一口气查找并替换HTML标记内的所有字符串

Question

I need a regular expression to search for and replace multiple occurrences of a text string within a delimited section of text.

Let's say there is HTML code with one or more spans that have a certain class. Each span may have none, one or multiple occurrences of the string {abc} inside, eg

<p>lorem ipsum dolor <span class="xyz">sid amet{abc}et pluribus {abc} unum{abc} diex 
et mon droit</span> you'll never walk alone</p>

Thus I need a regex pair to replace all occurrences of {abc} within <span id="xyz"> with {def} in a single pass .

This is for use in a text editor such as Notepad++ and the like and needs to be be a PCRE/UNIX-style regular expression.

What I have is,

find: (<span class="xyz">)([^<]*)\\{abc\\}([^<]*<)

replace: \\1\\2{def}\\3

This does work for one occurrence within a span, but in case of more occurrences, I have to run replacement multiple times, in cycle, while I need that to be one-pass .

I wonder how can I achieve that. I suppose this is a pretty common case, somehow I could not find similar things concerning the need to be one-pass, no cycles, no code, and I'd like to get an idea how this could be done in principle.

Answer 1

This seems to work in Notepad++

Find what : (?:<span class="xyz">|\\G)[^<]*?\\K\\{abc\\}(?=[^<]*<\\/span>)

Replace with : {def}

Search mode : Regular expression

Note that because of the [^<]* there is an assumption that there are no other tags within the span tag.