Question

我不知道我尝试执行的查询是否可行，但是如果你们中的LINQ to SQL / XML专家之一能弄清楚这一点，我将非常感谢并向LINQ神致敬。我的最终目标是识别所有重复的XML模型，并显示除一个以外的所有重复的CECID。所以可以说我有一个看起来像这样的Xdocument：

<ApplianceModels xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema" ApplianceType="IceMakers">
    <Model>
        <ReferenceNumber>201877149</ReferenceNumber>
        <Action>C</Action>
        <Brand>4564</Brand>
        <ModelNumber>1234212</ModelNumber>
        <EquipmentType>A</EquipmentType>
        <CoolingType>W</CoolingType>
        <IceType>C</IceType>
        <IceMakerProcessType>B</IceMakerProcessType>
        <TestLabCode>ARN3190</TestLabCode>
        <ManufacturerCode>ARN2396</ManufacturerCode>
        <HarvestRateLbs24Hr>56</HarvestRateLbs24Hr>
        <EnergyCons_kWhPer100Lbs>4.00</EnergyCons_kWhPer100Lbs>
        <WaterCons_galPer100Lbs>12</WaterCons_galPer100Lbs>
        <IceHardnessAdjustmentFactor xsi:nil="true" />
        <RegulatoryStatus>I</RegulatoryStatus>
        <CECID>d579ae7a-f3f7-4627-a3f1-f17b23aa28e3</CECID>
    </Model>
    <Model>
        <ReferenceNumber>201877143</ReferenceNumber>
        <Action>C</Action>
        <Brand>4564</Brand>
        <ModelNumber>12342</ModelNumber>
        <EquipmentType>A</EquipmentType>
        <CoolingType>W</CoolingType>
        <IceType>C</IceType>
        <IceMakerProcessType>B</IceMakerProcessType>
        <TestLabCode>ARN3190</TestLabCode>
        <ManufacturerCode>ARN2396</ManufacturerCode>
        <HarvestRateLbs24Hr>56</HarvestRateLbs24Hr>
        <EnergyCons_kWhPer100Lbs>4.00</EnergyCons_kWhPer100Lbs>
        <WaterCons_galPer100Lbs>12</WaterCons_galPer100Lbs>
        <IceHardnessAdjustmentFactor xsi:nil="true" />
        <RegulatoryStatus>I</RegulatoryStatus>
        <CECID>94c6d6e6-5b6a-4f45-a7ff-70a64e50e4e6</CECID>
    </Model>
    <Model>
        <ReferenceNumber>201877152</ReferenceNumber>
        <Action>C</Action>
        <Brand>4564</Brand>
        <ModelNumber>1231114234</ModelNumber>
        <EquipmentType>A</EquipmentType>
        <CoolingType>W</CoolingType>
        <IceType>C</IceType>
        <IceMakerProcessType>C</IceMakerProcessType>
        <TestLabCode>ARN3190</TestLabCode>
        <ManufacturerCode>ARN2396</ManufacturerCode>
        <HarvestRateLbs24Hr>81</HarvestRateLbs24Hr>
        <EnergyCons_kWhPer100Lbs>1.10</EnergyCons_kWhPer100Lbs>
        <WaterCons_galPer100Lbs>12</WaterCons_galPer100Lbs>
        <IceHardnessAdjustmentFactor>4.45</IceHardnessAdjustmentFactor>
        <RegulatoryStatus>I</RegulatoryStatus>
        <CECID>d97a603c-1836-43a3-b564-ab8d1bdec65f</CECID>
    </Model>
</ApplianceModels>

然后在SQL Server中，有一个名为tApplianceTypeColumns的表，对于给定的设备类型，该表如下所示：

ApplianceTypeID       ApplianceColumnUnique        ApplianceColumnName
10                    0                            ReferenceNumber
10                    1                            Brand
10                    1                            ModelNumber
10                    0                            EquipmentType
10                    0                            CoolingType
10                    0                            IceType
10                    0                            IceMakerProcessType
10                    0                            HarvestRateLbs24Hr
10                    0                            EnergyCons_kWhPer100Lbs
10                    0                            WaterCons_galPer100lbs
10                    1                            RegulatoryStatus

所以这是我的开始，但距离还很遥远：

var DupeItems = from m in doc.Descendants("Model").Elements()
                join at in entities.tApplianceTypeColumns on m.Name equals at.ApplianceColumnName
                group m by m.Element(at.ApplianceColumnName).Value into d
                where at.ApplianceTypeID == ApplianceTypeID

所以我真的希望能够按品牌，型号和监管状态分组，这些是tApplianceTypeColumns表中将ApplianceColumnUnique位列设置为true的列。真实位数可能会有所不同，具体取决于我在该表中查找的ApplianceTypeID。

此外，我还需要在分组中包括两个永远不在tApplianceTypeColumns表中的元素，这些元素是Action然后是ManufacturerCode，然后是tApplianceTypeColumns中的所有其他唯一元素（没有特定顺序）。

ApplianceTypeID是一个已知参数，将传递给查询。因此，对于任何重复项，我都需要显示第二个及后续重复项的CECID，以便我可以获取这些CECID并在其他表中进行查找以更改其状态。但是第一步很难。我不在乎哪些重复项不会显示。我只需要显示除1以外的所有其他内容。我希望我已经对此进行了充分解释。

Answer 1

任务可以分为3个步骤：

找到要与之分组的唯一列：

所以我真的希望能够按品牌，型号和监管状态分组，这些是tApplianceTypeColumns表中将ApplianceColumnUnique位列设置为true的列。 真实位数可能会有所不同，具体取决于我在该表中查找的ApplianceTypeID。 另外，我还需要在分组中包括两个元素，这些元素永远不在tApplianceTypeColumns表中，并且这些元素依次为 Action和ManufacturerCode tApplianceTypeColumns中的所有其他唯一元素都没有特定顺序。
```
Enumerable.Concat(
    "Action,ManufacturerCode".Split(','),
    applianceTypeColumns
        .Where(at => at.ApplianceColumnUnique)
        .Select(at => at.ApplianceColumnName)
);
```

按上一步中的列将模型分组：

我们将列名投影到每个模型的列值中

applianceModels.GroupBy(
    model => uniqueColumns.Select(columnName => model.Element(columnName)?.Value).ToArray()

但是，我们不能仅按字符串数组进行分组，因此我们需要提供一个自定义IEqualityComparer：

new LambdaComparer<string[]>((a, b) => a.SequenceEqual(b), x => x.Aggregate(13, (hash, y) => hash * 7 + y?.GetHashCode() ?? 0))

汇总重复项：

.Select(g => new { g.Key, Duplicates = g.Select(x => x.Element("CECID")?.Value) })

一切融合在一起：

void Main()
{
    const int ApplianceTypeID = 10;

    var applianceModels = GetApplianceModels().XPathSelectElements("Model"); //.Dump();
    var applianceTypeColumns = GetApplianceTypeColumns().Where(x => x.ApplianceTypeID == ApplianceTypeID); //.Dump();

    var uniqueColumns = Enumerable.Concat(
        "Action,ManufacturerCode".Split(','),
        applianceTypeColumns
            .Where(at => at.ApplianceColumnUnique)
            .Select(at => at.ApplianceColumnName)
    );

    var query = applianceModels
        .GroupBy(
            model => uniqueColumns.Select(columnName => model.Element(columnName)?.Value).ToArray(),
            new LambdaComparer<string[]>((a, b) => a.SequenceEqual(b), x => x.Aggregate(13, (hash, y) => hash * 7 + y?.GetHashCode() ?? 0))
        )
        .Where(x => x.Count() > 1)
        .Select(g => new { g.Key, Duplicates = g.Select(x => x.Element("CECID")?.Value) });
        //.Dump();
}

// Define other methods and classes here
XElement GetApplianceModels()
{
    return XElement.Parse(
@"<ApplianceModels xmlns:xsi=""http://www.w3.org/2001/XMLSchema-instance"" xmlns:xsd=""http://www.w3.org/2001/XMLSchema"" ApplianceType=""IceMakers"">
    <Model>
        <ReferenceNumber>201877149</ReferenceNumber>
        <Action>C</Action>
        <Brand>4564</Brand>
        <ModelNumber>1234212</ModelNumber>
        <EquipmentType>A</EquipmentType>
        <CoolingType>W</CoolingType>
        <IceType>C</IceType>
        <IceMakerProcessType>B</IceMakerProcessType>
        <TestLabCode>ARN3190</TestLabCode>
        <ManufacturerCode>ARN2396</ManufacturerCode>
        <HarvestRateLbs24Hr>56</HarvestRateLbs24Hr>
        <EnergyCons_kWhPer100Lbs>4.00</EnergyCons_kWhPer100Lbs>
        <WaterCons_galPer100Lbs>12</WaterCons_galPer100Lbs>
        <IceHardnessAdjustmentFactor xsi:nil=""true"" />
        <RegulatoryStatus>I</RegulatoryStatus>
        <CECID>d579ae7a-f3f7-4627-a3f1-f17b23aa28e3</CECID>
    </Model>
    <Model>
        <ReferenceNumber>201877143</ReferenceNumber>
        <Action>C</Action>
        <Brand>4564</Brand>
        <ModelNumber>12342</ModelNumber>
        <EquipmentType>A</EquipmentType>
        <CoolingType>W</CoolingType>
        <IceType>C</IceType>
        <IceMakerProcessType>B</IceMakerProcessType>
        <TestLabCode>ARN3190</TestLabCode>
        <ManufacturerCode>ARN2396</ManufacturerCode>
        <HarvestRateLbs24Hr>56</HarvestRateLbs24Hr>
        <EnergyCons_kWhPer100Lbs>4.00</EnergyCons_kWhPer100Lbs>
        <WaterCons_galPer100Lbs>12</WaterCons_galPer100Lbs>
        <IceHardnessAdjustmentFactor xsi:nil=""true"" />
        <RegulatoryStatus>I</RegulatoryStatus>
        <CECID>94c6d6e6-5b6a-4f45-a7ff-70a64e50e4e6</CECID>
    </Model>
    <Model>
        <ReferenceNumber>201877152</ReferenceNumber>
        <Action>C</Action>
        <Brand>4564</Brand>
        <ModelNumber>1231114234</ModelNumber>
        <EquipmentType>A</EquipmentType>
        <CoolingType>W</CoolingType>
        <IceType>C</IceType>
        <IceMakerProcessType>C</IceMakerProcessType>
        <TestLabCode>ARN3190</TestLabCode>
        <ManufacturerCode>ARN2396</ManufacturerCode>
        <HarvestRateLbs24Hr>81</HarvestRateLbs24Hr>
        <EnergyCons_kWhPer100Lbs>1.10</EnergyCons_kWhPer100Lbs>
        <WaterCons_galPer100Lbs>12</WaterCons_galPer100Lbs>
        <IceHardnessAdjustmentFactor>4.45</IceHardnessAdjustmentFactor>
        <RegulatoryStatus>I</RegulatoryStatus>
        <CECID>d97a603c-1836-43a3-b564-ab8d1bdec65f</CECID>
    </Model>
</ApplianceModels>");
}
IEnumerable<(int ApplianceTypeID, bool ApplianceColumnUnique, string ApplianceColumnName)> GetApplianceTypeColumns()
{
    var data =
@"ApplianceTypeID       ApplianceColumnUnique        ApplianceColumnName
10                    0                            ReferenceNumber
10                    1                            Brand
10                    1                            ModelNumber
10                    0                            EquipmentType
10                    0                            CoolingType
10                    0                            IceType
10                    0                            IceMakerProcessType
10                    0                            HarvestRateLbs24Hr
10                    0                            EnergyCons_kWhPer100Lbs
10                    0                            WaterCons_galPer100lbs
10                    1                            RegulatoryStatus";
    return Regex.Matches(data, @"^(\d+)\s+(\d+)\s+(\w+)", RegexOptions.Multiline)
        .Cast<Match>()
        .Select(x => 
        (
            /*ApplianceTypeID = */int.Parse(x.Groups[1].Value),
            /*ApplianceColumnUnique = */int.Parse(x.Groups[2].Value) != 0,
            /*ApplianceColumnName = */x.Groups[3].Value
        ));
}

class LambdaComparer<T> : IEqualityComparer<T>
{
    private readonly Func<T, T, bool> equals;
    private readonly Func<T, int> getHashCode;

    public LambdaComparer(Func<T, T, bool> equals, Func<T, int> getHashCode)
    {
        this.equals = equals;
        this.getHashCode = getHashCode;
    }

    public bool Equals(T x, T y) => equals(x, y);
    public int GetHashCode(T obj) => getHashCode(obj);
}

Answer 2

这是我基于Xiaoy312解决方案的最终代码。再次感谢你。它运作良好。我向LINQ神致敬：

private List<string> XMLDuplicatesToEliminate(XDocument doc, Guid ApplianceTypeID)
{
    var entities = new DbContextFactory().MAEDBSEntities;

    var applianceModels = doc.Descendants("Model");
    var applianceTypeColumns =
    (
        from at in entities.tApplianceTypeColumns
        where
            at.ApplianceTypeID == ApplianceTypeID &&
            at.ApplianceColumnUnique == true
        select new { at.ApplianceColumnName }
    ).ToList();

    var uniqueColumns = Enumerable.Concat(
        "Action,ManufacturerCode".Split(','),
        applianceTypeColumns
            .Select(at => at.ApplianceColumnName)
    );

    List<string> DuplicatesToEliminate = new List<string>();
    var duplicates = applianceModels
        .GroupBy(
            model => uniqueColumns.Select(columnName => model.Element(columnName)?.Value).ToArray(),
            new LambdaComparer<string[]>((a, b) => a.SequenceEqual(b), x => x.Aggregate(13, (hash, y) => hash * 7 + y?.GetHashCode() ?? 0)))
        .Where(x => x.Count() > 1)
        .Select(g => new { g.Key, Duplicates = g.Select(x => x.Element("CECID")?.Value) })
        .ToList();

    foreach (var duperow in duplicates)
    {
        string firstdupe = duperow.Duplicates.First();
        IEnumerable<string> allbutone = duperow.Duplicates.Where(x => x != firstdupe);
        foreach (string dupeitem in allbutone)
        {
            DuplicatesToEliminate.Add(dupeitem);
        }
    }

    return DuplicatesToEliminate;
}

复杂的LINQ to XML查询辅助

2 个答案: