I am stuck trying to get Swashbuckle 5 to generate complete help pages for an ApiController with a Post request using multipart/form-data parameters. The help page for the action comes up in the browser, but there is not included information on the parameters passed in the form. I have created an operation filter and enabled it in SwaggerConfig, the web page that includes the URI parameters, return type and other info derived from XML comments shows in the browser help pages; however, nothing specified in the operation filter about the parameters is there, and the help page contains no information about the parameters.
I must be missing something. Are there any suggestion on what I may have missed?
Operation filter code:
public class AddFormDataUploadParamTypes : IOperationFilter
{
public void Apply(Operation operation, SchemaRegistry schemaRegistry, ApiDescription apiDescription) {
if (operation.operationId == "Documents_Upload")
{
operation.consumes.Add("multipart/form-data");
operation.parameters = new[]
{
new Parameter
{
name = "anotherid",
#in = "formData",
description = "Optional identifier associated with the document.",
required = false,
type = "string",
format = "uuid"
},
new Parameter
{
name = "documentid",
#in = "formData",
description = "The document identifier of the slot reserved for the document.",
required = false,
type = "string",
format = "uuid"
},
new Parameter
{
name = "documenttype",
#in = "formData",
description = "Specifies the kind of document being uploaded. This is not a file name extension.",
required = true,
type = "string"
},
new Parameter
{
name = "emailfrom",
#in = "formData",
description = "A optional email origination address used in association with the document if it is emailed to a receiver.",
required = false,
type = "string"
},
new Parameter
{
name = "emailsubject",
#in = "formData",
description = "An optional email subject line used in association with the document if it is emailed to a receiver.",
required = false,
type = "string"
},
new Parameter
{
name = "file",
#in = "formData",
description = "File to upload.",
required = true,
type = "file"
}
};
}
}
}
With Swashbuckle v5.0.0-rc4 methods listed above do not work. But by reading OpenApi spec I have managed to implement a working solution for uploading a single file. Other parameters can be easily added:
public class FileUploadOperationFilter : IOperationFilter
{
public void Apply(OpenApiOperation operation, OperationFilterContext context)
{
var isFileUploadOperation =
context.MethodInfo.CustomAttributes.Any(a => a.AttributeType == typeof(YourMarkerAttribute));
if (!isFileUploadOperation) return;
var uploadFileMediaType = new OpenApiMediaType()
{
Schema = new OpenApiSchema()
{
Type = "object",
Properties =
{
["uploadedFile"] = new OpenApiSchema()
{
Description = "Upload File",
Type = "file",
Format = "binary"
}
},
Required = new HashSet<string>()
{
"uploadedFile"
}
}
};
operation.RequestBody = new OpenApiRequestBody
{
Content =
{
["multipart/form-data"] = uploadFileMediaType
}
};
}
}
I presume you figured out what your problem was. I was able to use your posted code to make a perfect looking 'swagger ui' interface complete with the file [BROWSE...] input controls.
I only modified your code slightly so it is applied when it detects my preferred ValidateMimeMultipartContentFilter attribute stolen from Damien Bond. Thus, my slightly modified version of your class looks like this:
public class AddFormDataUploadParamTypes<T> : IOperationFilter
{
public void Apply(Operation operation, SchemaRegistry schemaRegistry, ApiDescription apiDescription)
{
var actFilters = apiDescription.ActionDescriptor.GetFilterPipeline();
var supportsDesiredFilter = actFilters.Select(f => f.Instance).OfType<T>().Any();
if (supportsDesiredFilter)
{
operation.consumes.Add("multipart/form-data");
operation.parameters = new[]
{
//other parameters omitted for brevity
new Parameter
{
name = "file",
#in = "formData",
description = "File to upload.",
required = true,
type = "file"
}
};
}
}
}
Here's my Swagger UI:
FWIW:
My NuGets
<package id="Swashbuckle" version="5.5.3" targetFramework="net461" />
<package id="Swashbuckle.Core" version="5.5.3" targetFramework="net461" />
Swagger Config Example
public class SwaggerConfig
{
public static void Register()
{
var thisAssembly = typeof(SwaggerConfig).Assembly;
GlobalConfiguration.Configuration
.EnableSwagger(c =>
{
c.Schemes(new[] { "https" });
// Use "SingleApiVersion" to describe a single version API. Swagger 2.0 includes an "Info" object to
// hold additional metadata for an API. Version and title are required but you can also provide
// additional fields by chaining methods off SingleApiVersion.
//
c.SingleApiVersion("v1", "MyCorp.WebApi.Tsl");
c.OperationFilter<MyCorp.Swashbuckle.AddFormDataUploadParamTypes<MyCorp.Attr.ValidateMimeMultipartContentFilter>>();
})
.EnableSwaggerUi(c =>
{
// If your API supports ApiKey, you can override the default values.
// "apiKeyIn" can either be "query" or "header"
//
//c.EnableApiKeySupport("apiKey", "header");
});
}
}
UPDATE March 2019
I don't have quick access to the original project above, but, here's an example API controller from a different project...
Controller signature:
[ValidateMimeMultipartContentFilter]
[SwaggerResponse(HttpStatusCode.OK, Description = "Returns JSON object filled with descriptive data about the image.")]
[SwaggerResponse(HttpStatusCode.NotFound, Description = "No appropriate equipment record found for this endpoint")]
[SwaggerResponse(HttpStatusCode.BadRequest, Description = "This request was fulfilled previously")]
public async Task<IHttpActionResult> PostSignatureImage(Guid key)
You'll note that there's no actual parameter representing my file in the signature, you can see below that I just spin up a MultipartFormDataStreamProvider to suck out the incoming POST'd form data.
Controller Body:
var signatureImage = await db.SignatureImages.Where(img => img.Id == key).FirstOrDefaultAsync();
if (signatureImage == null)
{
return NotFound();
}
if (!signatureImage.IsOpenForCapture)
{
ModelState.AddModelError("CaptureDateTime", $"This equipment has already been signed once on {signatureImage.CaptureDateTime}");
}
if (!ModelState.IsValid)
{
return BadRequest(ModelState);
}
string fileName = String.Empty;
string ServerUploadFolder = System.Web.Hosting.HostingEnvironment.MapPath("~/App_Data/");
DirectoryInfo di = new DirectoryInfo(ServerUploadFolder + key.ToString());
if (di.Exists == true)
ModelState.AddModelError("id", "It appears an upload for this item is either in progress or has already occurred.");
else
di.Create();
var fullPathToFinalFile = String.Empty;
var streamProvider = new MultipartFormDataStreamProvider(di.FullName);
await Request.Content.ReadAsMultipartAsync(streamProvider);
foreach (MultipartFileData fileData in streamProvider.FileData)
{
if (string.IsNullOrEmpty(fileData.Headers.ContentDisposition.FileName))
{
return StatusCode(HttpStatusCode.NotAcceptable);
}
fileName = cleanFileName(fileData.Headers.ContentDisposition.FileName);
fullPathToFinalFile = Path.Combine(di.FullName, fileName);
File.Move(fileData.LocalFileName, fullPathToFinalFile);
signatureImage.Image = File.ReadAllBytes(fullPathToFinalFile);
break;
}
signatureImage.FileName = streamProvider.FileData.Select(entry => cleanFileName(entry.Headers.ContentDisposition.FileName)).First();
signatureImage.FileLength = signatureImage.Image.LongLength;
signatureImage.IsOpenForCapture = false;
signatureImage.CaptureDateTime = DateTimeOffset.Now;
signatureImage.MimeType = streamProvider.FileData.Select(entry => entry.Headers.ContentType.MediaType).First();
db.Entry(signatureImage).State = EntityState.Modified;
try
{
await db.SaveChangesAsync();
//cleanup...
File.Delete(fullPathToFinalFile);
di.Delete();
}
catch (DbUpdateConcurrencyException)
{
if (!SignatureImageExists(key))
{
return NotFound();
}
else
{
throw;
}
}
char[] placeHolderImg = paperClipIcon_svg.ToCharArray();
signatureImage.Image = Convert.FromBase64CharArray(placeHolderImg, 0, placeHolderImg.Length);
return Ok(signatureImage);
Extending #bkwdesign very useful answer...
His/her code includes:
//other parameters omitted for brevity
You can actually pull all the parameter information (for the non-multi-part form parameters) from the parameters to the filter. Inside the check for supportsDesiredFilter, do the following:
if (operation.parameters.Count != apiDescription.ParameterDescriptions.Count)
{
throw new ApplicationException("Inconsistencies in parameters count");
}
operation.consumes.Add("multipart/form-data");
var parametersList = new List<Parameter>(apiDescription.ParameterDescriptions.Count + 1);
for (var i = 0; i < apiDescription.ParameterDescriptions.Count; ++i)
{
var schema = schemaRegistry.GetOrRegister(apiDescription.ParameterDescriptions[i].ParameterDescriptor.ParameterType);
parametersList.Add(new Parameter
{
name = apiDescription.ParameterDescriptions[i].Name,
#in = operation.parameters[i].#in,
description = operation.parameters[i].description,
required = !apiDescription.ParameterDescriptions[i].ParameterDescriptor.IsOptional,
type = apiDescription.ParameterDescriptions[i].ParameterDescriptor.ParameterType.FullName,
schema = schema,
});
}
parametersList.Add(new Parameter
{
name = "fileToUpload",
#in = "formData",
description = "File to upload.",
required = true,
type = "file"
});
operation.parameters = parametersList;
first it checks to make sure that the two arrays being passed in are consistent. Then it walks through the arrays to pull out the required info to put into the collection of Swashbuckle Parameters.
The hardest thing was to figure out that the types needed to be registered in the "schema" in order to have them show up in the Swagger UI. But, this works for me.
Everything else I did was consistent with #bkwdesign's post.
Related
I've written a trigger under the LeadConvert update event as follows:
trigger WebhookSenderTriggerLeadConvert on Lead (after update) {
if (Trigger.new.size() == 1) {
if (Trigger.old[0].isConverted == false && Trigger.new[0].isConverted == true) {
if (Trigger.new[0].ConvertedAccountId != null) {
String url = 'https://mydomain.io';
String content = WebhookSender.jsonContent(Trigger.new, Trigger.old);
WebhookSender.callout(url, content);
}
}
}
}
This works for me on a dev Salesforce, and in the payload I correctly receive:
{
"new":[
{
"attributes":{
"type":"Lead",
"url":"/services/data/v56.0/sobjects/Lead/B00000000000000000"
},
"Id":"B00000000000000000",
...(+30 more fields)
}
],
"old":[
{
"attributes":{
"type":"Lead",
"url":"/services/data/v56.0/sobjects/Lead/B00000000000000000"
},
"Id":"B00000000000000000",
...(+30 more fields)
}
],
"userId":"A00000000000000000"
}
However in another third party Salesforce account I get the following:
{
"new":[
{
"attributes":{
"type":"Lead",
"url":"/services/data/v56.0/sobjects/Lead/C00000000000000000"
},
...(9 more fields)
}
],
"old":[
{
"attributes":{
},
...(9 more fields)
}
],
"userId":"D00000000000000000"
}
I've obfuscated a lot of the fields here as a lot of it is sensitive, but what i'm unable to determine is what exactly causing a large portion of fields in the third-party Salesforce to not be there, including the Id field, where in the dev Salesforce everything is present.
Is there anything that may be doing this?
EDIT:
Posting WebhookSender, as it's been brought up in comments
public class WebhookSender {
public static String jsonContent(List<Object> triggerNew, List<Object> triggerOld) {
String newObjects = '[]';
if (triggerNew != null) {
newObjects = JSON.serialize(triggerNew);
}
String oldObjects = '[]';
if (triggerOld != null) {
oldObjects = JSON.serialize(triggerOld);
}
String userId = JSON.serialize(UserInfo.getUserId());
String content = '{"new": ' + newObjects + ', "old": ' + oldObjects + ', "userId": ' + userId + '}';
return content;
}
#future(callout=true)
public static void callout(String url, String content) {
Http h = new Http();
HttpRequest req = new HttpRequest();
req.setEndpoint(url);
req.setMethod('POST');
req.setHeader('Content-Type', 'application/json');
req.setHeader('Authorization', 'someKey');
req.setBody(content);
if (!Test.isRunningTest()) {h.send(req);}
}
public static Map<String, Object> ParseRequest(RestRequest req) {
String body = req.requestBody.toString();
Map<String, Object> data = (Map<String, Object>) JSON.deserializeUntyped(body);
return data;
}
}
Well, is the WebhookSender class identical in both? Does it dump all fields it received to JSON or does it contain some security-related code such as "stripInaccessible"? Maybe the fields are there but your Profile doesn't see them so strip... cuts them out?
Can it be that your dev org simply has 20+ custom fields in Lead table more than the other org?
Are the fields you're missing coming from managed package? They'd have namespace__FieldName__c format, with 4 underscores total. Maybe the package isn't installed. If you know the fields are there but your user doesn't have license for that managed package - it's possible they'll be hidden.
I'm making a .net Core WebApi using .Net Core 2.2. The API is ready but the failure message and response is where I'm stuck at.
Right now, I'm getting respose like below
json
{
"empId":1999,
"empName":"Conroy, Deborah",
"enrollmentStatus":true,
"primaryFingerprintScore":65,
"secondaryFingerprintScore":60,
"primaryFingerprint":null,
"secondaryFingerprint":null,
"primaryFingerprintType":null,
"secondaryFingerprintType":null}
}
I created a json formatter class and wrote the below code
public class SuperJsonOutputFormatter : JsonOutputFormatter
{
public SuperJsonOutputFormatter(
JsonSerializerSettings serializerSettings,
ArrayPool<char> charPool) : base(serializerSettings, charPool)
{
}
public override async Task WriteResponseBodyAsync(
OutputFormatterWriteContext context,
Encoding selectedEncoding)
{
if (context == null)
throw new ArgumentNullException(nameof(context));
if (selectedEncoding == null)
throw new ArgumentNullException(nameof(selectedEncoding));
using (TextWriter writer =
context.WriterFactory(
context.HttpContext.Response.Body,
selectedEncoding))
{
var rewrittenValue = new
{
resultCode = context.HttpContext.Response.StatusCode,
resultMessage =
((HttpStatusCode)context.HttpContext.Response.StatusCode)
.ToString(),
result = context.Object
};
this.WriteObject(writer, rewrittenValue);
await writer.FlushAsync();
}
}
I expect all the error codes to be sent as generic error messages like the JSON below.
FOR STATUS OKAY:
{
"status" : True,
"error" : null,
"data" : {
{
"empId":1999,
"empName":"Conroy, Deborah",
"enrollmentStatus":true,
"primaryFingerprintScore":65,
"secondaryFingerprintScore":60,
"primaryFingerprint":null,
"secondaryFingerprint":null,
"primaryFingerprintType":null,
"secondaryFingerprintType":null}
}
}
}
FOR OTHER STATUS LIKE 404, 500, 400, 204
{
"status" : False,
"error" : {
"error code" : 404,
"error description" : Not Found
},
"data" : null
}
I expect all the error codes to be sent as generic error messages like the JSON below
You're almost there. What you need to do is enabling your SuperJsonOutputFormatter.
A Little Change to Your Formatter
Firstly, your formatter didn't return a json with the same schema as you want. So I create a dummy class to hold the information for error code and error description:
public class ErrorDescription{
public ErrorDescription(HttpStatusCode statusCode)
{
this.Code = (int)statusCode;
this.Description = statusCode.ToString();
}
[JsonProperty("error code")]
public int Code {get;set;}
[JsonProperty("error description")]
public string Description {get;set;}
}
And change your WriteResponseBodyAsync() method as below:
...
using (TextWriter writer = context.WriterFactory(context.HttpContext.Response.Body, selectedEncoding)) {
var statusCode = context.HttpContext.Response.StatusCode;
var rewrittenValue = new {
status = IsSucceeded(statusCode),
error = IsSucceeded(statusCode) ? null : new ErrorDescription((HttpStatusCode)statusCode),
data = context.Object,
};
this.WriteObject(writer, rewrittenValue);
await writer.FlushAsync();
}
Here the IsSucceeded(statusCode) is a simple helper method that you can custom as you need:
private bool IsSucceeded(int statusCode){
// I don't think 204 indicates that's an error.
// However, you could comment out it if you like
if(statusCode >= 400 /* || statusCode==204 */ ) { return false; }
return true;
}
Enable your Formatter
Secondly, to enable your custom Formatter, you have two approaches: One way is to register it as an global Formatter, the other way is to enable it for particular Controller or Action. Personally, I believe the 2nd way is better. So I create a Action Filter to enable your formatter.
Here's an implementation of the Filter that enables your custom formatter dynamically:
public class SuperJsonOutputFormatterFilter : IAsyncActionFilter{
private readonly SuperJsonOutputFormatter _formatter;
// inject your SuperJsonOutputFormatter service
public SuperJsonOutputFormatterFilter(SuperJsonOutputFormatter formatter){
this._formatter = formatter;
}
// a helper method that provides an ObjectResult wrapper over the raw object
private ObjectResult WrapObjectResult(ActionExecutedContext context, object obj){
var wrapper = new ObjectResult(obj);
wrapper.Formatters.Add(this._formatter);
context.Result= wrapper;
return wrapper;
}
public async Task OnActionExecutionAsync(ActionExecutingContext context, ActionExecutionDelegate next)
{
ActionExecutedContext resultContext = await next();
// in case we get a 500
if(resultContext.Exception != null && ! resultContext.ExceptionHandled){
var ewrapper = this.WrapObjectResult(resultContext, new {});
ewrapper.StatusCode = (int) HttpStatusCode.InternalServerError;
resultContext.ExceptionHandled = true;
return;
}
else {
switch(resultContext.Result){
case BadRequestObjectResult b : // 400 with an object
var bwrapper=this.WrapObjectResult(resultContext,b.Value);
bwrapper.StatusCode = b.StatusCode;
break;
case NotFoundObjectResult n : // 404 with an object
var nwrapper=this.WrapObjectResult(resultContext,n.Value);
nwrapper.StatusCode = n.StatusCode;
break;
case ObjectResult o : // plain object
this.WrapObjectResult(resultContext,o.Value);
break;
case JsonResult j : // plain json
this.WrapObjectResult(resultContext,j.Value);
break;
case StatusCodeResult s: // other statusCodeResult(including NotFound,NoContent,...), you might want to custom this case
var swrapper = this.WrapObjectResult(resultContext, new {});
swrapper.StatusCode = s.StatusCode;
break;
}
}
}
}
And don't forget to register your formatter as a service :
services.AddScoped<SuperJsonOutputFormatter>();
Finally, when you want to enable your formatter, just add a [TypeFilter(typeof(SuperJsonOutputFormatterFilter))] annotation for the controller or action.
Demo
Let's create an action method for Test:
[TypeFilter(typeof(SuperJsonOutputFormatterFilter))]
public IActionResult Test(int status)
{
// test json result(200)
if(status == 200){ return Json(new { Id = 1, }); }
// test 400 object result
else if(status == 400){ return BadRequest( new {}); }
// test 404 object result
else if(status == 404){ return NotFound(new { Id = 1, }); }
// test exception
else if(status == 500){ throw new Exception("unexpected exception"); }
// test status code result
else if(status == 204){ return new StatusCodeResult(204); }
// test normal object result(200)
var raw = new ObjectResult(new XModel{
empId=1999,
empName = "Conroy, Deborah",
enrollmentStatus=true,
primaryFingerprintScore=65,
secondaryFingerprintScore=60,
primaryFingerprint = null,
secondaryFingerprint= null,
primaryFingerprintType=null,
secondaryFingerprintType=null
});
return raw;
}
Screenshot:
I have an api end point that takes a custom header.
This header is a object that looks like this:
User: {"UserId":"someguid"}
If I have the type as a parameter in a api method I can do as follows:
public class AddFileParamTypes : IOperationFilter
{
public void Apply(Operation operation, SchemaRegistry schemaRegistry, ApiDescription apiDescription)
{
if (operation.operationId == "FileDelivery_Post")
{
operation.consumes.Add("multipart/form-data");
operation.parameters.Add(new Parameter
{
name = "file",
required = true,
type = "file",
#in = "formData"
});
operation.parameters.Add(new Parameter
{
name = "User",
required = true,
schema = new Schema() { #ref = "#/definitions/User" },
#in = "header"
});
}
}
}
But the User type will not be a parameter for a method, so how do I add the definition to swashbuckle?
Found out, it was simple but lacking in documentation:
public class AddFileParamTypes : IOperationFilter
{
public void Apply(Operation operation, SchemaRegistry schemaRegistry, ApiDescription apiDescription)
{
if (operation.operationId == "FileDelivery_Post")
{
operation.consumes.Add("multipart/form-data");
operation.parameters.Add(new Parameter
{
name = "file",
required = true,
type = "file",
#in = "formData"
});
operation.parameters.Add(new Parameter
{
name = "User",
required = true,
schema = schemaRegistry.GetOrRegister(typeof(User)),
#in = "header"
});
}
}
}
I have my web api and I added the web api help pages to auto-generate my documentation. It's working great for methods where my parameters are listed out, but I have a method like this:
public SessionResult PostLogin(CreateSessionCommand request)
And, on my help page, it is only listing the command parameter in the properties section. However, in the sample request section, it lists out all of the properties of my CreateSessionCommand class.
Parameters
Name | Description | Additional information
request | No documentation available. | Define this parameter in the request body.
I would like it instead to list all of the properties in my CreateSessionCommand class. Is there an easy way to do this?
So, I managed to devise a workaround for this problem, in case anyone is interested.
In HelpPageConfigurationExtensions.cs I added the following extension method:
public static void AlterApiDescription(this ApiDescription apiDescription, HttpConfiguration config)
{
var docProvider = config.Services.GetDocumentationProvider();
var addParams = new List<ApiParameterDescription>();
var removeParams = new List<ApiParameterDescription>();
foreach (var param in apiDescription.ParameterDescriptions)
{
var type = param.ParameterDescriptor.ParameterType;
//string is some special case that is not a primitive type
//also, compare by full name because the type returned does not seem to match the types generated by typeof
bool isPrimitive = type.IsPrimitive || String.Compare(type.FullName, typeof(string).FullName) == 0;
if (!isPrimitive)
{
var properties = from p in param.ParameterDescriptor.ParameterType.GetProperties()
let s = p.SetMethod
where s.IsPublic
select p;
foreach (var property in properties)
{
var documentation = docProvider.GetDocumentation(new System.Web.Http.Controllers.ReflectedHttpParameterDescriptor()
{
ActionDescriptor = param.ParameterDescriptor.ActionDescriptor,
ParameterInfo = new CustomParameterInfo(property)
});
addParams.Add(new ApiParameterDescription()
{
Documentation = documentation,
Name = property.Name,
Source = ApiParameterSource.FromBody,
ParameterDescriptor = param.ParameterDescriptor
});
}
//since this is a complex type, select it to be removed from the api description
removeParams.Add(param);
}
}
//add in our new items
foreach (var item in addParams)
{
apiDescription.ParameterDescriptions.Add(item);
}
//remove the complex types
foreach (var item in removeParams)
{
apiDescription.ParameterDescriptions.Remove(item);
}
}
And here is the Parameter info instanced class I use
internal class CustomParameterInfo : ParameterInfo
{
public CustomParameterInfo(PropertyInfo prop)
{
base.NameImpl = prop.Name;
}
}
Then, we call the extension in another method inside the extensions class
public static HelpPageApiModel GetHelpPageApiModel(this HttpConfiguration config, string apiDescriptionId)
{
object model;
string modelId = ApiModelPrefix + apiDescriptionId;
if (!config.Properties.TryGetValue(modelId, out model))
{
Collection<ApiDescription> apiDescriptions = config.Services.GetApiExplorer().ApiDescriptions;
ApiDescription apiDescription = apiDescriptions.FirstOrDefault(api => String.Equals(api.GetFriendlyId(), apiDescriptionId, StringComparison.OrdinalIgnoreCase));
if (apiDescription != null)
{
apiDescription.AlterApiDescription(config);
HelpPageSampleGenerator sampleGenerator = config.GetHelpPageSampleGenerator();
model = GenerateApiModel(apiDescription, sampleGenerator);
config.Properties.TryAdd(modelId, model);
}
}
return (HelpPageApiModel)model;
}
The comments that are used for this must be added to the controller method and not the properties of the class object. This might be because my object is part of an outside library
this should go as an addition to #Josh answer. If you want not only to list properties from the model class, but also include documentation for each property, Areas/HelpPage/XmlDocumentationProvider.cs file should be modified as follows:
public virtual string GetDocumentation(HttpParameterDescriptor parameterDescriptor)
{
ReflectedHttpParameterDescriptor reflectedParameterDescriptor = parameterDescriptor as ReflectedHttpParameterDescriptor;
if (reflectedParameterDescriptor != null)
{
if (reflectedParameterDescriptor.ParameterInfo is CustomParameterInfo)
{
const string PropertyExpression = "/doc/members/member[#name='P:{0}']";
var pi = (CustomParameterInfo) reflectedParameterDescriptor.ParameterInfo;
string selectExpression = String.Format(CultureInfo.InvariantCulture, PropertyExpression, pi.Prop.DeclaringType.FullName + "." + pi.Prop.Name);
XPathNavigator methodNode = _documentNavigator.SelectSingleNode(selectExpression);
if (methodNode != null)
{
return methodNode.Value.Trim();
}
}
else
{
XPathNavigator methodNode = GetMethodNode(reflectedParameterDescriptor.ActionDescriptor);
if (methodNode != null)
{
string parameterName = reflectedParameterDescriptor.ParameterInfo.Name;
XPathNavigator parameterNode = methodNode.SelectSingleNode(String.Format(CultureInfo.InvariantCulture, ParameterExpression, parameterName));
if (parameterNode != null)
{
return parameterNode.Value.Trim();
}
}
}
}
return null;
}
and CustomParameterInfo class should keep property info as well:
internal class CustomParameterInfo : ParameterInfo
{
public PropertyInfo Prop { get; private set; }
public CustomParameterInfo(PropertyInfo prop)
{
Prop = prop;
base.NameImpl = prop.Name;
}
}
This is currently not supported out of the box. Following bug is kind of related to that:
http://aspnetwebstack.codeplex.com/workitem/877
I modified the "Read" operation on my Windows Azure Mobile Services Preview table (named "Item") as follows:
Javascript:
function read(query, user, request)
{
var howRead;
if(howRead == "unique")
{
var sqlUnique = "SELECT DISTINCT ? FROM Item WHERE qProjectCode = ?";
mssql.query(sqlUnique)
request.execute();
}
else if (howRead == "column")
{
var sqlColumn = "SELECT ? FROM Item WHERE qProjectCode = ?";
mssql.query(sqlColumn)
request.execute();
}
else if (howRead == "all")
{
var sqlAll = "SELECT * FROM Item WHERE qProjectCode = ?";
mssql.query(sqlAll)
request.execute();
}
}
This simply species when I want a unique list of a single column's values returned, all items in a single column, or all columns, respectively, all while limiting the read to those records with a given project code.
Right now, this works in C#, but scans the entire table (with other project codes) and always returns all columns. This is inherently inefficient.
c#
var client = new MobileServiceClient("[https path", "[key]");
var table = client.GetTable<Item>();
var query1 = table.Where(w => w.QProjectCode == qgv.projCode && w.QRecord == (int)lbRecord.Items[uStartRecordIndex]);
var query1Enum = await query1.ToEnumerableAsync();
foreach (var i in query1Enum)
{
// process data
}
How do I alter the c# code to deal with the Javascript code? Feel free to critique the overall approach, since I am not a great programmer and can always use advice!
Thanks
A few things:
In your server code, the mssql calls are not doing anything (useful). If you want to get their results, you need to pass a callback (the call is asynchronous) to it.
Most of your scenarios can be accomplished at the client side. The only for which you'll need server code is the one with the DISTINCT modifier.
For that scenario, you'll need to pass a custom parameter to the server script. You can use the WithParameters method in the MobileServiceTableQuery<T> object to define parameters to pass to the service.
Assuming this data class:
public class Item
{
public int Id { get; set; }
public string Name { get; set; }
public string Description { get; set; }
public string Other { get; set; }
public string ProjectCode { get; set; }
}
The code below can be used to accomplish the scenarios 2 and 3 at the client side only (no script needed at the server side). The other one will need some script, which I'll cover later.
Task<IEnumerable<string>> ReadingByColumn(IMobileServiceTable<Item> table, string projectCode)
{
return table
.Where(i => i.ProjectCode == projectCode)
.Select(i => i.Name)
.ToEnumerableAsync();
}
Task<IEnumerable<Item>> ReadingAll(IMobileServiceTable<Item> table, string projectCode)
{
return table.Where(i => i.ProjectCode == projectCode).ToEnumerableAsync();
}
Task<IEnumerable<string>> ReadingByColumnUnique(IMobileServiceTable<Item> table, string projectCode)
{
var dict = new Dictionary<string, string>
{
{ "howRead", "unique" },
{ "projectCode", projectCode },
{ "column", "Name" },
};
return table
.Select(i => i.Name)
.WithParameters(dict)
.ToEnumerableAsync();
}
Now, to support the last method (which takes the parameters, we'll need to do this on the server script:
function read(query, user, request)
{
var howRead = request.parameters.howRead;
if (howRead) {
if (howRead === 'unique') {
var column = request.parameters.column; // WARNING: CHECK FOR SQL INJECTION HERE!!! DO NOT USE THIS IN PRODUCTION!!!
var sqlUnique = 'SELECT DISTINCT ' + column + ' FROM Item WHERE ProjectCode = ?';
mssql.query(sqlUnique, [request.parameters.projectCode], {
success: function(distinctColumns) {
var results = distinctColumns.map(function(item) {
var result = [];
result[column] = item; // mapping to the object shape
return result;
});
request.respond(statusCodes.OK, results);
}
});
} else {
request.respond(statusCodes.BAD_REQUEST, {error: 'Script does not support option ' + howRead});
}
} else {
// no server-side action needed
request.execute();
}
}