Notepad ++中的正则表达式可替换CDATA中的<和>

Question

I'm using Notepad++ to fix a huge XML export file and one of the challenges here is to replace all < and > characters to < and > . The thing is, I can't simply use the replace all action since the XML file is full of < and > that cannot be changed.

Luckly all the < and > that I need to change are wrapped by CDATA tags, like this:

<![CDATA[Text here... <span class="vSpecial"><p>Special Offer - more text here!</p></span>]]>

I was wondering if there'd be a Regular Expression to identify < and > wrapped in CDATA content, so I could easily use the Replace All to change only them.

UPDATE

The content of CDATA can contain line breaks.

Answer 1

Code

See regex in use here

<!\[CDATA\[)(?:(?!\]\]>).)*?\K(?:(<)|(>))

Replacement: (?{1}<)(?{2}>)

Note : For display purposes the link above uses \\G(?!\\A) . This is not supported in Notepad++, thus it's been dropped in the actual answer. I added it to the link to show what it basically does.

See the Notepadd++ documentation for more information. It mentions the following:

For those readers familiar with Perl, \\G is not supported.

---

Results

Before

After

---

Explanation

Click Replace All repeatedly until the message at the bottom shows Replace All: 0 occurrences were replaced. It will replace the first occurrence, then the second occurrence, then third, etc. for each CDATA that is found until there are no more matches.

Pattern

• <!\\[CDATA\\[ Matches <![[CDATA[ literally
• (?:(?!\\]\\]>).)*? Tempered lazy token matching any character any number of times, but as few as possible ensuring what follows doesn't match ]]>
• \\K Resets the starting point of the reported match. Any previously consumed characters are no longer included in the final match
• (?:(<)|(>)) Match either of the following
  • (<) Capture < literally into capture group 1
  • (>) Capture > literally into capture group 2

Replacement

Notepad++ allows conditional replacements, so (?{1}<) makes reference to capture group one and (?{2}>) makes reference to capture group 2.