I am learning M (Power Query Language). I'd like to use M to parse JSON from REST APIs. For example one can use the Stack Overflow REST API. I can see how to drill down into a simple JSON string using say
let
Source = Json.Document("{ ""glossary"": { ""title"": ""example glossary"", ""GlossDiv"": { ""title"": ""S"", ""GlossList"": { ""GlossEntry"":
{ ""ID"": ""SGML"", ""SortAs"": ""SGML"", ""GlossTerm"": ""Standard Generalized Markup Language"", ""Acronym"": ""SGML"",
""Abbrev"": ""ISO 8879:1986"", ""GlossDef"": { ""para"": ""A meta-markup language, used to create markup languages such as DocBook."",
""GlossSeeAlso"": [""GML"", ""XML""] }, ""GlossSee"": ""markup"" } } } } }"),
glossary = Source[glossary],
GlossDiv = glossary[GlossDiv],
GlossList = GlossDiv[GlossList],
GlossEntry = GlossList[GlossEntry],
ConvertedToTable = Record.ToTable(GlossEntry)
in
ConvertedToTable
But what happens when I have a list from which I want to drill in and fetch a subproperty and then I want to return all of those like a SQL UNION query. Actually it is more of a For Each type query.
So here is my non-working query that does not do a union but unfortunately glues the second record onto the side
let
Source = "{""items"":["
{""tags"":[""vba"",""permissions""],""owner"":
{""reputation"":49,""user_id"":9073241,""user_type"":""registered"",""accept_rate"":86,""display_name"":""Kam""},
""is_answered"":false,""view_count"":4,""answer_count"":0,""score"":0,""question_id"":48229549},
{""tags"":[""excel"",""vba"",""excel-vba""],""owner"":
{""reputation"":18,""user_id"":9057704,""user_type"":""registered"",""accept_rate"":29,""display_name"":""Gregory""},
""is_answered"":false,""view_count"":6,""answer_count"":0,""score"":0,""question_id"":48229590}
]}",
#"Parsed JSON" = Json.Document(Source),
items = #"Parsed JSON"[items],
item0 = items{0},
owner0 = item0[owner],
item1 = items{1},
owner1 = item1[owner],
#"Converted to Table" = Table.Combine( {Record.ToTable(owner0), Record.ToTable(owner1) })
in
#"Converted to Table"
What I am really aiming for is this output but not limited to 2 records, but all the records from the list. (The above sample source has been simplified from this REST API StackOverflow questions tagged VBA)
reputation user_id user_type accept_rate display_name
49 9073241 registered 86 Kam
18 9057704 registered 29 Gregory
I think you want to pivot your tables before you try to combine them. Try this query, for example.
let
Source1 = Json.Document("{""tags"":[""vba"",""permissions""],""owner"":
{""reputation"":49,""user_id"":9073241,""user_type"":""registered"",""accept_rate"":86,""display_name"":""Kam""},
""is_answered"":false,""view_count"":4,""answer_count"":0,""score"":0,""question_id"":48229549}"),
Owner1 = Table.Pivot(Record.ToTable(Source1[owner]), List.Distinct(Record.ToTable(Source1[owner])[Name]), "Name", "Value"),
Source2 = Json.Document("{""tags"":[""excel"",""vba"",""excel-vba""],""owner"":
{""reputation"":18,""user_id"":9057704,""user_type"":""registered"",""accept_rate"":29,""display_name"":""Gregory""},
""is_answered"":false,""view_count"":6,""answer_count"":0,""score"":0,""question_id"":48229590}"),
Owner2 = Table.Pivot(Record.ToTable(Source2[owner]), List.Distinct(Record.ToTable(Source2[owner])[Name]), "Name", "Value"),
#"Appended Query" = Table.Combine({Owner1, Owner2})
in
#"Appended Query"
If you just want to expand all of the owners, try a query more like this:
let
Source = "{""items"":[{""tags"":[""vba"",""permissions""],""owner"":{""reputation"":49,""user_id"":9073241,""user_type"":""registered"",""accept_rate"":86,""display_name"":""Kam""},""is_answered"":false,""view_count"":4,""answer_count"":0,""score"":0,""question_id"":48229549},{""tags"":[""excel"",""vba"",""excel-vba""],""owner"":{""reputation"":18,""user_id"":9057704,""user_type"":""registered"",""accept_rate"":29,""display_name"":""Gregory""},""is_answered"":false,""view_count"":6,""answer_count"":0,""score"":0,""question_id"":48229590}]}",
#"Parsed JSON" = Json.Document(Source),
items = #"Parsed JSON"[items],
#"Converted to Table" = Table.FromList(items, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"Expanded Column1" = Table.ExpandRecordColumn(#"Converted to Table", "Column1", {"owner"}, {"owner"}),
#"Expanded owner" = Table.ExpandRecordColumn(#"Expanded Column1", "owner", {"reputation", "user_id", "user_type", "display_name"}, {"reputation", "user_id", "user_type", "display_name"})
in
#"Expanded owner"
Related
I have to get data from 2 API (one is input and another is hardcoded).
My scenario is I would input my 1st key (as Authentication key) to run 1st API and inside I have a column that calls the 2nd API (which I hardcoded apikey) but the system still alerts me as wrong credential. If I run separately it will run normally. I've done a similar case but in that only used 1 APIkey. I don't know how to resolve it.
ExpandData = (dataList as list) as table =>
let
FromList = Table.FromList(dataList, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"Expanded Column1" = Table.ExpandRecordColumn(FromList, "Column1", {"search-results"}, {"search-results"}),
#"Expanded search-results" = Table.ExpandRecordColumn(#"Expanded Column1", "search-results", {"entry"}, {"entry"}),
#"Expanded entry" = Table.ExpandListColumn(#"Expanded search-results", "entry"),
#"Expanded entry1" = Table.ExpandRecordColumn(#"Expanded entry", "entry", {"dc:identifier", "prism:doi"}, {"identifier", "doi"}),
#"Reordered Columns" = Table.ReorderColumns(#"Expanded entry1",{"doi", "identifier"}),
#"Add Column" = Table.AddColumn(#"Reordered Columns", "Detail", each Text.Range(Text.From([identifier]),10)),
#"Add Column2" = Table.AddColumn(#"Add Column", "Detail2", each ExpandDataPublish(Text.From([Detail])))
in
#"Add Column2";
ExpandDataPublish = (data as text) as table =>
let
Source = Json.Document(Web.Contents("https://******" & data,
[Headers=[#"Accept" = "application/json;odata.metadata=minimal",
#"X-ELS-APIKey"="7f59af901d2d86f78a1fd60c1bf9426a"]])),
#"Converted to Table" = Table.FromRecords({Source})
in
#"Converted to Table";
It runs normally if not call 2nd API
But fail if call
I am trying to get financial data from Financial Modeling Prep's API into an excel spreadsheet. I am beginning to think that Power Query just does not do what I am looking for. I want to have one column with a static list of stock symbols (DAL, GOOG, AAL etc) and populate each row with financial data from various api calls such as the Net Income field from https://financialmodelingprep.com/api/v3/financials/income-statement/DAL and the current stock price from https://financialmodelingprep.com/api/v3/stock/real-time-price/DAL
What exactly have you tried? It's very simple to extract data from the first link you gave with the M Code below (all UI based, nothing advanced about that at all). Converting that into a function to go to the relevant URL for each code and do the same transformation is also trivial
let
Source = Json.Document(Web.Contents("https://financialmodelingprep.com/api/v3/financials/income-statement/DAL ")),
financials = Source[financials],
#"Converted to Table" = Table.FromList(financials, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"Expanded Column1" = Table.ExpandRecordColumn(#"Converted to Table", "Column1", {"date", "Net Income"}, {"date", "Net Income"}),
#"Changed Type" = Table.TransformColumnTypes(#"Expanded Column1",{{"Net Income", type number}, {"date", type date}})
in
#"Changed Type"
Here is my solution in Python:
Create parameters:
company = "NVDA"
years = 5
Add Key:
api_key = 'YOUR_KEY'
Request:
r = requests.get(f'https://financialmodelingprep.com/api/v3/income-statement/{company}?limit={years}&apikey={api_key}')
data = r.json()
data
Extract data
date = []
symbol = []
revenue = []
costOfRevenue = []
grossProfit = []
for finance in data:
date.append(finance["date"])
symbol.append(finance["symbol"])
revenue.append(finance["revenue"])
costOfRevenue.append(finance["costOfRevenue"])
grossProfit.append(finance["grossProfit"])
ncome_nvda_dict = {
"Date" : date,
"Ticket": symbol,
"Revenue" : revenue,
"CostOfRevenue" : costOfRevenue,
"grossProfit" : grossProfit,
From Object To Pands
income_nvda_df = pd.DataFrame(income_nvda_dict, columns = ['Date', 'Ticket', 'Revenue', 'CostOfRevenue', 'grossProfit'])
I am using the following code in my own powerBI Data connector to get some date from a json document:
{
"Customers": [
{
"CustomerId": "8cd72f16-8d7b-48b0-90d9-71df011502c8",
"CustomerTitle": "Test Customer",
}
}
Code:
GetCustomerTable = (url as text) as table =>
let
source = Test.Feed(url & "/overview"),
value = source[Customers],
toTable = Table.FromList(value, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"expandColumn" = Table.ExpandRecordColumn(toTable, "Column1", {"CustomerId", "CustomerTitle"}, {"CustomerId", "CustomerTitle"}),
#"ChangedType" = Table.TransformColumnTypes(#"expandColumn",{{"CustomerTitle", type text}, {"CustomerId", type text})
in
ChangedType;
The column "CustomerId" referes to another url where the actual data about the customer is available in json format:
URL: /Details/8cd72f16-8d7b-48b0-90d9-71df011502c8
{
"Category": "B",
}
What is the best approach to use data from another url with the ExpandRecordColumn function?
So what you need is another custom function to obtain the customer details with each CustomerId, as one of the step:
GetCustomerDetails = (url as text, customer_id as text) =>
let
Source = Json.Document(Web.Contents(url & "/Details/" & customer_id)),
#"Converted to Table" = Table.FromList(Source, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"Expanded Column" = Table.ExpandRecordColumn(#"Converted to Table", "Column1", {"Category"}, {"Category"})
in
#"Expanded Column"
And then you can invoke this function in your original code by passing url and the CustomerId column:
GetCustomerTable = (url as text) as table =>
let
source = Test.Feed(url & "/overview"),
value = source[Customers],
toTable = Table.FromList(value, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
"expandColumn" = Table.ExpandRecordColumn(toTable, "Column1", {"CustomerId", "CustomerTitle"}, {"CustomerId", "CustomerTitle"}),
"ChangedType" = Table.TransformColumnTypes(#"expandColumn",{{"CustomerTitle", type text}, {"CustomerId", type text}),
#"Invoked Custom Function" = Table.AddColumn(#"ChangedType", "GetCustomerDetails", each GetCustomerDetails("http://testing.com/", [CustomerId]))
in
#"Invoked Custom Function"
You may need to make some adjustments to the code, depending on how it exactly looks like, but I hope you get the point.
I'm connecting to Azure resources via API at resources.azure.com, from there I'm taking the API for Microsoft.Compute and importing all the VM details into Power BI via JSON.
The import works fine, however with some situations of the data there is case discrepancy. For example, when working with the tags value, some people have typed the same word but in different case, such as;
"tags": {
"Project": "DT",
"SLStandard": "Yes"
compared to
"tags": {
"project": "DT",
"SlStandard": "Yes"
When expanding the columns out in Power BI it will consider the items listed above as two different value.
Ideally I would like to have the JSON imported and the 'case' ignored, or perhaps mark all incoming as either upper or lower case.
I have read the two links below, but I'm new to Power BI and I'm unsure how to implement it, or even if it is what I need.
Case sensitivity in Power BI
and
Power BI changing text case automatically
and
http://www.thebiccountant.com/2016/10/27/tame-case-sensitivity-power-query-powerbi/
Here is my Advanced Editor code:
let
iterations = 10,
url =
"https://management.azure.com/subscriptions/< subscription id >/providers/Microsoft.Compute/virtualMachines?api-version=2017-12-01",
FnGetOnePage =
(url) as record =>
let
Source = Json.Document(Web.Contents(url)),
data = try Source[value] otherwise null,
next = try Source[nextLink] otherwise null,
res = [Data=data, Next=next]
in
res,
GeneratedList =
List.Generate(
()=>[i=0, res = FnGetOnePage(url)],
each [i]<iterations and [res][Data]<>null,
each [i=[i]+1, res = FnGetOnePage([res][Next])],
each [res][Data]),
#"Converted to Table" = Table.FromList(GeneratedList, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"Expanded Column1" = Table.ExpandListColumn(#"Converted to Table", "Column1"),
#"Expanded Column2" = Table.ExpandRecordColumn(#"Expanded Column1", "Column1", {"tags"}, {"Column1.tags"}),
#"Expanded Column1.tags" = Table.ExpandRecordColumn(#"Expanded Column2", "Column1.tags", {"Project", "project", "SLStandard", "sLStandard", "BIOffline", "bIStandard", "AutomationBI", "biStandard", "BIStandard", "asdf-U001", "TestVM"}, {"Column1.tags.Project.1", "Column1.tags.project", "Column1.tags.SLStandard.1", "Column1.tags.sLStandard", "Column1.tags.BIOffline", "Column1.tags.bIStandard.1", "Column1.tags.AutomationBI", "Column1.tags.biStandard.2", "Column1.tags.BIStandard", "Column1.tags.asdf-U001", "Column1.tags.TestVM"})
in
#"Expanded Column1.tags"
If you're wondering about why my query is so long for import, then check out my previous post here: Power BI - Call Azure API with nextLink (next page)
Any help would be greatly appreciated.
I'm struggling with the same issue today. One workaround, although not elegant, is to lowercase or UPPERCASE the JSON upstream.
Worked for me as a temporary solution.
Hope it helps,
Chris
It seems that your issue is coming directly from the data source and the discrepancy between the field names at the data source level.
Here's a sample code on how you can force/make sure that they all have the same casing:
let
Source = {[A=24,b=53], [a=43,B=43], [a=3,b=3]},
Custom1 = List.Transform(Source, (_)=> Record.RenameFields(_, List.Zip({Record.FieldNames(_), List.Transform(Record.FieldNames(_), Text.Lower)})))
I've had this answered on another forum and tried it, and it does work.
https://community.powerbi.com/t5/Desktop/Power-BI-case-insensitive-with-JSON/m-p/360134
"You can rename your record fields (by applying Text.Proper for example) before expanding (in the last step):"
Table.TransformColumns(#"Expanded Column2", {{"tags", each Record.RenameFields(_, List.Zip({Record.FieldNames(_), List.Transform(Record.FieldNames(_), (name)=> Text.Proper(name))}))}}),
An the final output looks like this:
let
iterations = 10,
url =
"https://management.azure.com/subscriptions/< subscription >/providers/Microsoft.Compute/virtualMachines?api-version=2017-12-01",
FnGetOnePage =
(url) as record =>
let
Source = Json.Document(Web.Contents(url)),
data = try Source[value] otherwise null,
next = try Source[nextLink] otherwise null,
res = [Data=data, Next=next]
in
res,
GeneratedList =
List.Generate(
()=>[i=0, res = FnGetOnePage(url)],
each [i]<iterations and [res][Data]<>null,
each [i=[i]+1, res = FnGetOnePage([res][Next])],
each [res][Data]),
#"Converted to Table" = Table.FromList(GeneratedList, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
#"Expanded Column1" = Table.ExpandListColumn(#"Converted to Table", "Column1"),
#"Expanded Column2" = Table.ExpandRecordColumn(#"Expanded Column1", "Column1", {"properties", "tags"}, {"properties", "tags"}),
#"Expanded properties" = Table.TransformColumns(#"Expanded Column2", {{"tags", each Record.RenameFields(_, List.Zip({Record.FieldNames(_), List.Transform(Record.FieldNames(_), (name)=> Text.Proper(name))}))}}),
#"Expanded properties1" = Table.ExpandRecordColumn(#"Expanded properties", "properties", {"vmId"}, {"properties.vmId"}),
#"Expanded tags" = Table.ExpandRecordColumn(#"Expanded properties1", "tags", {"Project", "Slstandard", "Bioffline", "Bistandard", "Automationbi", "asdf-U001", "Testvm"}, {"tags.Project", "tags.Slstandard", "tags.Bioffline", "tags.Bistandard", "tags.Automationbi", "tags.asdf-U001", "tags.Testvm"})
in
#"Expanded tags"
Hope this helps other people!
I have developed a microservice using Spring Boot. The documentation for the REST API is made with Swagger. Some REST resources make use of Spring concepts to provide pagination for free. Below is an example:
#RequestMapping(value = "/buckets", method = GET)
public PagedResources list(Pageable pageable, PagedResourcesAssembler assembler) {
return bucketService.listBuckets(pageable, assembler);
}
If I open the Swagger page, the following form is available for the resource:
The issue I have is that the pageable parameter is detected with content-type application/json and I don't know how to pass a value to change the page size for example. All values seem to be ignored.
Is it possible to pass the query parameters as JSON object? or is it possible to configure Swagger to generate independent query parameter fields for getters contained by the Pageable interface?
Please note that I am using Springfox with Gradle:
compile 'io.springfox:springfox-spring-web:2.3.1'
compile 'io.springfox:springfox-swagger2:2.3.1'
compile 'io.springfox:springfox-swagger-ui:2.3.1'
This is a known issue with Spring-Fox. See Issue #755. Based on zdila's comment 2 at this time alternative is to add #ApiImplicitParams which is not ideal but it does work.
#ApiImplicitParams({
#ApiImplicitParam(name = "page", dataType = "integer", paramType = "query",
value = "Results page you want to retrieve (0..N)"),
#ApiImplicitParam(name = "size", dataType = "integer", paramType = "query",
value = "Number of records per page."),
#ApiImplicitParam(name = "sort", allowMultiple = true, dataType = "string", paramType = "query",
value = "Sorting criteria in the format: property(,asc|desc). " +
"Default sort order is ascending. " +
"Multiple sort criteria are supported.")
})
[
1 https://github.com/springfox/springfox/issues/755
2 https://github.com/springfox/springfox/issues/755#issuecomment-135059871
Building on Vineet Bhatia's answer, you can wrap the solution up in a custom annotation for reusability:
#Target({ ElementType.METHOD, ElementType.ANNOTATION_TYPE, ElementType.TYPE })
#Retention(RetentionPolicy.RUNTIME)
#ApiImplicitParams({
#ApiImplicitParam(name = "page", dataType = "int", paramType = "query", value = "Results page you want to retrieve (0..N)"),
#ApiImplicitParam(name = "size", dataType = "int", paramType = "query", value = "Number of records per page."),
#ApiImplicitParam(name = "sort", allowMultiple = true, dataType = "string", paramType = "query", value = "Sorting criteria in the format: property(,asc|desc). "
+ "Default sort order is ascending. " + "Multiple sort criteria are supported.") })
#interface ApiPageable {
}
Which can then be used like so:
#ApiPageable
public Page<Data> getData(Pageable pageRequest) {
Vineet Bhatia's answer with #ApiImplicitParams looks fine. But I faced with situation, when #ApiIgnor and #ApiParam(hidden = true) doesn't work and you can still observe the asembler and pageable params. I fixed this problem by adding next line
docket.ignoredParameterTypes(Pageable.class, PagedResourcesAssembler.class);
to the Docket bean in my SwaggerConfig.
Java example:
Bean:
#Bean
public Docket api() {
return new Docket(DocumentationType.SWAGGER_2)
.select()
.paths(PathSelectors.any())
.build()
.directModelSubstitute(Pageable.class, SwaggerPageable.class);
}
SwaggerPageable:
#Getter
private static class SwaggerPageable {
#ApiParam(value = "Number of records per page", example = "0")
#Nullable
private Integer size;
#ApiParam(value = "Results page you want to retrieve (0..N)", example = "0")
#Nullable
private Integer page;
#ApiParam(value = "Sorting criteria in the format: property(,asc|desc). Default sort order is ascending. Multiple sort criteria are supported.")
#Nullable
private String sort;
}
Swagger:
Here is the version of the annotation that was integrated into springdoc-openapi-data-rest for OpenAPI v3:
#Target({ ElementType.METHOD, ElementType.ANNOTATION_TYPE })
#Retention(RetentionPolicy.RUNTIME)
#Parameter(in = ParameterIn.QUERY
, description = "Zero-based page index (0..N)"
, name = "page"
, content = #Content(schema = #Schema(type = "integer", defaultValue = "0")))
#Parameter(in = ParameterIn.QUERY
, description = "The size of the page to be returned"
, name = "size"
, content = #Content(schema = #Schema(type = "integer", defaultValue = "20")))
#Parameter(in = ParameterIn.QUERY
, description = "Sorting criteria in the format: property(,asc|desc). "
+ "Default sort order is ascending. " + "Multiple sort criteria are supported."
, name = "sort"
, content = #Content(array = #ArraySchema(schema = #Schema(type = "string"))))
public #interface PageableAsQueryParam {
}
See https://springdoc.github.io/springdoc-openapi-demos/faq.html#how-can-i-map-pageable-spring-date-commons-object-to-correct-url-parameter-in-swagger-ui
Open API 3.0 has seamless integration.
For example,
#GetMapping("/filter")
public Page<Employee> filterEmployees(Pageable pageable) {
return repository.getEmployees(pageable);
}
Add springdoc-openapi-data-rest dependency
implementation 'org.springdoc:springdoc-openapi-data-rest:1.5.2'
Note: You may add '#ParameterObject' if you have multiple parameters
public Page<Employee> filterEmployees(#ParameterObject Pageable pageable)
This solution works without the need to annotate every single API method in every single controller. First we create a replacement for Pageable class with the correct property names and descriptions (Kotlin code, you can use a Interface for Java):
data class SwaggerPageable(
#ApiModelProperty("Number of records per page", example = "20")
val size: Int?,
#ApiModelProperty("Results page you want to retrieve (0..N)", example = "0")
val page: Int?,
#ApiModelProperty("Sorting criteria in the format: property(,asc|desc)." +
"Default sort order is ascending. Multiple sort criteria are supported.")
var sort: String?
)
Then in the Swagger config, just add a direct substitute from Pageable to this class (again Kotlin code, but Java should be pretty similar):
#Bean
fun api(): Docket {
return Docket(DocumentationType.SWAGGER_2)
.select()
.paths(PathSelectors.any())
.build()
.directModelSubstitute(Pageable::class.java, SwaggerPageable::class.java)
}
The result looks like this:
The downside is not being able to define the default value in the ApiModelProperty, but this is more than good enough for my project.
Answer of Vineet Bhatia will have validation problem when you are not running on localhost. It will argue for integer parameters that they are not corresponding to json schema.
So I changed integer to string:
#ApiImplicitParams({
#ApiImplicitParam(name = "page", dataType = "string", paramType = "query",
value = "Results page you want to retrieve (0..N)"),
#ApiImplicitParam(name = "size", dataType = "string", paramType = "query",
value = "Number of records per page."),
#ApiImplicitParam(name = "sort", allowMultiple = true, dataType = "string", paramType = "query",
value = "Sorting criteria in the format: property(,asc|desc). " +
"Default sort order is ascending. " +
"Multiple sort criteria are supported.")
})
For people who wants to solve this problem in 2019. This configuration via springfox documentation works fine except you can't set description for parameters.
Code is here.
https://github.com/springfox/springfox/blob/ef1721afc4c910675d9032bee59aea8e75e06d27/springfox-data-rest/src/main/java/springfox/documentation/spring/data/rest/configuration/SpringDataRestConfiguration.java
here is a solution for springdoc-openapi-ui
SpringDocUtils.getConfig()
.replaceWithClass(org.springframework.data.domain.Pageable.class, SwaggerPageable.class);
#Getter
private static class SwaggerPageable {
#ApiParam(value = "Number of records per page", example = "0")
#Nullable
private Integer size;
#ApiParam(value = "Results page you want to retrieve (0..N)", example = "0")
#Nullable
private Integer page;
#ApiParam(value = "Sorting criteria in the format: property(,asc|desc). Default sort order is ascending. Multiple sort criteria are supported.")
#Nullable
private String sort;
}
Answer to validation problem indicated by Evgeny.
Using
#ApiImplicitParams({
#ApiImplicitParam(name = "page", dataType = "int", paramType = "query", value = "Results page you want to retrieve (0..N)"),
#ApiImplicitParam(name = "size", dataType = "int", paramType = "query", value = "Number of records per page."),
#ApiImplicitParam(name = "sort", allowMultiple = true, dataType = "string", paramType = "query", value = "Sorting criteria in the format: property(,asc|desc). "
+ "Default sort order is ascending. " + "Multiple sort criteria are supported.") })
throws an exception:
Illegal DefaultValue for parameter type integer
java.lang.NumberFormatException: For input string: ""
at java.base/java.lang.NumberFormatException.forInputString(NumberFormatException.java:68)
at java.base/java.lang.Long.parseLong(Long.java:709)
at java.base/java.lang.Long.valueOf(Long.java:1151)
at io.swagger.models.parameters.AbstractSerializableParameter.getExample(AbstractSerializableParameter.java:412)
at jdk.internal.reflect.GeneratedMethodAccessor366.invoke(Unknown Source)
at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:567)
at com.fasterxml.jackson.databind.ser.BeanPropertyWriter.serializeAsField(BeanPropertyWriter.java:688)
at com.fasterxml.jackson.databind.ser.std.BeanSerializerBase.serializeFields(BeanSerializerBase.java:719)
at com.fasterxml.jackson.databind.ser.BeanSerializer.serialize(BeanSerializer.java:155)
at com.fasterxml.jackson.databind.ser.impl.IndexedListSerializer.serializeContents(IndexedListSerializer.java:119)
at com.fasterxml.jackson.databind.ser.impl.IndexedListSerializer.serialize(IndexedListSerializer.java:79)
(at least, it does with springfox-swagger2 and springfox-swagger2-ui version 2.9.2)
You can avoid the exception by following Evgeny's answer or by adding default values and example values for the integer parameters:
#ApiImplicitParams({
#ApiImplicitParam(name = "page", dataType = "int", paramType = "query", value = "Results page you want to retrieve (0..N)", defaultValue = "0", example = "2"),
#ApiImplicitParam(name = "size", dataType = "int", paramType = "query", value = "Number of records per page.", defaultValue = "20", example = "10"),
#ApiImplicitParam(name = "sort", allowMultiple = true, dataType = "string", paramType = "query", value = "Sorting criteria in the format: property(,asc|desc). "
+ "Default sort order is ascending. " + "Multiple sort criteria are supported.") })
An update:
Use org.springdoc.core.converters.models.DefaultPageable instead of Pageable. It already has the #ParameterObject which makes it breakdown in to 3 parameters when the Swagger JSON is generated.
Although the solution with the implicit parameters works, it introduces a lot of extra, brittle code. In the end we went with the following solution:
#GetMapping(value = "/")
public HttpEntity<PagedResources<Item>> getItems(
#RequestParam(value = "page", required = false) Integer page,
#RequestParam(value = "size", required = false) Integer size,
PagedResourcesAssembler assembler) {
Page<Item> itemPage = itemService.listItems(PageRequest.of(page, size, Sort.unsorted()));
return new ResponseEntity<>(assembler.toResource(itemPage), HttpStatus.OK);
}
We pass a PageRequest (which implements Pageable) to our service, which returns a Page. (all from org.springframework.data.domain).
The org.springframework.data.web.PagedResourcesAssembler gets injected automagically via the controller method and allows mapping Items to org.springframework.hateoas.PagedResources
We didn't require dynamic sorting so we omitted that; it poses some challenges to add sorting since springfox does not play nice with org.springframework.data.domain.Sort.