在Notepad ++中用正则表达式查找和替换标题标签

Question

There's a OCR scanned book and there's a tool which converts the OCR'd PDF to XML but most of the XML tags are wrong so there's another tool to fix it. But I need to break the lines from <h1> to <h5> , 1. & 1.1. & 1.1.1. so its easy to re-tag using the tool.

The XML code looks like this:

`<h1>text</h1><h2>text</h3><h3>text</h3>"

and

1.text.2.text.3.text.1.1.text.1.1.1.text 

And I need to break the lines like this using a Regex in notepad++.

<h1>text</h1>
<h2>text</h2>
<h3>text</h3>

and

1.text.
2.text.
3.text.

and

1.1.text.
1.1.1.text.

I used </h1>\\s* to find an </h1>\\n but it only breaks h1 tags. I need to break all "H" tags and 1., 2., 1.1., 1.1.1. tags too.

Answer 1

At the risk of getting downvoted, i think you may be better served by a parser. In the past when I've had to manage similar tasks, I would write a small script/program to parse the file and re-write it as needed. Parsing the xml first, and then reformatting using regex might be easier to accomplish your goal.

Answer 2

You can use this search and replace (if your h1, h2, ... tags don't contain other tags) :

search:  (?<!^)(<h[1-6][^<]*|(?<![0-9]\.)[0-9]+\.)
replace: \n$1

note: if you need Windows newlines, you must change \\n with \\r\\n .

pattern details:

(?<!^)   # not preceded by the begining of the string

(                         # open the capture group 1
    <h[1-6][^<]*          # <h, a digit between 1 to 6, all characters until 
                          # the next < (to skip all the content between
                          # h1, h2... tags) 
  |                     # OR
    (?<![0-9]\.)[0-9]+\.  # one or more digits and a dot not preceded by a digit
                          # and a dot 
)                         # close the capture group 1

$1 is a reference to the content of the capture group 1