Question

我有一个包含50000行营业时间数据的excel电子表格（以各种可能的格式）：

THU 4P-9P F 9A-9P SAT，SUN 9A-6P
WED＆amp; THU 10A - 3P FRI 10A - 1P
MON - FRI 6P - 10P
M - SA 9A - 9P
SUN-SAT 9-5
SU - SA 8A-10P
TUE - FRI 10A - 6P SAT 12P - 4P
MON-FRI 730A-4P / SAT 9A-12P
SUN 6A-5P / M-F 6A-9P / SAT 5A-9P

我需要将其转换为：

$conn=mysql_connect('localhost','root','');
mysql_select_db('alinDataBase');

$idcat = $_GET['id'];
$query =  "SELECT * FROM electrocasnice WHERE subcat = '$idcat' ";
$result = mysql_query($query,$conn);
$output='';
$records = []; //define it globally.

while($row = mysql_fetch_array($result)){
    $output=array(
    "nume" => $row["nume"],
    "pret" => $row["pret"],
    "imagine" => $row["imaginepath"]
    );
    array_push($records, $output); //Push each objects into array
}
echo json_encode($records);
mysql_close($conn);

我想到了：

Days Open A Week: 2 
Hours Open a Week: 15

每一天，但我想知道除了硬编码的可能性之外，是否有更简单的方法来清理这些数据。

谢谢！

Answer 1

这是一些处理提供的样本数据的VBA。它几乎肯定不适用于更大的样本，但它是一个开始。

Public Function DaysAndHours(ByVal sInput As String)

    Dim vaTokens As Variant
    Dim i As Long, j As Long
    Dim vaSeps As Variant
    Dim dtTime As Date, dtStart As Date, dtEnd As Date
    Dim dHours As Double, dTodayHours As Double
    Dim lStartDay As Long, lEndDay As Long
    Dim bThrough As Boolean
    Dim dcDay As Scripting.Dictionary
    Dim vItem As Variant

    Set dcDay = New Scripting.Dictionary

    'These are all the characters that split the data
    'as you discover more characters, add them here
    vaSeps = Split("- , / &")

    'If the data has times like 730, Excel can't tell it's a time
    'so this adds a colon before 30 and 15 assuming it's unlikely
    'anyone would open or close on other than a quarter hour
    'then if there was already a colon there, it would be doubled
    'so remove double colons
    sInput = Replace(sInput, "30", ":30")
    sInput = Replace(sInput, "15", ":15")
    sInput = Replace(sInput, "::", ":")

    'Some separators have spaces around them and some don't. This changes
    'all separators so they have spaces. This is so our split creates
    'proper tokens
    For j = LBound(vaSeps) To UBound(vaSeps)
        sInput = Replace(sInput, vaSeps(j), Space(1) & vaSeps(j) & Space(1))
    Next j

    'If the separators already had spaces around them, they would be
    'doubled. Trim removes double spaces
    sInput = Application.Trim(sInput)

    vaTokens = Split(sInput, Space(1))

    'Assume the first token is a day, and put it in the
    'dictionary at zero hours
    lStartDay = GetDayFromInit(vaTokens(LBound(vaTokens)))
    dcDay.Add lStartDay, 0

    For i = LBound(vaTokens) + 1 To UBound(vaTokens)
        'Some separators are "through" meaning that all the days in between
        'the two days are included. Other separators just list discrete days
        If IsSep(vaTokens(i), vaSeps) Then
            Select Case vaTokens(i)
                Case "-"
                    bThrough = True
                Case Else
                    bThrough = False
            End Select
        Else
            'Excel won't convert a straight number to a time, so this
            'adds :00 to make it look like a time
            If IsNumeric(vaTokens(i)) Then
                vaTokens(i) = vaTokens(i) & ":00"
            End If

            'Try to change the token into a time. If it
            'works, we're dealing with times, otherwise days
            On Error Resume Next
                dtTime = TimeValue(vaTokens(i))
            On Error GoTo 0

            If dtTime > 0 Then 'the current token is a time
                If dtStart > 0 Then 'we've already converted a time, so this must be the end time
                    dtEnd = dtTime
                    If dtEnd < dtStart Then dtEnd = dtEnd + TimeSerial(12, 0, 0) 'make sure the end time is after the start time
                    dTodayHours = dtEnd - dtStart 'compute the hours open

                    'For every day that we haven't filled a time, put this time
                    For j = 0 To dcDay.Count - 1
                        If dcDay.Items(j) = 0 Then
                            dcDay.Item(dcDay.Keys(j)) = dTodayHours
                        End If
                    Next j
                    dtStart = 0: dtEnd = 0: dtTime = 0 'reset
                    bThrough = False 'reset
                Else 'We haven't already filled a time, so this must be the start time
                    dtStart = dtTime
                End If
            Else 'the current token isn't a time, it must be day
                'we've encountered a through separator, so we've alreay got a start day
                'and this token is the end day
                If bThrough Then
                    lEndDay = GetDayFromInit(vaTokens(i))
                    'If the days are in the right order, just add them
                    'in order to the dictionary
                    If lStartDay < lEndDay Then
                        For j = lStartDay To lEndDay
                            If Not dcDay.Exists(j) Then
                                dcDay.Add j, 0
                            End If
                        Next j
                    Else 'Days are in the wrong order (where Sunday = 1)
                        For j = 1 To lEndDay
                            If Not dcDay.Exists(j) Then
                                dcDay.Add j, 0
                            End If
                        Next j
                        For j = lStartDay To 7
                            If Not dcDay.Exists(j) Then
                                dcDay.Add j, 0
                            End If
                        Next j
                    End If
                Else 'We haven't encountered a through operator, so this is a lone day or the first of a range
                    lStartDay = GetDayFromInit(vaTokens(i))
                    If Not dcDay.Exists(lStartDay) Then
                        dcDay.Add lStartDay, 0
                    End If
                End If
            End If
        End If
    Next i

    DaysAndHours = dcDay.Count & " days, " & Application.Sum(dcDay.Items) * 24 & " hours"

End Function

Public Function GetDayFromInit(ByVal sInit As String) As Long

    Dim vaDays As Variant
    Dim i As Long

    vaDays = Split("SUNDAY MONDAY TUESDAY WEDNESDAY THURSDAY FRIDAY SATURDAY")

    For i = 0 To 6
        If UCase(sInit) = Left$(vaDays(i), Len(sInit)) Then
            GetDayFromInit = i + 1
            Exit For
        End If
    Next i

End Function

Public Function IsSep(ByVal sChar As String, ByRef vSeps As Variant) As Boolean

    IsSep = InStr(1, Join(vSeps), sChar)

End Function

清洁数据：＆＃34;周一至周五＆＃34;或者＆＃34;周六到周四＆＃34;

1 个答案: