PowerBuilder

Regular Expression with Multiple Matches and Capture Groups

See more Regular Expressions Examples

Demonstrates a regular expression with multiple matches and capture groups for each match.

Note: Chilkat uses PCRE2. See PCRE2 Regular Expressions
Also see: PCRE2 Performance

PowerBuilder

integer li_rc
integer li_Success
string ls_Pattern
oleobject loo_Sb
integer li_Crlf
oleobject loo_Json
integer li_TimeoutMs
integer li_NumMatches
string ls_Cap
integer i
integer li_MatchCount
integer j
integer li_NumCaptureGroups

li_Success = 0

ls_Pattern = "Name:\s+(\w+)\s+(\w+),\s+Email:\s+(\S+)"

loo_Sb = create oleobject
li_rc = loo_Sb.ConnectToNewObject("Chilkat.StringBuilder")
if li_rc < 0 then
    destroy loo_Sb
    MessageBox("Error","Connecting to COM object failed")
    return
end if
li_Crlf = 1
loo_Sb.AppendLine("Name: John Smith, Email: john.smith@example.com",li_Crlf)
loo_Sb.AppendLine("Name: Jack Johnson, Email: jack.johnson@example.com",li_Crlf)
loo_Sb.AppendLine("Name: Mary Adams, Email: mary.adams@example.com",li_Crlf)

Write-Debug loo_Sb.GetAsString()

// We have the following string:
// Name: John Smith, Email: john.smith@example.com
// Name: Jack Johnson, Email: jack.johnson@example.com
// Name: Mary Adams, Email: mary.adams@example.com

loo_Json = create oleobject
li_rc = loo_Json.ConnectToNewObject("Chilkat.JsonObject")

loo_Json.EmitCompact = 0

li_TimeoutMs = 2000
li_NumMatches = loo_Sb.RegexMatch(ls_Pattern,loo_Json,li_TimeoutMs)
if li_NumMatches < 0 then
    // Probably an error in the regular expression.
    // Suggestion: Use AI to help create and/or diagnose regular expressions.
    Write-Debug loo_Sb.LastErrorText
    destroy loo_Sb
    destroy loo_Json
    return
end if

// Examine the matches:
Write-Debug loo_Json.Emit()

// This is the JSON with the match information.
// See the JSON parsing code below to get the matched capture group values.

// {
//   "match": [
//     {
//       "group": [
//         {
//           "cap": "Name: John Smith, Email: john.smith@example.com",
//           "idx": 0,
//           "len": 47
//         },
//         {
//           "cap": "John",
//           "idx": 6,
//           "len": 4
//         },
//         {
//           "cap": "Smith",
//           "idx": 11,
//           "len": 5
//         },
//         {
//           "cap": "john.smith@example.com",
//           "idx": 25,
//           "len": 22
//         }
//       ]
//     },
//     {
//       "group": [
//         {
//           "cap": "Name: Jack Johnson, Email: jack.johnson@example.com",
//           "idx": 49,
//           "len": 51
//         },
//         {
//           "cap": "Jack",
//           "idx": 55,
//           "len": 4
//         },
//         {
//           "cap": "Johnson",
//           "idx": 60,
//           "len": 7
//         },
//         {
//           "cap": "jack.johnson@example.com",
//           "idx": 76,
//           "len": 24
//         }
//       ]
//     },
//     {
//       "group": [
//         {
//           "cap": "Name: Mary Adams, Email: mary.adams@example.com",
//           "idx": 102,
//           "len": 47
//         },
//         {
//           "cap": "Mary",
//           "idx": 108,
//           "len": 4
//         },
//         {
//           "cap": "Adams",
//           "idx": 113,
//           "len": 5
//         },
//         {
//           "cap": "mary.adams@example.com",
//           "idx": 127,
//           "len": 22
//         }
//       ]
//     }
//   ]
// }

// Important:  Capture group 0 always contains the entire match — that is, the portion of the input string that matches the full regular expression.

i = 0
li_MatchCount = loo_Json.SizeOfArray("match")
do while i < li_MatchCount
    Write-Debug "Match " + string(i + 1) + ":"
    loo_Json.I = i
    j = 0
    li_NumCaptureGroups = loo_Json.SizeOfArray("match[i].group")
    do while j < li_NumCaptureGroups
        loo_Json.J = j
        ls_Cap = loo_Json.StringOf("match[i].group[j].cap")
        Write-Debug string(j) + ": " + ls_Cap
        j = j + 1
    loop
    i = i + 1
loop

// Capture group 0 always contains the entire match — that is, the portion of the input string that matches the full regular expression.

// Output

// Match 1:
// 0: Name: John Smith, Email: john.smith@example.com
// 1: John
// 2: Smith
// 3: john.smith@example.com
// Match 2:
// 0: Name: Jack Johnson, Email: jack.johnson@example.com
// 1: Jack
// 2: Johnson
// 3: jack.johnson@example.com
// Match 3:
// 0: Name: Mary Adams, Email: mary.adams@example.com
// 1: Mary
// 2: Adams
// 3: mary.adams@example.co


destroy loo_Sb
destroy loo_Json