有没有办法在VBA中使用带有正则表达式的命名组?
我想写一个与文件名中的日期匹配的Excel VBA Sub,并按指定的数量递减这些日期。我需要能够区分dd / mm和mm / dd格式 - 以及其他不规则性 - 并且使用这样的命名组可以解决问题:
(?:<month>\d\d)(?:<day>\d\d)
建议表示赞赏
答案 0 :(得分:3)
不,VBScript regular expressions中没有命名组。
VBScript使用与JScript相同的正则表达式引擎,因此它与JavaScript正则表达式兼容,后者也没有命名组。
你必须使用未命名的组,并按照它们在表达式上出现的顺序来运行它后通过索引检索它们。
通常,dd / mm和mm / dd无法自动区分,因为有效日期可能是。 (例如01/04可能是1月4日或4月1日)。我不认为你能用正则表达式解决这个问题。
答案 1 :(得分:1)
这里是我今天制作的使用VBA的命名组的实现。希望这对其他人有用!:
'Description:
' An implementation of Regex which includes Named Groups
' and caching implemented in VBA
'Example:
' Dim match as Object
' set match = RegexMatch("01/01/2019","(?<month>\d\d)\/(?<day>\d\d)\/(?<year>\d\d\d\d)")
' debug.print match("day") & "/" & match("month") & "/" & match("year")
'Options:
' "i" = IgnoreCase
'Return value:
' A dictionary object with the following keys:
' 0 = Whole match
' 1,2,3,... = Submatch 1,2,3,...
' "Count" stores the count of matches
' "<<NAME>>" stores the match of a specified name
Function RegexMatch(ByVal haystack As String, ByVal pattern As String, Optional ByVal options As String) As Object
'Cache regexes for optimisation
Static CachedRegex As Object
Static CachedNames As Object
If CachedRegex Is Nothing Then Set CachedRegex = CreateObject("Scripting.Dictionary")
If CachedNames Is Nothing Then Set CachedNames = CreateObject("Scripting.Dictionary")
'Named regexp used to detect capturing groups and named capturing groups
Static NamedRegexp As Object
If NamedRegexp Is Nothing Then
Set NamedRegexp = CreateObject("VBScript.RegExp")
NamedRegexp.pattern = "\((?:\?\<(.*?)\>)?"
NamedRegexp.Global = True
End If
'If cached pattern doesn't exist, create it
If Not CachedRegex(pattern) Then
'Create names/capture group object
Dim testPattern As String, oNames As Object
testPattern = pattern
testPattern = Replace(testPattern, "\\", "asdasd")
testPattern = Replace(testPattern, "\(", "asdasd")
'Store names for optimisation
Set CachedNames(options & ")" & pattern) = NamedRegexp.Execute(testPattern)
'Create new VBA valid pattern
Dim newPattern As String
newPattern = NamedRegexp.Replace(pattern, "(")
'Create regexp from new pattern
Dim oRegexp As Object
Set oRegexp = CreateObject("VBScript.RegExp")
oRegexp.pattern = newPattern
'Set regex options
Dim i As Integer
For i = 1 To Len(flags)
Select Case Mid(flags, i, 1)
Case "i"
oRegexp.ignoreCase = True
Case "g"
oRegexp.Global = True
End Select
Next
'Store regex for optimisation
Set CachedRegex(options & ")" & pattern) = oRegexp
End If
'Get matches object
Dim oMatches As Object
Set oMatches = CachedRegex(options & ")" & pattern).Execute(haystack)
'Get names object
Dim CName As Object
Set CName = CachedNames(options & ")" & pattern)
'Create dictionary to return
Dim oRet As Object
Set oRet = CreateObject("Scripting.Dictionary")
'Fill dictionary with names and indexes
'0 = Whole match
'1,2,3,... = Submatch 1,2,3,...
'"Count" stores the count of matches
'"<<NAME>>" stores the match of a specified name
For i = 1 To CName.Count
oRet(i) = oMatches(0).Submatches(i - 1)
If Not IsEmpty(CName(i - 1).Submatches(0)) Then oRet(CName(i - 1).Submatches(0)) = oMatches(0).Submatches(i - 1)
Next i
oRet(0) = oMatches(0)
oRet("Count") = CName.Count
Set RegexMatch = oRet
End Function
答案 2 :(得分:0)
感谢@Sancarn的代码!
出于某些原因,我已对其进行了修订。我所做的更改记录在代码中:
' Procedure for testing 'RegexMatch'.
' - It shows how to convert a date from 'mm/dd/yyyy' to 'dd.mm.yyyy' format.
' - It shows how to retrieve named groups by real name: 'Match.Item("group name")'
' as well as by number: 'Match.Items(group number)'.
' - It shows how to retrieve unnamed groups by number-generated name as well as by number.
' - It shows how to retrieve group count and the whole match by number-generated name as well as by number.
' - It shows that non-capturing groups like '(?:y)?' won't be listed.
' - It shows that left parenthesis inside a character class like '([x(])?' won't disturbe.
' Take notice of:
' - the small difference between 'Item' and 'Items'
' - the quotes in 'Match.Item("number of an unnamed group")'
Sub TestRegexMatch()
Dim Match As Scripting.Dictionary
Set Match = RegexMatch("01/23/2019z", "(?<month>\d\d)\/([x(])?(?<day>\d\d)\/(?:y)?(?<year>\d\d\d\d)(z)?")
Debug.Print Match.Item("day") & "." & Match.Item("month") & "." & Match.Item("year") & " vs. " & Match.Items(2) & "." & Match.Items(0) & "." & Match.Items(3)
Debug.Print "'" & Match.Item("1") & "'" & ", '" & Match.Item("4") & "' vs. '" & Match.Items(1) & "', '" & Match.Items(4) & "'"
Debug.Print Match.Item("98") & " vs. " & Match.Items(Match.Count - 2)
Debug.Print Match.Item("99") & " vs. " & Match.Items(Match.Count - 1)
End Sub
' An implementation of regex which includes named groups and caching implemented in VBA.
' The 'Microsoft VBScript Regular Expressions 5.5' library must be referenced (in VBA-editor: Tools -> References).
' Parameters:
' - haystack: the string the regex is applied on.
' - originalPattern: the regex pattern with or without named groups.
' The group naming has to follow .net regex syntax: '(?<group name>group content)'.
' Group names may contain the following characters: a-z, A-Z, _ (underscore).
' Group names must not be an empty string.
' - options: a string that may contain:
' - 'i' (the regex will work case-insensitive)
' - 'g' (the regex will work globally)
' - 'm' (the regex will work in multi-line mode)
' or any combination of these.
' Returned value: a Scripting.Dictionary object with the following entries:
' - Item 0 or "0", 1 or "1" ... for the groups content/submatches,
' following the convention of VBScript_RegExp_55.SubMatches collection, which is 0-based.
' - Item Match.Count - 2 or "98" for the whole match, assuming that the number of groups is below.
' - Item Match.Count - 1 or "99" for number of groups/submatches.
' Changes compared to the original version:
' - Handles non-capturing and positive and negative lookahead groups.
' - Handles left parenthesis inside a character class.
' - Named groups do not count twice.
' E.g. in the original version the second named group occupies items 3 and 4 of the returned
' dictionary, in this revised version only item 1 (item 0 is the first named group).
' - Additional 'm' option.
' - Fixed fetching cached regexes.
' - Early binding.
' - Some code cleaning.
' For an example take a look at the 'TestRegexMatch' procedure above.
Function RegexMatch(ByVal haystack As String, ByVal originalPattern As String, Optional ByVal options As String) As Scripting.Dictionary
Dim GroupsPattern As String
Dim RealPattern As String
Dim RealRegExp As VBScript_RegExp_55.RegExp
Dim RealMatches As VBScript_RegExp_55.MatchCollection
Dim ReturnData As Scripting.Dictionary
Dim GroupNames As VBScript_RegExp_55.MatchCollection
Dim Ctr As Integer
' Cache regexes and group names for optimisation.
Static CachedRegExps As Scripting.Dictionary
Static CachedGroupNames As Scripting.Dictionary
' Group 'meta'-regex used to detect named and unnamed capturing groups.
Static GroupsRegExp As VBScript_RegExp_55.RegExp
If CachedRegExps Is Nothing Then Set CachedRegExps = New Scripting.Dictionary
If CachedGroupNames Is Nothing Then Set CachedGroupNames = New Scripting.Dictionary
If GroupsRegExp Is Nothing Then
Set GroupsRegExp = New VBScript_RegExp_55.RegExp
' Original version: GroupsRegExp.Pattern = "\((?:\?\<(.*?)\>)?"
GroupsRegExp.Pattern = "\((?!(?:\?:|\?=|\?!|[^\]\[]*?\]))(?:\?<([a-zA-Z0-9_]+?)>)?"
GroupsRegExp.Global = True
End If
' If the pattern isn't cached, create it.
If Not CachedRegExps.Exists("(" & options & ")" & originalPattern) Then
' Prepare the pattern for retrieving named and unnamed groups.
GroupsPattern = Replace(Replace(Replace(Replace(originalPattern, "\\", "X"), "\(", "X"), "\[", "X"), "\]", "X")
' Store group names for optimisation.
CachedGroupNames.Add "(" & options & ")" & originalPattern, GroupsRegExp.Execute(GroupsPattern)
' Create new VBScript regex valid pattern and set regex for this pattern.
RealPattern = GroupsRegExp.Replace(originalPattern, "(")
Set RealRegExp = New VBScript_RegExp_55.RegExp
RealRegExp.Pattern = RealPattern
' Set regex options.
For Ctr = 1 To Len(options)
Select Case Mid(options, Ctr, 1)
Case "i"
RealRegExp.IgnoreCase = True
Case "g"
RealRegExp.Global = True
Case "m"
RealRegExp.MultiLine = True
End Select
Next
' Store this regex for optimisation.
CachedRegExps.Add "(" & options & ")" & originalPattern, RealRegExp
End If
' Get matches.
Set RealMatches = CachedRegExps.Item("(" & options & ")" & originalPattern).Execute(haystack)
' Get group names.
Set GroupNames = CachedGroupNames.Item("(" & options & ")" & originalPattern)
' Create dictionary to return.
Set ReturnData = New Scripting.Dictionary
' Fill dictionary with names and indexes as descibed in the remarks introducing this procedure.
For Ctr = 1 To GroupNames.Count
If IsEmpty(GroupNames(Ctr - 1).SubMatches(0)) Then
ReturnData.Add CStr(Ctr - 1), RealMatches(0).SubMatches(Ctr - 1)
Else
ReturnData.Add GroupNames(Ctr - 1).SubMatches(0), RealMatches(0).SubMatches(Ctr - 1)
End If
Next
ReturnData.Add "98", RealMatches.Item(0)
ReturnData.Add "99", GroupNames.Count
' Return the result.
Set RegexMatch = ReturnData
End Function
为进一步改进,此代码可以作为替换VBScript正则表达式的类模块的基础。