I am validating the content for file import and I have an IsValid property for each line.
public class Header
{
public int LineNumber { get; set; }
public string Property1 { get; set; }
public string Property2 { get; set; }
public bool IsValid { get; set; }
}
public class Detail
{
public int LineNumber { get; set; }
public string Property1 { get; set; }
public string Property2 { get; set; }
public string Property3 { get; set; }
public bool IsValid { get; set; }
}
public class Trailer
{
public int LineNumber { get; set; }
public string Property1 { get; set; }
public bool IsValid { get; set; }
}
public class ImportFile
{
public Header Header { get; set; }
public List<Detail> Details { get; set; }
public Trailer Trailer { get; set; }
}
and my validators look somewhat like:
public class DetailValidator : AbstractValidator<Detail>
{
public DetailValidator()
{
RuleFor(d => d.Property1)
.Cascade(CascadeMode.Stop)
.NotEmpty()
.WithState(d => d.LineNumber)
.Length(3)
.WithState(d => d.LineNumber);
RuleFor(d => d.Property2)
.Cascade(CascadeMode.Stop)
.NotEmpty()
.WithState(d => d.LineNumber)
.MaximumLength(50)
.WithState(d => d.LineNumber);
...
}
}
public class ImportFileValidator : AbstractValidator<ImportFile>
{
public ImportFileValidator()
{
RuleFor(f => f.Header)
.SetValidator(new HeaderValidator());
RuleForEach(f => f.Details)
.SetValidator(new DetailsValidator());
...
}
}
After I call the validation, I wanted to set the IsValid property of each line of the file (be it header, detail or trailer) base from the result of the validation.
What is possible for now is, since I am using WithState to store the LineNumber, I can match the ValidationResult against the ImportFile instance to set each line's validity like below:
ImportFile file = // parsed file content
var result = new ImportFileValidator().Validate(file);
foreach (var detail in file.Details)
{
var error = result.Errors.FirstOrDefault(e =>
Convert.ToInt32(e.CustomState) == detail.LineNumber);
detail.IsValid = error == null;
}
And I have to check for the header and trailer as well.
Is there a way I can do this inside the validators? I am trying to explore the FluentValidation's documentation, but I can't seem to find what I needed there.
As I was exploring the available methods in FluentValidation, I saw OnFailure and OnAnyFailure methods. This methods might be a good help to what I needed to do, but the problem is they're obsolete as of 10.3.0 and will be removed on version 11. They're suggesting to use a custom validator instead.
The Header, Detail and Trailer Abstract Validators remain as is.
I created custom validator extensions for those 3.
Each extension methods creates an instance of the corresponding validator and executes it. I can make them generic for header, detail and trailer since they will do the same thing, set IsValid property to the validation result.
public static IRuleBuilderOptionsConditions<ImportFile, T> IsHeaderValid<T>(this IRuleBuilder<ImportFile, T> ruleBuilder)
where T : Header
{
return builder.Custom((header, context) =>
{
// Create the Header Abstract Validator Instance
var validator = new HeaderValidator();
var result = validator.Validate(Header);
header.IsValid = result.IsValid;
// Pass the errors to the context
result.Errors.ForEach(context.AddFailure);
}
}
I had to change the ImportFileValidator to call the custom validators, instead of using setvalidator.
The ImportFileValidator looks like this:
public class ImportFileValidator : AbstractValidator<ImportFile>
{
public ImportFileValidator()
{
RuleFor(f => f.Header)
.IsHeaderValid();
RuleForEach(f => f.Details)
.IsDetailValid();
...
}
}
This is pretty much how I was able to set the IsValid property without having to do the matching I initially did in the question.
Related
I want to able to select certain entity properties (columns from db) in the include statement of queryable object. My query looks like below but I m getting error Lambda expression used inside Include is not valid
var samuraiWithQuotesQueryable = _context.Samurais.AsQueryable()
.Include(s => s.Quotes.Select(x => new { x.Text }));
// additional filters followed by getting the list
var samuraiList = samuraiWithQuotesQueryable.ToList();
Samurai and Quote entities look like below
public class Samurai
{
public Samurai()
{
Quotes = new List<Quote>();
}
public int Id { get; set; }
public string Name { get; set; }
public List<Quote> Quotes { get; set; }
}
public class Quote
{
public int Id { get; set; }
public string Text { get; set; }
public Samurai Samurai { get; set; }
public int SamuraiId { get; set; }
}
Wondering if this is possible with the IQueryable object?
I found this great post by Chris Sainty: Creating Bespoke Input Components for Blazor from Scratch. It is exactly what I need, but not with string, but with uploaded files IBrowserFile. So I have adapted and extended the example for me. The customized component displays the new files and saves it in my model, but in the CSS the status unfortunately stays on class="modified invalid".
I must be missing a small detail here. What is it? Thanks in advance for any hints.
Here is my code reduced to the essentials.
Selection.razor
#page "/selection"
#inherits ParentComponent<SelectionTestModel>
<PageComponent #ref="Page" Model="Model" StatusCode="StatusCode" PageType="PageType.Touch">
<PageBody>
<EditForm Model="Model" OnValidSubmit="Save">
<DataAnnotationsValidator />
<DocumentComponent #ref="DocumentUpload" #bind-Documents="Model.Files" />
</EditForm>
</PageBody>
</PageComponent>
#code {
private DocumentComponent DocumentUpload;
}
SelectionTestModel.cs
public class SelectionTestModel
{
public int? KeyID { get; set; }
/* ... */
[System.ComponentModel.DisplayName("Document")]
[System.ComponentModel.DataAnnotations.Display(Name = "Document")]
[System.ComponentModel.DataAnnotations.Range(2, 2, ErrorMessage = "You have to bring exactly two files!")]
public List<DocumentModel> Files { get; set; } = new List<DocumentModel>();
}
DocumentModel
public class DocumentModel
{
public int? Id { get; set; }
public string Reference { get; set; }
public string Name { get; set; }
public long Size { get; set; }
public string ContentType { get; set; }
public string Content { get; set; } /*file as base64 string*/
}
DocumentComponent.razor
#using System.Linq.Expressions
<div class="dropzone rounded #_dropClass #_validClass">
<InputFile id="inputDrop" multiple
ondragover="event.preventDefault()"
ondragstart="event.dataTransfer.setData('', event.target.id)"
accept="#AllowedFileTypes"
OnChange="OnInputFileChange"
#ondragenter="HandleDragEnter"
#ondragleave="HandleDragLeave" />
#*...*#
</div>
#code {
[CascadingParameter] public EditContext EditContext { get; set; }
[Parameter] public List<DocumentModel> Documents { get; set; } = new List<DocumentModel>();
[Parameter] public EventCallback<List<DocumentModel>> DocumentsChanged { get; set; }
[Parameter] public Expression<Func<List<DocumentModel>>> DocumentsExpression { get; set; }
/*...*/
public List<string> AllowedFileTypes { get; set; } = new List<string> { ".pdf", /*...*/ };
private FieldIdentifier _fieldIdentifier;
private string _validClass => EditContext?.FieldCssClass(_fieldIdentifier) ?? null;
protected override void OnInitialized()
{
base.OnInitialized();
_fieldIdentifier = FieldIdentifier.Create(DocumentsExpression);
}
private async Task OnInputFileChange(InputFileChangeEventArgs e)
{
// validation: do we accept the file (content type, amount of files, size)
if (e.FileCount == 1) // keep it simple for this example
{
// read from IBrowserFile and return DocumentModel in memory only
Documents.Add(await SaveFile(e.File));
await DocumentsChanged.InvokeAsync(Documents);
EditContext?.NotifyFieldChanged(_fieldIdentifier);
}
}
/*...*/
}
How does it behave in the browser (Chrome)
After loading the page everything looks as expected.
After that I upload a single file. So I have one file and I expect two. The validation turns red and I get "modified invalid". So far everything is great.
Finally I drag another file into the component and get two files. I can also see this in the model. But unfortunately the class attribute "modified valid" is not set.
Thanks again for any advice
I dug too deep in the wrong direction and didn't see the obvious.
The problem is that there is an attribute set in the model that does not throw an error, but also cannot validate.
The Range attribute is not for lists and therefore the model could never validate. With an own attribute I could work around this.
SelectionTestModel.cs
[Library.Validation.Attribute.ListRange(2, 2)]
public List<DocumentModel> Files { get; set; } = new List<DocumentModel>();
ListRangeAttribute.cs
namespace Library.Validation.Attribute
{
public class ListRangeAttribute : ValidationAttribute
{
public int Minimum { get; set; }
public int Maximum { get; set; }
public ListRangeAttribute(int minimum = 0, int maximum = int.MaxValue)
{
Minimum = minimum > 0 ? minimum : 0;
Maximum = maximum;
}
public string GetErrorMessage(string displayName) { /* ... */ }
protected override ValidationResult IsValid(object value, ValidationContext validationContext)
{
var list = value as IList;
if (list == null)
{
throw new InvalidOperationException($"Attribute {nameof(ListRangeAttribute)} must be on a property of type {nameof(IList)}.");
}
if ((list?.Count ?? 0) < Minimum || (list?.Count ?? int.MaxValue) > Maximum)
{
return new ValidationResult(GetErrorMessage(validationContext.DisplayName), new[] { validationContext.MemberName });
}
return ValidationResult.Success;
}
}
}
I hope this post can help others.
Remaining: Now I am left with a new mystery.
Why does the validation text disappear after a save button click, which could not be saved due to an invalid state of the model!?
I'm using an external library to return data to me, the library has a lot of fields in it and goes quite deep with nested objects.
My class looks something like this;
public class Dto
{
public Dto(Val val)
{
Val = val;
}
[Key]
public int Id { get; set; }
public Val Value{ get; set; }
}
And when trying to save the changes to EF I get the error;
Cannot insert explicit value for identity column in table 'Value' when IDENTITY_INSERT is set to OFF.
When looking through the migrations it appears as if the Primary Key has been set on a few properties within the Value object. I've looked and looked but can't find a way to stop the migration automatically assigning primary keys. I've found stuff such as in the example above using the;
[Key]
attribute and adding bits and pieces into the OnModelCreating override but nothing has came up with a successful result.
Edit to give some more clarity instead of an example.
[DbContext(typeof(SummonerDtoContext))]
[Migration("20180314210242_Migration8")]
partial class Migration8
{
protected override void BuildTargetModel(ModelBuilder modelBuilder)
{
#pragma warning disable 612, 618
modelBuilder
.HasAnnotation("ProductVersion", "2.0.2-rtm-10011")
.HasAnnotation("SqlServer:ValueGenerationStrategy", SqlServerValueGenerationStrategy.IdentityColumn);
modelBuilder.Entity("LccWebAPI.Models.SummonerDto", b =>
{
b.Property<int>("Id")
.ValueGeneratedOnAdd();
b.Property<long?>("SummonerId");
b.HasKey("Id");
b.HasIndex("SummonerId");
b.ToTable("Summoners");
});
modelBuilder.Entity("RiotSharp.SummonerEndpoint.Summoner", b =>
{
b.Property<long>("Id")
.ValueGeneratedOnAdd();
b.Property<long>("AccountId");
b.Property<long>("Level");
b.Property<string>("Name");
b.Property<int>("ProfileIconId");
b.Property<int>("Region");
b.Property<DateTime>("RevisionDate");
b.HasKey("Id");
b.ToTable("Summoner");
});
modelBuilder.Entity("LccWebAPI.Models.SummonerDto", b =>
{
b.HasOne("RiotSharp.SummonerEndpoint.Summoner", "Summoner")
.WithMany()
.HasForeignKey("SummonerId");
});
#pragma warning restore 612, 618
}
}
}
and my actual data object I'm trying to store;
public class SummonerDto
{
public SummonerDto(Summoner summoner)
{
Summoner = summoner;
}
public int Id { get; set; }
public Summoner Summoner { get; set; }
}
and my context;
public class SummonerDtoContext : DbContext
{
public SummonerDtoContext(DbContextOptions<SummonerDtoContext> options)
: base(options)
{ }
public DbSet<SummonerDto> Summoners { get; set; }
}
I don't have access to modify the inside of the Summoner object itself to add annotations to ignore them as keys.
And the model structure of the Summoner object which I have no access to amend;
public class Summoner : SummonerBase
{
[JsonProperty("profileIconId")]
public int ProfileIconId { get; set; }
[JsonConverter(typeof(DateTimeConverterFromLong))]
[JsonProperty("revisionDate")]
public DateTime RevisionDate { get; set; }
[JsonProperty("summonerLevel")]
public long Level { get; set; }
}
And the base;
public class SummonerBase
{
public Region Region { get; set; }
[JsonProperty("id")]
public long Id { get; set; }
public long AccountId { get; set; }
[JsonProperty("name")]
public string Name { get; set; }
}
Edit : RESOLVED
After 3 hours of trying to figure this out I've finally managed to solve it, since I don't have access to annotate the models directly, I'd previously tried accessing the Ids through my SummonerDto model. Instead I accecssed them directly and it's worked.
protected override void OnModelCreating(ModelBuilder modelBuilder)
{
modelBuilder.Entity<SummonerDto>()
.HasKey(c => c.Id);
modelBuilder.Entity<Summoner>().Property(x => x.Id).ValueGeneratedNever();
modelBuilder.Entity<Summoner>().Property(x => x.ProfileIconId).ValueGeneratedNever();
modelBuilder.Entity<SummonerBase>().Property(x => x.Id).ValueGeneratedNever();
base.OnModelCreating(modelBuilder);
}
I have a document that contains an array of tags. I need to create a suggestion field corresponding to this tag field (to generate tag suggestions based on the values in the tag array). I am using NEST to interact with elastic search mostly. But I am not able to updated the suggestion property. The class used for the document contains following
Document structure:
public class SuggestField
{
public IEnumerable<string> Input { get; set; }
public string Output { get; set; }
public object Payload { get; set; }
public int? Weight { get; set; }
}
public class Source{
[ElasticProperty(Index = FieldIndexOption.NotAnalyzed)]
public string[] tags { get; set; }
public SuggestField[] tag_suggest { get; set; }
}
I add the mapping as follows:
var response = client.Map<Source>(m => m
.MapFromAttributes()
.Properties(p => p
.Completion(c => c
.Name(cp => cp.tag_suggest)
.Payloads()
)));
For updating tags, I use external scripts. I was hoping to change this same script to add changes to tag_suggest field also. But I tried the following but it is not working. Following is the script I tried:
if (ctx._source.tags.indexOf(newTag) < 0) {
ctx._source.tags[ctx._source.tags.length] = newTag;
ctx._source.tag_suggest[ctx._source.tag_suggest.length] = { input :newTag }
}
I would change type of tag_suggest property from SuggestField[] to SuggestField. You can store all tags in SuggestField.Input.
public class Source
{
[ElasticProperty(Index = FieldIndexOption.NotAnalyzed)]
public string[] tags { get; set; }
public SuggestField tag_suggest { get; set; }
}
Regarding your update script, after this change you can modify it to:
if (ctx._source.tags.indexOf(newTag) < 0) {
ctx._source.tags[ctx._source.tags.length] = newTag;
ctx._source.tag_suggest.input[ctx._source.tag_suggest.length] = newTag;
}
Hope it helps.
By injecting values into my domain object, I would keep the values of some properties.
Example:
Domain model
public class Person
{
public string Name { get; set; }
public Guid ID { get; set; }
public DateTime CreateAt { get; set; }
public string Notes { get; set; }
public IList<string> Tags { get; set; }
}
View Model
public class PersonViewMode
{
public string Name { get; set; }
public Guid ID { get; set; }
public DateTime CreateAt { get; set; }
public string Notes { get; set; }
public IList<string> Tags { get; set; }
public PersonViewMode() { ID = Guid.NewGuid(); } //You should use this value when it is the Target
}
Sample
var p = new Person
{
ID = Guid.NewGuid() //Should be ignored!
,
Name = "Riderman"
,
CreateAt = DateTime.Now
,
Notes = "teste de nota"
,
Tags = new[] {"Tag1", "Tag2", "Tag3"}
};
var pvm = new PersonViewMode();
pvm.InjectFrom(p); //Should use the ID value generated in the class constructor PersonViewMode
if you delete the set; from from the ViewModel's ID then it won't be set;
otherwise you could save the value of ID in a separate variable and put it back after injecting,
or you can create a custom valueinjection that would ignore "ID" or would receive a list of properties to ignore as a parameter
here's the example for a custom injection that receives a list of property names to ignore:
public class MyInj : ConventionInjection
{
private readonly string[] ignores = new string[] { };
public MyInj(params string[] ignores)
{
this.ignores = ignores;
}
protected override bool Match(ConventionInfo c)
{
if (ignores.Contains(c.SourceProp.Name)) return false;
return c.SourceProp.Name == c.TargetProp.Name && c.SourceProp.Type == c.TargetProp.Type;
}
}
and use it like this:
pvm.InjectFrom(new MyInj("ID"), p);
if you need to ignore more, you can do like this:
pvm.InjectFrom(new MyInj("ID","Prop2","Prop3"), p);