如何在excel或python中提取单词周围的文本？

Question

I have a thousands lines text that goes like:

ksjd 234first special 34-37xy kjsbn
sde 89second special 22-23xh ewio
647red special 55fg dsk
uuire another special 98
another special 107r
green special 55-59 ewk
blue special 31-39jkl

I need to extract a word before "special" and number (or number range) from the right. In other words, I want:

converted into a table:

Answer 1

A fast way to do this is to use regular expressions:

In [1]: import re

In [2]: text = '''234first special 34-37xy                          
   ...: 89second special 22-23xh
   ...: 647red special 55fg
   ...: another special 98
   ...: another special 107r
   ...: green special 55-59
   ...: blue special 31-39jkl'''

In [3]: [re.findall('\d*\s*(\S+)\s+(special)\s+(\d+(?:-\d+)?)', line)[0] for line in text.splitlines()]
Out[3]: 
[('first', 'special', '34-37'),
 ('second', 'special', '22-23'),
 ('red', 'special', '55'),
 ('another', 'special', '98'),
 ('another', 'special', '107'),
 ('green', 'special', '55-59'),
 ('blue', 'special', '31-39')]

Answer 2

In Excel, you can using a formula to extract text between two words by doing as follow:

1. Select a blank cell and type this formula =MID(A1,SEARCH("KTE",A1)+3,SEARCH("feature",A1)-SEARCH("KTE",A1)-4) into it , then press Enter button.

2. Drag the fill handle to fill the range you want to apply this formula. Now the text strings between "KTE" and "feature" are extracted only.

Notes:

1. In this formula, A1 is the cell you want to extract text from.

2. KTE and feature are the words you want to extract text between.

3. The number 3 is the characters length of KTE, and number 4 is equal to the characters length of KTE plus one.

Answer 3

In addition what @RolandSmith wrote, here is a way of using Regular Expressions in Excel - VBA

---

Option Explicit
Function ExtractSpecial(S As String, Index As Long) As String
    Dim RE As Object, MC As Object
    Const sPat As String = "([a-z]+)\s+(special)\s+([^a-z]+)"

Set RE = CreateObject("vbscript.regexp")
With RE
    .Global = True
    .ignorecase = True
    .MultiLine = False
    .Pattern = sPat
    If .test(S) = True Then
        Set MC = .Execute(S)
        ExtractSpecial = MC(0).submatches(Index - 1)
    End If
End With

End Function

---

The Index argument in this UDF corresponds to returning either the 1st, 2nd or 3rd submatch from the match collection, so you can easily split the original string into your three desired components.

Since you write you have "thousands of lines", you may prefer to run a macro. The macro will process the data much more quickly, but is not dynamic. The macro below assumes your original data is in Column A on Sheet2, and will put the results in columns C:E on the same worksheet. You can easily change these parameters:

---

Sub ExtractSpec()
    Dim RE As Object, MC As Object
    Dim wsSrc As Worksheet, wsRes As Worksheet, rRes As Range
    Dim vSrc As Variant, vRes As Variant
    Dim I As Long

Set wsSrc = Worksheets("sheet2")
Set wsRes = Worksheets("sheet2")
    Set rRes = wsRes.Cells(1, 3)

With wsSrc
    vSrc = .Range(.Cells(1, 1), .Cells(.Rows.Count, 1).End(xlUp))
End With

Set RE = CreateObject("vbscript.regexp")
With RE
    .Global = True
    .MultiLine = False
    .ignorecase = True
    .Pattern = "([a-z]+)\s+(special)\s+([^a-z]+)"

ReDim vRes(1 To UBound(vSrc), 1 To 3)
For I = 1 To UBound(vSrc)
    If .test(vSrc(I, 1)) = True Then
        Set MC = .Execute(vSrc(I, 1))
        vRes(I, 1) = MC(0).submatches(0)
        vRes(I, 2) = MC(0).submatches(1)
        vRes(I, 3) = MC(0).submatches(2)
    End If
Next I
End With

Set rRes = rRes.Resize(UBound(vRes, 1), UBound(vRes, 2))
With rRes
    .EntireColumn.Clear
    .Value = vRes
    .EntireColumn.AutoFit
End With

End Sub

---