正则表达式查找缺少某些部分的字符串的子字符串

Question

I have a long string (html of website) and I want to scrape the substrings.

For example, some of the output contains something like this:

<span title="Minecraft: Pocket Edition" class="oneline-info title-info">
  <a href="/apps/ios/app/minecraft-pocket-edition/">Minecraft: Pocket Edition</a>
</span>


    <span title="Mojang" class="oneline-info add-info" data-items="1">
        <a href="/apps/ios/publisher/mojang/">Mojang</a>
    </span>

I want to scrape everything from <span title= to </span> (In the above example, that means it will be 2 different matches)

So, I have this code:

        var matches = Regex.Matches(s, @"<span title=(?<content>(?:(?!""</span>).)+)");
        scrapeTitles.AddRange(matches.Cast<Match>().Select(x => x.Groups["content"].Value).ToList());

But for some reason, it's not scraping all the data between those 2 words. It only gives me outputs like this:

"Minecraft: Pocket Edition" class="oneline-info title-info">
"Mojang" class="oneline-info add-info" data-items="1">
"Clash of Clans" class="oneline-info title-info">
"Supercell" class="oneline-info add-info" data-items="1">

I need to scrape all the data, including the <a> line as well.

"Mojang" class="oneline-info add-info" data-items="1">
            <a href="/apps/ios/publisher/mojang/">Mojang</a>

Answer 1

The problem is your match doesn't take proper care of the new line character.

Try this one:

<span title=(?<content>(?:(.|\n)(?!</span>))+)

See live version .

Disclaimer: I stronly recommend NOT to do HTML (SGML actually) parsing using regular expressions. It leads to broken behavior in the long run.

Answer 2

您没有捕获换行符，因此可以更新正则表达式以解析它们，或执行以下操作：

var matches = Regex.Matches(s.Replace(Environment.NewLine, string.Empty), @"<span title=(?<content>(?:(?!""</span>).)+)");