donnytian / npoi.mapper Goto Github PK

Hello, does the library support the table style? Such as background color or font color or font type and so on

Before using the mapper, I apply a .trim() to the headers to remove empty spaces. But the problem comes when trying to match with the rows that don't have the column name with the .trim() method used on them. Do you know if there is a method or a place where I can apply the trim to the column names of the rows?

Hi,

My demo excel has two headers, the first header for Chinese Name and the second one is the actual column name of the table.

I have tried to set HeaderRowIndex to 1 or 2, but the mapping result is not expected, which the value of the properties is the default value. But if I just keep single hearer, it works fine.

Below are my code and excel:
`

            var mapper = new Mapper(fullPath);
            mapper.HeaderRowIndex = 1;
            var objs = mapper.Take<ProductDTO>().ToList();

`

Mapper class Should implements IDispose , and set _workbook = null in Dispose method;
Otherwise it will cause memory leak.

Get attribute value as friendly display name if property has DisplayName attribute.

Or you design a custom attribute enhancement class that provides a custom friendly display name, column display order, and so on.

Or does Npoi.Mapper already have a similar function? Do you have a better idea?

Sample data model code:

public class CatalogCareViewModel
{
	/// <summary>
	/// ISO/国际标准
	/// </summary>
	public virtual string ISO { get; set; }

	/// <summary>
	/// German/德语
	/// </summary>
	[DisplayName("German")]
	public virtual string DE { get; set; }

	/// <summary>
	/// English/英语
	/// </summary>
	[DisplayName("English")]
	public virtual string EN { get; set; }

	/// <summary>
	/// French/法语
	/// </summary>
	[DisplayName("French")]
	public virtual string FR { get; set; }

	/// <summary>
	/// Russion/俄语
	/// </summary>
	[DisplayName("Russion")]
	public virtual string RU { get; set; }

	/// <summary>
	/// Italian/意大利语
	/// </summary>
	[DisplayName("Italian")]
	public virtual string IT { get; set; }

	/// <summary>
	/// Dutch/荷兰语
	/// </summary>
	[DisplayName("Dutch")]
	public virtual string NL { get; set; }

	/// <summary>
	/// Norwegian/挪威
	/// </summary>
	[DisplayName("Norwegian")]
	public virtual string NO { get; set; }
}

Sample excel sheet table

Was trying to insert a DateTime using Put, but It always returned me the numeric value of the Date.

I even reproduced it on the tests.

https://i.imgur.com/trrQhor.png

I'm trying to import a Excel sheet which contains several columns which can have date values as double. Here is my approach:

var mapper = new Mapper(filePath);
mapper.TrackObjects = false;
mapper.Map<Model>(0, o => o.Property1);
mapper.Map<Model>(1, o => o.Property2);
...
mapper.Map(column =>
{
    if (column.Attribute.Index == 10 && column.CurrentValue is double value)
    {
        column.CurrentValue = DateTime.FromOADate(value);
    }
    
    return true;
});

var result = mapper.Take<Model>().ToList();

When debugging, the debugger dont even stop in the mapper.Map(column => ...) method.

All I got as a result is a list of errors for the specific columns:
"Object reference not set to an instance of an object."

How do I have configure the mapper to convert the double to date?
Any help is welcomed!

In MapHelper.cs, there is a static list to contains cached custom styles. In an multi-thread scenario, styles will be messed by different workbooks.

We should cache these styles as instance property.

Without setting a format on DateTime properties, excel seems to format them as integers. For example,
10/26/2016 prints as 42669.

I see that I can easily format a DateTime property if I know it in advance, like this:
mapper.Format<MyObject>("yyyy-MM-dd", x => x.MyDateField)

I tried to make it generic, so I don't need a custom method for every one of my types:

foreach (var property in typeof(T).GetProperties())
{
    if (property.PropertyType == typeof(DateTime) ||
        property.PropertyType == typeof(DateTime?))
    {
        mapper.Format<T>("yyyy-MM-dd", x => property.GetValue(x));
    }
}

But this fails with "Unable to cast object of type 'System.Linq.Expressions.InstanceMethodCallExpressionN' to type 'System.Linq.Expressions.UnaryExpression'."

Would you help me figure this out, or consider building in a default format for DateTime that isn't an integer?

这是Bug吗？还是我的用法不对？

When import from file for nullable value types like DateTime? and int?, if the value from Excel file cannot be converted to the target type, no error reported, just set null to the object.

The expected result is to report errors or use setting to control whether or not report error for specific column/ property.

用户故意上传了个文件格式合法，但列格式不合法的文件。
此时，我希望给用户一个友好的提示（列格式有误），而不是让用户顺利上传成功。
例如：

public class MyEntity
{
    public string Name { get; set; }
    public string Address { get; set; }
}

Excel

Comany|State|City

通过你的类库，能上传成功，但获取的实体类列表都是空值。
我希望有列名校验之类的功能。不知道你的Mapper.Map扩展方法能否做到。
或者你有更好的建议？？

FEATURE REQUEST

Is there a support to map string to enum values, based on the defined DescriptionAttribute or DisplayAttribute (would choose probably this) on then enum values?

Common scenario is to use abbreviations with enums as well. Also the enum's property is used inside the code (like gold/silver/bronze client) and in excel we might consider to use some abbreviations (G/S/B) which when importing from excel should be converted to the proper enum value.

base on my test

1. the whole row will be null if your .Task<Import> , the Import contains DateTimeOffset?
2. DateTime? is Ok

I think the DateTimeOffset should be support base on Current TimeZone of the thread

	var reader = new Mapper(@"D:\111111.xlsx");
	var set = reader.Take<dynamic>().Dump().Select(x => x.Value).ToList();

	var writer = new Mapper();
	writer.Put(set, "Sheet1");
	writer.Save(@"D:\222222.xlsx");

D:\111111.xlsx只有标题行，没有数据
D:\222222.xlsx完全空白，标题行都没有了

Appears to be an issue in Mapper.cs -> Put( ...

SetCell not always called if using resolver.

From this:-

                if (column.Resolver != null)
                {
                    column.Resolver.TryPutCell(column, out value, o);
                }
                else
                {
                    SetCell(cell, value, column, setStyle: overwrite);
                }

To this:-

                if (column.Resolver != null)
                {
                    column.Resolver.TryPutCell(column, out value, o);
                }

                SetCell(cell, value, column, setStyle: overwrite);

Regards,

Gary

FEATURE REQUEST

In some cases (localization) the Y/N value do change (A/N or T/F or I/N).

Is there an extended Nullable mapping support which could handle locale string value to Nullable conversion?

Haven't found the possibility to get a list of all column headers.
Is there any way to get them using the NPOI Workbook maybe?

Best regards

使用Npoi IWorkbook的方式导出的excel在两者中打开都正常，使用3.3.0版本的Npoi.Mapper导出，在Microsoft Office中打开会提示如图
无论点击是或者否都不能正常打开。
` //var stream = new MemoryStream();
//var mapper = new Mapper();
//mapper.Format("yyyy-MM-dd", ps=> ps.DeliverPeriod)
// .Format("yyyy-MM-dd", ps=>ps.ApplyTime)
// // 将各个属性映射为中文列名
// .Map("序号", o=>o.SeqNO)
// .Map("合同编号", o=>o.ContractNO)
// .Map("产品编号", o=>o.ProductNO)
// .Map("客户", o=>o.CustomerName)
// .Map("营销代表", o=>o.MarketerName)
// .Map("交货日期", o=>o.DeliverPeriod)
// .Map("物料代码", o=>o.Code)
// .Map("产品型号", o=>o.ProductModelName)
// .Map("规格参数", o=>o.Spec)
// .Map("数量", o=>o.Qty)
// .Map("创建时间", o=>o.ApplyTime)
// .Map("备注", o=>o.Remarks)
// .Map("是否禁用", o=>o.Disabled);

        //// 将数据保存到stream中
        //mapper.Save(stream, objs, "生产计划评审");
        //// 获取stream的buffer
        //var buffer = stream.GetBuffer();
        //stream.Dispose();
        //return buffer;`

其中objs为List的数据

能否在没有自定义映射规则的情况下，将"列标题映射的默认命名约定"作用于Excel导出。

举例：
我没有调用Map<TModel>(parameters...)，但我希望导出后，
DisplayNameProperty属性在Excel表格中显示的列名是Display Name，而不是DisplayNameProperty。

而ColumnNameAttributeProperty属性在Excel表格中的列名是正确显示的——By Name。
换句话说，在导出功能中，ColumnAttribute起作用了，但DisplayAttribute没起作用。

实体类SampleClass

using System;
using System.Collections.Generic;
using System.ComponentModel.DataAnnotations;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using Npoi.Mapper.Attributes;

namespace Sample
{
    public class SampleClass
    {
        public string StringProperty { get; set; }

        [Column("By Name")]
        public string ColumnNameAttributeProperty { get; set; }

        [Column(11)]
        public string ColumnIndexAttributeProperty { get; set; }

        [Ignore]
        public string IgnoredAttributeProperty { get; set; }

        [Display(Name = "Display Name")]
        public string DisplayNameProperty { get; set; }

        [Column(CustomFormat = "0%")]
        public double CustomFormatProperty { get; set; }
    }
}

你好，很高兴你使用CI持续发布NuGet包。
现在有个问题：

你现在的Excel导出是根据类属性顺序导出的，而且是先输出派生类属性列，后输出基类属性列。
能否根据Mapper.Attributes集合的顺序来导出Excel?

Hi there!
Is there a way to ignore the base classes? I have already tried:
mapper.Ignore and mapper.Ignore but it is still saving it to the excel file.

Any clues on how to overcome this?

Not sure what I'm doing wrong... trying to set up a mapping to my Customer class which has some nested properties to BillingAddress, ServiceAddress, BillingContact, etc. - my excel file is 17,000 lines, with some whitespace rows in certain regions.

System.IndexOutOfRangeException: Index was outside the bounds of the array.

StackTrace:

at Npoi.Mapper.MapHelper.GetPropertyInfoByExpression[T](Expression`1 propertySelector)
at Npoi.Mapper.MapExtensions.Map[T](Mapper mapper, String columnName, Expression`1 propertySelector, Func`3 tryTake, Func`3 tryPut)

The above error is thrown, seemingly, when I try to use the mapper .. here's my code:

    Npoi.Mapper.Mapper m_mapper = new Npoi.Mapper.Mapper("exportContacts.xlsx");

    m_mapper.Map<Customer>("_kftAccountID", o => o.CustomerId)
        .Map<Customer>("accountType", o => o.AccountType)
        .Map<Customer>("addrBilling", o => o.BillingAddress.AddressLine1)
        .Map<Customer>("addrBillingCity", o => o.BillingAddress.City)
        .Map<Customer>("addrBillingCounty", o => o.BillingAddress.County)
        .Map<Customer>("addrBillingState", o => o.BillingAddress.State)
        .Map<Customer>("addrBillingZip", o => o.BillingAddress.ZipCode)
        .Map<Customer>("addrService", o => o.ServiceAddress.AddressLine1)
        .Map<Customer>("addrServiceCity", o => o.ServiceAddress.City)
        .Map<Customer>("addrServiceCounty", o => o.ServiceAddress.County)
        .Map<Customer>("addrServiceState", o => o.ServiceAddress.State)
        .Map<Customer>("addrServiceZip", o => o.ServiceAddress.ZipCode)
        .Map<Customer>("addrServiceAreaSubdivision", o => o.ServiceAreaSubdivision)
        .Map<Customer>("contactBillingCompName", o => o.BillingContact.CompanyName)
        .Map<Customer>("contactBillingEmail", o => o.BillingContact.Email)
        .Map<Customer>("contactBillingNameFirst", o => o.BillingContact.FirstName)
        .Map<Customer>("contactBillingNameLast", o => o.BillingContact.LastName)
        .Map<Customer>("contactBillingNamePrefix", o => o.BillingContact.Prefix)
        .Map<Customer>("contactBillingPhoneFax", o => o.BillingContact.FaxNumber)
        .Map<Customer>("contactBillingPhoneHome", o => o.BillingContact.PhoneHome)
        .Map<Customer>("contactBillingPhoneMobile", o => o.BillingContact.PhoneMobile)
        .Map<Customer>("contactBillingProfessionalTitle", o => o.BillingContact.ProfessionalTitle)
        .Map<Customer>("contactBillingSecondaryName", o => o.BillingContact.SecondaryName)
        .Map<Customer>("contactBillingSecondaryType", o => o.BillingContact.SecondaryType)
        .Map<Customer>("contactBillingWebSite", o => o.BillingContact.WebSite)
        .Map<Customer>("contactServiceName", o => o.ServiceContact.CompanyName)
        .Map<Customer>("contactServiceEmail", o => o.ServiceContact.Email)
        .Map<Customer>("contactServiceNameFirst", o => o.ServiceContact.FirstName)
        .Map<Customer>("contactServiceNamePrefix", o => o.ServiceContact.Prefix)
        .Map<Customer>("contactServicePhoneFax", o => o.ServiceContact.FaxNumber)
        .Map<Customer>("contactServicePhoneHome", o => o.ServiceContact.PhoneHome)
        .Map<Customer>("contactServicePhoneMobile", o => o.ServiceContact.PhoneMobile)
        .Map<Customer>("contactServicePropertyManager", o => o.ServiceContact.PropertyManager)
        .Map<Customer>("contactServiceType", o => o.ServiceContact.Type)
        .Map<Customer>("contactServiceWebSite", o => o.ServiceContact.WebSite);

Hi,
there is a message about missing method if you try to export to XLS/S with NPOI version 2.2.1 (newest) installed.
Workaround: use NPOI version 2.1.3.1.

Regards

throw System.ArgumentException: Sheet index (0) is out of range (no sheets).

/// <summary>
/// Put objects in the sheet with specified zero-based index.
/// </summary>
/// <typeparam name="T">Target object type</typeparam>
/// <param name="objects">The objects to save.</param>
/// <param name="sheetIndex">The sheet index, default is 0.</param>
/// <param name="overwrite"><c>true</c> to overwrite existing rows; otherwise append.</param>
public void Put<T>(IEnumerable<T> objects, int sheetIndex = 0, bool overwrite = true)
{
	if (Workbook == null) Workbook = new XSSFWorkbook();
	var sheet = Workbook.GetSheetAt(sheetIndex);
	Put(sheet, objects, overwrite);
}

The second line in the method needs to judge the NumberOfSheets property first.
When it‘s value is greater than 0, the GetSheetAt method is called, otherwise, the CreateSheet method is called.

You mentioned in Issue #4 the new mapper.UseFormat() method. It works great on non-nullable types and on nullable types if the data in the first row after the header is not null. I tested with DateTime and DateTime?.

However, it does not apply the format if the data in the first row is null. In the case of DateTime? this results in integers being used in Excel instead of dates. Would support for this be feasible?

Also, thank you for implementing my previously-requested feature so quickly and I feel bad asking for more!

I think having to create a custom mapper to parse System.Guid values from a spreadsheet is something the mapper should have built in

If this is not going to be supported, can an example be added to the README .. as I had a hard time making a custom mapper for doing this

.NET sample object:

    public class MySampleTupleObject
    {
        public string Id { get; set; }

        public System.Guid ProcessId { get; set; }
    }

IgnoreErrorsFor and Ignore method are not working for dynamic type.

When producing the object from a sheet only settable properties are used and, in particular, Readonly properties should be ignored. Presently, the mere presence of such properties causing a crash. Conversely, when writing a sheet Writeonly properties should be ignored. Consequently, two distinct BindingFlag seems necessary.

Please look at the code:

The original code, but ForHeader method does not work.

Mapper mapper = new Mapper();
mapper.Ignore<TModel>(t => t.Id).Put(list);

var headerCellStyle = mapper.Workbook.CreateCellStyle();
//Ignore style detail setting code...
mapper.ForHeader(cell =>
{
	cell.Sheet.AutoSizeColumn(cell.ColumnIndex);
	cell.CellStyle = headerCellStyle;
});

Modified Code, it throw NullReferenceException at the mapper.Workbook.CreateCellStyle() line

Mapper mapper = new Mapper();
mapper.Ignore<TModel>(t => t.Id);

var headerCellStyle = mapper.Workbook.CreateCellStyle();
//Ignore style detail setting code...
mapper.ForHeader(cell =>
{
	cell.Sheet.AutoSizeColumn(cell.ColumnIndex);
	cell.CellStyle = headerCellStyle;
});

mapper.Put(list);

Re-modified Code

IWorkbook workbook = new HSSFWorkbook();

Mapper mapper = new Mapper(workbook);
mapper.Ignore<TModel>(t => t.Id);

var headerCellStyle = mapper.Workbook.GetDefaultHeaderCellStyle();
//Ignore style detail setting code...
mapper.ForHeader(cell =>
{
	cell.Sheet.AutoSizeColumn(cell.ColumnIndex);
	cell.CellStyle = headerCellStyle;
});

mapper.Put(list);

I have to create an object for the Workbook property to work properly, but this is not what I want.
I looked at the code of the Mapper class, I think a IWorkbook object should be created inside the default constructor.

As follows:

/// <summary>
/// Initialize a new instance of <see cref="Mapper"/> class.
/// </summary>
public Mapper()
{
	Workbook = new HSSFWorkbook();
}

After that, the Workbook attribute non-null judgment code in your other methods can be deleted.
If you agree to do this, pay attention to your Save method

First off, awesome tool! As an enhancement, would it be possible to overload the Ignore method to pass properties by property name (as a string), or list of names. Passing in property selector is handy, but it'd be nice to pass in strings.

I know this isn't a code writing service, but, would you mind showing a sample custom resolver that converts a DateTime excel column and maps it into a SampleObj.MyDateTimeOffsetProperty - where the property is of type DateTimeOffset?

如题，WPF客户端项目中，还有很多.NET4.0的项目

There´s a column named "N.I.F.".
I set the mapper like:

_mapper = new Mapper(filePath) { IgnoredNameChars = new[] { '.' } };

The column is totally ignored when:

_mapper.Take<dynamic>()

Though it works if I change the column name to "NIF"

I want to get the title can not be estimated,
Although I found the value I needed on dynamic, I would like to get the title how to get it?

1. Is there any property to set to skip certain number of lines from beginning?
2. Does the column attribute mapping support chinese?

Thanks

挺不错的，但不支持.net framework 4.0

ForHeader委托应该在填充完数据后执行，而不是填充数据前.
这是有原因的。
如果仅仅只是设置头部样式，你的写法毫无问题。
但如果调用者想要为单元格设置自适应列宽，那么这样写就肯定有问题的。
因为NPOI提供的Sheet.AutoSizeColumn自适应列宽方法，是基于当前列最宽的那一个单元格的宽度设置整列的宽度。
所以如果调用者在ForHeader委托里调用Sheet.AutoSizeColumn自适应列宽方法，是不会提作用的。
调用者只能在调用了你的Put方法之后，自己另外写for调用Sheet.AutoSizeColumn方法。
而如果你把ForHeader委托的执行放在填充数据之后，那么调用者无须另写for代码。

所以请考虑将以下代码段移到Put方法底部。

// Injects custom action for headers.
if (overwrite && HasHeader && _headerAction != null)
{
	firstRow?.Cells.ForEach(c => _headerAction(c));
}

移到这个填充数据的代码段之后

foreach (var o in objectArray)
{
	var row = sheet.GetRow(rowIndex);

	if (overwrite && row != null)
	{
		sheet.RemoveRow(row);
		row = sheet.CreateRow(rowIndex);
	}

	row = row ?? sheet.CreateRow(rowIndex);

	foreach (var column in columns)
	{
		var pi = column.Attribute.Property;
		var value = pi?.GetValue(o);
		var cell = row.GetCell(column.Attribute.Index, MissingCellPolicy.CREATE_NULL_AS_BLANK);

		column.CurrentValue = value;
		if (column.Attribute.TryPut == null || column.Attribute.TryPut(column, o))
		{
			SetCell(cell, column.CurrentValue, column, setStyle: overwrite);
		}
	}

	rowIndex++;
}

谢谢

看了下源码，数据类型是根据列头以外的第一行数据取，假如第一行是字符串，则这一列的所有数据都会转字符串，假如是数值，这一列就全是double。
但是有一种情况是，这一列既第一行是数值，其他行有字符串，然后字符串转数值会转换失败（比如中文不能转换成数字）。
代码在类 MapHelper.InferColumnDataType
请问能不能 mapper.Take 的时候，所有列设置为字符串类型

Calling Mapper.Save and passing in a stream seems to close the stream. For files this is ok but if this is content we want to stream from a web server this causes errors when trying to write the data to the response. Would it be possible to create an overload which takes a "closeStream" boolean parameter? This would allow the existing stream to be seeked to zero and sent over the wire from asp.net core.

The firstRow variable does not get the object returned by the PopulateFirstRow method

private void Put<T>(ISheet sheet, IEnumerable<T> objects, bool overwrite)
{
	var sheetName = sheet.SheetName;
	var firstRow = sheet.GetRow(sheet.FirstRowNum);
	var objectArray = objects as T[] ?? objects.ToArray();
	var type = MapHelper.GetConcreteType(objectArray);

	var columns = GetTrackedColumns(sheetName, type) ??
				   GetColumns(firstRow ?? PopulateFirstRow(sheet, null, type), type);

	if (firstRow == null)
	{
		PopulateFirstRow(sheet, columns, type);
	}

	// Injects custom action for headers.
	if (overwrite && HasHeader && _headerAction != null)
	{
		firstRow?.Cells.ForEach(c => _headerAction(c));
	}

	var rowIndex = overwrite
		? HasHeader ? sheet.FirstRowNum + 1 : sheet.FirstRowNum
		: sheet.GetRow(sheet.LastRowNum) != null ? sheet.LastRowNum + 1 : sheet.LastRowNum;

	MapHelper.EnsureDefaultFormats(columns, TypeFormats);

	foreach (var o in objectArray)
	{
		var row = sheet.GetRow(rowIndex);

		if (overwrite && row != null)
		{
			sheet.RemoveRow(row);
			row = sheet.CreateRow(rowIndex);
		}

		row = row ?? sheet.CreateRow(rowIndex);

		foreach (var column in columns)
		{
			var pi = column.Attribute.Property;
			var value = pi?.GetValue(o);
			var cell = row.GetCell(column.Attribute.Index, MissingCellPolicy.CREATE_NULL_AS_BLANK);

			column.CurrentValue = value;
			if (column.Attribute.TryPut == null || column.Attribute.TryPut(column, o))
			{
				SetCell(cell, column.CurrentValue, column, setStyle: overwrite);
			}
		}

		rowIndex++;
	}

	// Remove not used rows if any.
	while (overwrite && rowIndex <= sheet.LastRowNum)
	{
		var row = sheet.GetRow(rowIndex);
		if (row != null) sheet.RemoveRow(row);
		rowIndex++;
	}
}

error:

if (firstRow == null)
{
	PopulateFirstRow(sheet, columns, type);
}

right:

if (firstRow == null)
{
	firstRow = PopulateFirstRow(sheet, columns, type);
}

sorry for my broken english
column header name is: NF
cell format is text
cell value is 2017

public class Entity
{
...
public string NF { get; set; }
...
}
var set = mapper.Take<Entity>();
var isnull = set.All(x => x.NF == null);

isnull is true
but this works well

Mapper.Take<dynamic>()

  <NetStandardImplicitPackageVersion>1.6.0</NetStandardImplicitPackageVersion>

this line should be remove because it add unnecessary/obsolete dependancy (current netstandard.library version is 1.6.1 ) , it should be remove even you target netstandard

  <GenerateAssemblyConfigurationAttribute>false</GenerateAssemblyConfigurationAttribute>
  <GenerateAssemblyCompanyAttribute>false</GenerateAssemblyCompanyAttribute>
  <GenerateAssemblyProductAttribute>false</GenerateAssemblyProductAttribute>

those line should be remove too, the default value is true, and Properties/AssemblyInfo.cs will be automatic generate ( in obj folder ) , (it will be compile if you remove Properties/AssemblyInfo.cs )

When I save the file everything work well, but when i want to read it I see:

System.MissingMethodException : No parameterless constructor defined for this object.
Stack Trace:
   at System.RuntimeTypeHandle.CreateInstance(RuntimeType type, Boolean publicOnly, Boolean wrapExceptions, Boolean& canBeCached, RuntimeMethodHandleInternal& ctor)
   at System.RuntimeType.CreateInstanceSlow(Boolean publicOnly, Boolean wrapExceptions, Boolean skipCheckThis, Boolean fillCache)
   at Npoi.Mapper.Mapper.Take[T](ISheet sheet, Int32 maxErrorRows, Func`1 objectInitializer)+MoveNext()
   at System.Linq.Enumerable.Count[TSource](IEnumerable`1 source)

var mapper = SummaryMapping.Mapper(PathHelper.directory + "/statement.xlsx");                                                                                                                                                                                  
var test = mapper.Take<SummaryExelPersistence>("sheet0").Count();

What can cause the issue?

Proposed changes to .Map

Although the set-up is a bit different, I think some ideas from AutoMapper could be borrowed...

Currently the .Map function is a bit confusing to use.

First of all it will only be executed for properties where map via annotation failed (somehow feels as if it'd be conflicting with those annotations). And second: the given interface is hard to use (needs a lot of extra code).

Would consider (at least excel to object) following changes:

No1
have a ShouldMap function to replace the current .Map - first callback function (should map column)

• should provide original excel column
• should provide mapped destination property (if a match via attributes was found) - along with some match details

should return

• null (if no mapping should occure)
• or the destination property with a transformation function (optional)

No2
the Map function should change as follows (to work similarly as in case of AutoMapper):

• should be aware of the original excel column
• as of the destination column

...and should allow to do any kind of data manipulation from the excel value, before assigning it to the destination property (even value type changes by parsing and so)

Currently we need to match column names as find property names before we can do any extra changes... This would be maybe the first thing i'd consider to change (to pass the matched propertyname to the call)!

No3
AfterMap / BeforeMap could also help to do extra manipulation

I want to add a cell style to the header.
Therefore I had first to Put data, Ignore some fields, and then get Workbook, and then foreach and set the first row of the cell style.

Can you provide this functionality?
I believe there are many developers with this kind of demand.
Thank you so much.