.compile(str_regex).matcher(text);
while(str_matcher.find()) {
    System.out.println(str_matcher.group()); // 可能会输出整个匹配到的字符串，但内部捕获组只对应最后一个
}

import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class RegexWordInQuotes {

    public static void main(String[] args) {
        String text = "variant "if and only if 5 divides by i without remainder, then it prints \"i + \" variant: \" + variant\"" variant "another variant here" and "yet another variant"";
        String targetWord = "variant";

        // 阶段一：匹配所有双引号字符串的正则表达式
        // (?s) - DOTALL模式，让.匹配所有字符包括换行符
        // (?<!\\) - 负向后瞻，确保开引号前没有单反斜杠
        // (?:\\{2})* - 匹配零个或多个双反斜杠，处理 \" 这样的情况
        // "[^"\\]*(?:\\.[^"\\]*)*" - 匹配引号内容，处理转义字符
        Pattern quotedStringPattern = Pattern.compile("(?s)(?<!\\)(?:\\{2})*"[^"\\]*(?:\\.[^"\\]*)*"");
        Matcher quotedStringMatcher = quotedStringPattern.matcher(text);

        int totalCount = 0;

        // 遍历所有匹配到的双引号字符串
        while (quotedStringMatcher.find()) {
            String quotedContent = quotedStringMatcher.group();
            // System.out.println("Found quoted string: " + quotedContent); // 调试用

            // 阶段二：在每个双引号字符串内部统计目标单词
            // 使用split方法计算出现次数
            // split("word", -1).length - 1 是一个常见的计数技巧
            int occurrencesInThisQuote = quotedContent.split(Pattern.quote(targetWord), -1).length - 1;
            totalCount += occurrencesInThisQuote;
        }

        System.out.println("The target word "" + targetWord + "" appears " + totalCount + " times inside double quotes.");
        // 预期输出: The target word "variant" appears 4 times inside double quotes.
        // 解释：
        // 第一个双引号字符串: "if and only if ... variant: " + variant"" 包含 2 个 "variant"
        // 第二个双引号字符串: "another variant here" 包含 1 个 "variant"
        // 第三个双引号字符串: "yet another variant" 包含 1 个 "variant"
        // 总计 2 + 1 + 1 = 4
    }
}