[javascript] How do I download a file with Angular2 or greater

I have a WebApi / MVC app for which I am developing an angular2 client (to replace MVC). I am having some troubles understanding how Angular saves a file.

The request is ok (works fine with MVC, and we can log the data received) but I can't figure out how to save the downloaded data (I am mostly following the same logic as in this post). I am sure it is stupidly simple, but so far I am simply not grasping it.

The code of the component function is below. I've tried different alternatives, the blob way should be the way to go as far as I understood, but there is no function createObjectURL in URL. I can't even find the definition of URL in window, but apparently it exists. If I use the FileSaver.js module I get the same error. So I guess this is something that changed recently or is not yet implemented. How can I trigger the file save in A2?

downloadfile(type: string){

    let thefile = {};
    this.pservice.downloadfile(this.rundata.name, type)
        .subscribe(data => thefile = new Blob([data], { type: "application/octet-stream" }), //console.log(data),
                    error => console.log("Error downloading the file."),
                    () => console.log('Completed file download.'));

    let url = window.URL.createObjectURL(thefile);
    window.open(url);
}

For the sake of completeness, the service that fetches the data is below, but the only thing it does is to issue the request and pass on the data without mapping if it succeeds:

downloadfile(runname: string, type: string){
   return this.authHttp.get( this.files_api + this.title +"/"+ runname + "/?file="+ type)
            .catch(this.logAndPassOn);
}

I got a solution for downloading from angular 2 without getting corrupt, using spring mvc and angular 2

1st- my return type is :-ResponseEntity from java end. Here I am sending byte[] array has return type from the controller.

2nd- to include the filesaver in your workspace-in the index page as:

<script src="https://cdnjs.cloudflare.com/ajax/libs/FileSaver.js/2014-11-29/FileSaver.min.js"></script>

3rd- at component ts write this code:

import {ResponseContentType} from '@angular.core';

let headers = new Headers({ 'Content-Type': 'application/json', 'MyApp-Application' : 'AppName', 'Accept': 'application/pdf' });
        let options = new RequestOptions({ headers: headers, responseType: ResponseContentType.Blob });
            this.http
            .post('/project/test/export',
                    somevalue,options)
              .subscribe(data => {

                  var mediaType = 'application/vnd.ms-excel';
                  let blob: Blob = data.blob();
                    window['saveAs'](blob, 'sample.xls');

                });

This will give you xls file format. If you want other formats change the mediatype and file name with right extension.

Update to Hector's answer using file-saver and HttpClient for step 2:

public downloadFile(file: File): Observable<Blob> {
    return this.http.get(file.fullPath, {responseType: 'blob'})
}

How about this?

this.http.get(targetUrl,{responseType:ResponseContentType.Blob})
        .catch((err)=>{return [do yourself]})
        .subscribe((res:Response)=>{
          var a = document.createElement("a");
          a.href = URL.createObjectURL(res.blob());
          a.download = fileName;
          // start download
          a.click();
        })

I could do with it.
no need additional package.

For newer angular versions:

npm install file-saver --save
npm install @types/file-saver --save


import {saveAs} from 'file-saver/FileSaver';

this.http.get('endpoint/', {responseType: "blob", headers: {'Accept': 'application/pdf'}})
  .subscribe(blob => {
    saveAs(blob, 'download.pdf');
  });

If you only send the parameters to a URL, you can do it this way:

downloadfile(runname: string, type: string): string {
   return window.location.href = `${this.files_api + this.title +"/"+ runname + "/?file="+ type}`;
}

in the service that receives the parameters

I am using Angular 4 with the 4.3 httpClient object. I modified an answer I found in Js' Technical Blog which creates a link object, uses it to do the download, then destroys it.

Client:

doDownload(id: number, contentType: string) {
    return this.http
        .get(this.downloadUrl + id.toString(), { headers: new HttpHeaders().append('Content-Type', contentType), responseType: 'blob', observe: 'body' })
}

downloadFile(id: number, contentType: string, filename:string)  {

    return this.doDownload(id, contentType).subscribe(  
        res => { 
            var url = window.URL.createObjectURL(res);
            var a = document.createElement('a');
            document.body.appendChild(a);
            a.setAttribute('style', 'display: none');
            a.href = url;
            a.download = filename;
            a.click();
            window.URL.revokeObjectURL(url);
            a.remove(); // remove the element
        }, error => {
            console.log('download error:', JSON.stringify(error));
        }, () => {
            console.log('Completed file download.')
        }); 

} 

The value of this.downloadUrl has been set previously to point to the api. I am using this to download attachments, so I know the id, contentType and filename: I am using an MVC api to return the file:

 [ResponseCache(Location = ResponseCacheLocation.None, NoStore = true)]
    public FileContentResult GetAttachment(Int32 attachmentID)
    { 
        Attachment AT = filerep.GetAttachment(attachmentID);            
        if (AT != null)
        {
            return new FileContentResult(AT.FileBytes, AT.ContentType);  
        }
        else
        { 
            return null;
        } 
    } 

The attachment class looks like this:

 public class Attachment
{  
    public Int32 AttachmentID { get; set; }
    public string FileName { get; set; }
    public byte[] FileBytes { get; set; }
    public string ContentType { get; set; } 
}

The filerep repository returns the file from the database.

Hope this helps someone :)

I was facing this same case today, I had to download a pdf file as an attachment (the file shouldn't be rendered in the browser, but downloaded instead). To achieve that I discovered I had to get the file in an Angular Blob, and, at the same time, add a Content-Disposition header in the response.

This was the simplest I could get (Angular 7):

Inside the service:

getFile(id: String): Observable<HttpResponse<Blob>> {
  return this.http.get(`./file/${id}`, {responseType: 'blob', observe: 'response'});
}

Then, when I need to download the file in a component, I can simply:

fileService.getFile('123').subscribe((file: HttpResponse<Blob>) => window.location.href = file.url);

UPDATE:

Removed unnecessary header setting from service

<a href="my_url" download="myfilename">Download file</a>

my_url should have the same origin, otherwise it will redirect to that location

As mentioned by Alejandro Corredor it is a simple scope error. The subscribe is run asynchronously and the open must be placed in that context, so that the data finished loading when we trigger the download.

That said, there are two ways of doing it. As the docs recommend the service takes care of getting and mapping the data:

//On the service:
downloadfile(runname: string, type: string){
  var headers = new Headers();
  headers.append('responseType', 'arraybuffer');
  return this.authHttp.get( this.files_api + this.title +"/"+ runname + "/?file="+ type)
            .map(res => new Blob([res],{ type: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet' }))
            .catch(this.logAndPassOn);
}

Then, on the component we just subscribe and deal with the mapped data. There are two possibilities. The first, as suggested in the original post, but needs a small correction as noted by Alejandro:

//On the component
downloadfile(type: string){
  this.pservice.downloadfile(this.rundata.name, type)
      .subscribe(data => window.open(window.URL.createObjectURL(data)),
                  error => console.log("Error downloading the file."),
                  () => console.log('Completed file download.'));
  }

The second way would be to use FileReader. The logic is the same but we can explicitly wait for FileReader to load the data, avoiding the nesting, and solving the async problem.

//On the component using FileReader
downloadfile(type: string){
    var reader = new FileReader();
    this.pservice.downloadfile(this.rundata.name, type)
        .subscribe(res => reader.readAsDataURL(res), 
                    error => console.log("Error downloading the file."),
                    () => console.log('Completed file download.'));

    reader.onloadend = function (e) {
        window.open(reader.result, 'Excel', 'width=20,height=10,toolbar=0,menubar=0,scrollbars=no');
  }
}

Note: I am trying to download an Excel file, and even though the download is triggered (so this answers the question), the file is corrupt. See the answer to this post for avoiding the corrupt file.

The following code worked for me

let link = document.createElement('a');
link.href = data.fileurl; //data is object received as response
link.download = data.fileurl.substr(data.fileurl.lastIndexOf('/') + 1);
link.click();

I found the answers so far lacking insight as well as warnings. You could and should watch for incompatibilities with IE10+ (if you care).

This is the complete example with the application part and service part after. Note that we set the observe: "response" to catch the header for the filename. Also note that the Content-Disposition header has to be set and exposed by the server, otherwise the current Angular HttpClient will not pass it on. I added a dotnet core piece of code for that below.

public exportAsExcelFile(dataId: InputData) {
    return this.http.get(this.apiUrl + `event/export/${event.id}`, {
        responseType: "blob",
        observe: "response"
    }).pipe(
        tap(response => {
            this.downloadFile(response.body, this.parseFilename(response.headers.get('Content-Disposition')));
        })
    );
}

private downloadFile(data: Blob, filename: string) {
    const blob = new Blob([data], {type: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet;charset=utf-8;'});
    if (navigator.msSaveBlob) { // IE 10+
        navigator.msSaveBlob(blob, filename);
    } else {
        const link = document.createElement('a');
        if (link.download !== undefined) {
            // Browsers that support HTML5 download attribute
            const url = URL.createObjectURL(blob);
            link.setAttribute('href', url);
            link.setAttribute('download', filename);
            link.style.visibility = 'hidden';
            document.body.appendChild(link);
            link.click();
            document.body.removeChild(link);
        }
    }
}

private parseFilename(contentDisposition): string {
    if (!contentDisposition) return null;
    let matches = /filename="(.*?)"/g.exec(contentDisposition);

    return matches && matches.length > 1 ? matches[1] : null;
}

Dotnet core, with Content-Disposition & MediaType

 private object ConvertFileResponse(ExcelOutputDto excelOutput)
    {
        if (excelOutput != null)
        {
            ContentDisposition contentDisposition = new ContentDisposition
            {
                FileName = excelOutput.FileName.Contains(_excelExportService.XlsxExtension) ? excelOutput.FileName : "TeamsiteExport.xlsx",
                Inline = false
            };
            Response.Headers.Add("Access-Control-Expose-Headers", "Content-Disposition");
            Response.Headers.Add("Content-Disposition", contentDisposition.ToString());
            return File(excelOutput.ExcelSheet, "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet");
        }
        else
        {
            throw new UserFriendlyException("The excel output was empty due to no events.");
        }
    }

It will be better if you try to call the new method inside you subscribe

this._reportService.getReport()
    .subscribe((data: any) => {
        this.downloadFile(data);
    },
        (error: any) => ?onsole.log(error),
        () => console.log('Complete')
    );

Inside downloadFile(data) function we need to make block, link, href and file name

downloadFile(data: any, type: number, name: string) {
    const blob = new Blob([data], {type: 'text/csv'});
    const dataURL = window.URL.createObjectURL(blob);

    // IE doesn't allow using a blob object directly as link href
    // instead it is necessary to use msSaveOrOpenBlob
    if (window.navigator && window.navigator.msSaveOrOpenBlob) {
      window.navigator.msSaveOrOpenBlob(blob);
      return;
    }

    const link = document.createElement('a');
    link.href = dataURL;
    link.download = 'export file.csv';
    link.click();

    setTimeout(() => {

      // For Firefox it is necessary to delay revoking the ObjectURL
      window.URL.revokeObjectURL(dataURL);
      }, 100);
    }
}

If a tab opens and closes without downloading anything, i tried following with mock anchor link and it worked.

downloadFile(x: any) {
var newBlob = new Blob([x], { type: "application/octet-stream" });

    // IE doesn't allow using a blob object directly as link href
    // instead it is necessary to use msSaveOrOpenBlob
    if (window.navigator && window.navigator.msSaveOrOpenBlob) {
      window.navigator.msSaveOrOpenBlob(newBlob);
      return;
    }

    // For other browsers: 
    // Create a link pointing to the ObjectURL containing the blob.
    const data = window.URL.createObjectURL(newBlob);

    var link = document.createElement('a');
    link.href = data;
    link.download = "mapped.xlsx";
    // this is necessary as link.click() does not work on the latest firefox
    link.dispatchEvent(new MouseEvent('click', { bubbles: true, cancelable: true, view: window }));

    setTimeout(function () {
      // For Firefox it is necessary to delay revoking the ObjectURL
      window.URL.revokeObjectURL(data);
      link.remove();
    }, 100);  }

Download *.zip solution for angular 2.4.x: you must import ResponseContentType from '@angular/http' and change responseType to ResponseContentType.ArrayBuffer (by default it ResponseContentType.Json)

getZip(path: string, params: URLSearchParams = new URLSearchParams()): Observable<any> {
 let headers = this.setHeaders({
      'Content-Type': 'application/zip',
      'Accept': 'application/zip'
    });

 return this.http.get(`${environment.apiUrl}${path}`, { 
   headers: headers, 
   search: params, 
   responseType: ResponseContentType.ArrayBuffer //magic
 })
          .catch(this.formatErrors)
          .map((res:Response) => res['_body']);
}

If you don't need to add headers in the request, to download a file in Angular2 you can do a simple (KISS PRINCIPLE):

window.location.href='http://example.com/myuri/report?param=x';

in your component.

To download and show PDF files, a very similar code snipped is like below:

  private downloadFile(data: Response): void {
    let blob = new Blob([data.blob()], { type: "application/pdf" });
    let url = window.URL.createObjectURL(blob);
    window.open(url);
  }

  public showFile(fileEndpointPath: string): void {
    let reqOpt: RequestOptions = this.getAcmOptions();  //  getAcmOptions is our helper method. Change this line according to request headers you need.
    reqOpt.responseType = ResponseContentType.Blob;
    this.http
      .get(fileEndpointPath, reqOpt)
      .subscribe(
        data => this.downloadFile(data),
        error => alert("Error downloading file!"),
        () => console.log("OK!")
      );
  }

This is for folks looking how to do it using HttpClient and file-saver:

1. Install file-saver

npm install file-saver --save

npm install @types/file-saver --save

API Service class:

export() {
    return this.http.get(this.download_endpoint, 
        {responseType: 'blob'});
}

Component:

import { saveAs } from 'file-saver';
exportPdf() {
    this.api_service.export().subscribe(data => saveAs(data, `pdf report.pdf`));
}

 let headers = new Headers({
                'Content-Type': 'application/json',
                'MyApp-Application': 'AppName',
                'Accept': 'application/vnd.ms-excel'
            });
            let options = new RequestOptions({
                headers: headers,
                responseType: ResponseContentType.Blob
            });


this.http.post(this.urlName + '/services/exportNewUpc', localStorageValue, options)
                .subscribe(data => {
                    if (navigator.appVersion.toString().indexOf('.NET') > 0)
                    window.navigator.msSaveBlob(data.blob(), "Export_NewUPC-Items_" + this.selectedcategory + "_" + this.retailname +"_Report_"+this.myDate+".xlsx");

                    else {
                        var a = document.createElement("a");
                        a.href = URL.createObjectURL(data.blob());
                        a.download = "Export_NewUPC-Items_" + this.selectedcategory + "_" + this.retailname +"_Report_"+this.myDate+ ".xlsx";
                        a.click();
                    }
                    this.ui_loader = false;
                    this.selectedexport = 0;
                }, error => {
                    console.log(error.json());
                    this.ui_loader = false;
                    document.getElementById("exceptionerror").click();
                });

Try this!

1 - Install dependencies for show save/open file pop-up

npm install file-saver --save
npm install @types/file-saver --save

2- Create a service with this function to recive the data

downloadFile(id): Observable<Blob> {
    let options = new RequestOptions({responseType: ResponseContentType.Blob });
    return this.http.get(this._baseUrl + '/' + id, options)
        .map(res => res.blob())
        .catch(this.handleError)
}

3- In the component parse the blob with 'file-saver'

import {saveAs as importedSaveAs} from "file-saver";

  this.myService.downloadFile(this.id).subscribe(blob => {
            importedSaveAs(blob, this.fileName);
        }
    )

This works for me!

Here's something I did in my case -

// service method
downloadFiles(vendorName, fileName) {
    return this.http.get(this.appconstants.filesDownloadUrl, { params: { vendorName: vendorName, fileName: fileName }, responseType: 'arraybuffer' }).map((res: ArrayBuffer) => { return res; })
        .catch((error: any) => _throw('Server error: ' + error));
}

// a controller function which actually downloads the file
saveData(data, fileName) {
    var a = document.createElement("a");
    document.body.appendChild(a);
    a.style = "display: none";
    let blob = new Blob([data], { type: "octet/stream" }),
        url = window.URL.createObjectURL(blob);
    a.href = url;
    a.download = fileName;
    a.click();
    window.URL.revokeObjectURL(url);
}

// a controller function to be called on requesting a download
downloadFiles() {
    this.service.downloadFiles(this.vendorName, this.fileName).subscribe(data => this.saveData(data, this.fileName), error => console.log("Error downloading the file."),
        () => console.info("OK"));
}

The solution is referenced from - here

You may also download a file directly from your template where you use download attribute and to [attr.href] you can provide a property value from the component. This simple solution should work on most browsers.

<a download [attr.href]="yourDownloadLink"></a>

Reference: https://www.w3schools.com/tags/att_a_download.asp

I share the solution that helped me (any improvement is greatly appreciated)

On your service 'pservice' :

getMyFileFromBackend(typeName: string): Observable<any>{
    let param = new URLSearchParams();
    param.set('type', typeName);
    // setting 'responseType: 2' tells angular that you are loading an arraybuffer
    return this.http.get(http://MYSITE/API/FILEIMPORT, {search: params, responseType: 2})
            .map(res => res.text())
            .catch((error:any) => Observable.throw(error || 'Server error'));
}

Component Part :

downloadfile(type: string){
   this.pservice.getMyFileFromBackend(typename).subscribe(
                    res => this.extractData(res),
                    (error:any) => Observable.throw(error || 'Server error')
                );
}

extractData(res: string){
    // transforme response to blob
    let myBlob: Blob = new Blob([res], {type: 'application/vnd.oasis.opendocument.spreadsheet'}); // replace the type by whatever type is your response

    var fileURL = URL.createObjectURL(myBlob);
    // Cross your fingers at this point and pray whatever you're used to pray
    window.open(fileURL);
}

On the component part, you call the service without subscribing to a response. The subscribe for a complete list of openOffice mime types see : http://www.openoffice.org/framework/documentation/mimetypes/mimetypes.html

Simply put the url as href as below .

<a href="my_url">Download File</a>

This answer suggests that you cannot download files directly with AJAX, primarily for security reasons. So I'll describe what I do in this situation,

01. Add href attribute in your anchor tag inside the component.html file,
eg:-

<div>
       <a [href]="fileUrl" mat-raised-button (click)='getGenaratedLetterTemplate(element)'> GENARATE </a>
</div>

02. Do all following steps in your component.ts to bypass the security level and bring the save as popup dialog,
eg:-

import { environment } from 'environments/environment';
import { DomSanitizer } from '@angular/platform-browser';
export class ViewHrApprovalComponent implements OnInit {
private apiUrl = environment.apiUrl;
  fileUrl
 constructor(
    private sanitizer: DomSanitizer,
    private letterService: LetterService) {}
getGenaratedLetterTemplate(letter) {

    this.data.getGenaratedLetterTemplate(letter.letterId).subscribe(
      // cannot download files directly with AJAX, primarily for security reasons);
    console.log(this.apiUrl + 'getGeneratedLetter/' + letter.letterId);
    this.fileUrl = this.sanitizer.bypassSecurityTrustResourceUrl(this.apiUrl + 'getGeneratedLetter/' + letter.letterId);
  }

Note: This answer will work if you are getting an error "OK" with status code 200

Downloading file through ajax is always a painful process and In my view it is best to let server and browser do this work of content type negotiation.

I think its best to have

<a href="api/sample/download"></a> 

to do it. This doesn't even require any new windows opening and stuff like that.

The MVC controller as in your sample can be like the one below:

[HttpGet("[action]")]
public async Task<FileContentResult> DownloadFile()
{
    // ...
    return File(dataStream.ToArray(), "text/plain", "myblob.txt");
}

For those using Redux Pattern

I added in the file-saver as @Hector Cuevas named in his answer. Using Angular2 v. 2.3.1, I didn't need to add in the @types/file-saver.

The following example is to download a journal as PDF.

The journal actions

public static DOWNLOAD_JOURNALS = '[Journals] Download as PDF';
public downloadJournals(referenceId: string): Action {
 return {
   type: JournalActions.DOWNLOAD_JOURNALS,
   payload: { referenceId: referenceId }
 };
}

public static DOWNLOAD_JOURNALS_SUCCESS = '[Journals] Download as PDF Success';
public downloadJournalsSuccess(blob: Blob): Action {
 return {
   type: JournalActions.DOWNLOAD_JOURNALS_SUCCESS,
   payload: { blob: blob }
 };
}

The journal effects

@Effect() download$ = this.actions$
    .ofType(JournalActions.DOWNLOAD_JOURNALS)
    .switchMap(({payload}) =>
        this._journalApiService.downloadJournal(payload.referenceId)
        .map((blob) => this._actions.downloadJournalsSuccess(blob))
        .catch((err) => handleError(err, this._actions.downloadJournalsFail(err)))
    );

@Effect() downloadJournalSuccess$ = this.actions$
    .ofType(JournalActions.DOWNLOAD_JOURNALS_SUCCESS)
    .map(({payload}) => saveBlobAs(payload.blob, 'journal.pdf'))

The journal service

public downloadJournal(referenceId: string): Observable<any> {
    const url = `${this._config.momentumApi}/api/journals/${referenceId}/download`;
    return this._http.getBlob(url);
}

The HTTP service

public getBlob = (url: string): Observable<any> => {
    return this.request({
        method: RequestMethod.Get,
        url: url,
        responseType: ResponseContentType.Blob
    });
};

The journal reducer Though this only sets the correct states used in our application I still wanted to add it in to show the complete pattern.

case JournalActions.DOWNLOAD_JOURNALS: {
  return Object.assign({}, state, <IJournalState>{ downloading: true, hasValidationErrors: false, errors: [] });
}

case JournalActions.DOWNLOAD_JOURNALS_SUCCESS: {
  return Object.assign({}, state, <IJournalState>{ downloading: false, hasValidationErrors: false, errors: [] });
}

I hope this is helpful.

Well, I wrote a piece of code inspired by many of the above answers that should easily work in most scenarios where the server sends a file with a content disposition header, without any third-party installations, except rxjs and angular.

First, how to call the code from your component file

this.httpclient.get(
   `${myBackend}`,
   {
      observe: 'response',
      responseType: 'blob'
   }
).pipe(first())
.subscribe(response => SaveFileResponse(response, 'Custom File Name.extension'));

As you can see, it's basically pretty much the average backend call from angular, with two changes

1. I am observing the response instead of the body
2. I am being explicit about the response being a blob

Once the file is fetched from the server, I am in principle, delegating the entire task of saving the file to the helper function, which I keep in a separate file, and import into whichever component I need to

export const SaveFileResponse = 
(response: HttpResponse<Blob>, 
 filename: string = null) => 
{
    //null-checks, just because :P
    if (response == null || response.body == null)
        return;

    let serverProvidesName: boolean = true;
    if (filename != null)
        serverProvidesName = false;

    //assuming the header is something like
    //content-disposition: attachment; filename=TestDownload.xlsx; filename*=UTF-8''TestDownload.xlsx
    if (serverProvidesName)
        try {
            let f: string = response.headers.get('content-disposition').split(';')[1];
            if (f.includes('filename='))
                filename = f.substring(10);
        }
        catch { }
    SaveFile(response.body, filename);
}

//Create an anchor element, attach file to it, and
//programmatically click it. 
export const SaveFile = (blobfile: Blob, filename: string = null) => {
    const a = document.createElement('a');
    a.href = window.URL.createObjectURL(blobfile);
    a.download = filename;
    a.click();
}

There, no more cryptic GUID filenames! We can use whatever name the server provides, without having to specify it explicitly in the client, or, overwrite the filename provided by the server (as in this example). Also, one can easily, if need be, change the algorithm of extracting the filename from the content-disposition to suit their needs, and everything else will stay unaffected - in case of an error during such extraction, it will just pass 'null' as the filename.

As another answer already pointed out, IE needs some special treatment, as always. But with chromium edge coming in a few months, I wouldn't worry about that while building new apps (hopefully). There is also the matter of revoking the URL, but I'm kinda not-so-sure about that, so if someone could help out with that in the comments, that would be awesome.