Section:

Optimize / customize WebViewer Server usage

Normal Usage

WebViewer Server is designed to work hand-in-hand with the WebViewer client in order to seamlessly optimize the viewing experience. Once the server is running, simply supply the webviewerServerURL argument to WebViewer upon startup:

JavaScript (v8.0+)

1options.webviewerServerURL = 'http://<docker_address>'
2WebViewer(options, viewerElement);

JavaScript (v7.0+)

1options.pdftronServer = 'http://<docker_address>'
2WebViewer(options, viewerElement);

This will ensure an optimal client experience on any of your target platforms, optimizing the delivery method as required.

Caching and Performance

In order to handle requests as quickly as possible with minimal server load, WebViewer Server caches intermediate data opportunistically. For example, this might include includes caching parsed PDF content in memory for quick re-renders of the same doc, caching conversion results, fetched data, and render tiles, if used.

It will also batch requests together in order to minimize cpu-heavy operations like conversions -- if users Alice and Bob try to view the same Word document at the same time, then only one conversion to PDF will ever take place, and the will be broadcast to both users. For documents under high contention (as might happen when sharing a doc for review), the performance benefit can be tremendous.

In order for this caching and batching to work correctly, the server requires that incoming URLs map to documents in a one-to-one manner, including query parameters. If Alice and Bob access the same document but via different URLs, then the server will assume they are different documents and end up doing more work than is required. In order to ensure the cache is working as intended, it may be necessary to move unique URL query parameters (access tokens and time stamps, for example) into the request headers instead.

Controlling the Document Cache with CacheKey

We also offer method of controlling caching on the WebViewer Server. When requesting documents you can pass a cacheKey to the load document call. This will cause the document requested to be cached on the cacheKey you have provided. If a request is ever made to this cacheKey, the document first cached under that key will be returned.

JavaScript

1WebViewer(...)
2  .then(instance => {
3    const options = {
4      cacheKey: "unique_id_entry123",
5    };
6    const url = 'http://<documentserver>/FileDownload?param1=abcd&fetchid=dcba';
7    instance.loadDocument(url, options);
8  });
9});

Server authentication

When a document URL is provided to WebViewer, that URL will now be fetched by the docker server instead of the browser. If authorization information needs to be provided to the server providing the document, this can be done using the customHeaders option when creating a document.

You can also provide the filename option to ensure that the server knows what type of file you are attempting to view (in case it is not clear from the URL).

JavaScript

1WebViewer(...)
2  .then(instance => {
3    const options = {
4      customHeaders: { Cookie: "MYAUTHTOKEN=BF6CF50AB90C4025" },
5      filename: "Document.pdf"
6    };
7    const url = 'http://<documentserver>/FileDownload?param1=abcd&fetchid=dcba';
8    instance.loadDocument(url, options);
9  });
10});

When the docker server makes a request to fetch http://<documentserver>/FileDownload?param1=abcd&fetchid=dcba, the http request will include the values in options.customHeaders.

Custom Query Parameters

It is also possible to append a query parameter onto every request made to and from WebViewer Server. This can be used as a form of authentication or to handle some other system. To do so, call loadDocument with your custom query parameters like so:

JavaScript

1instance.loadDocument('http://pdftron.com/mydocument.pdf', { 
2  webViewerServerCustomQueryParameters: { param1: 'param1data', param2: 'param2data'}
3});

Now every link will have param1 and param2 appended to it as a query parameter.

Extended API

For most use cases no extra interaction with the server is required, beyond what is mentioned in the previous section. For special cases, these endpoints can be used to further optimize your application, or tailor it to your own particular use case

PreloadURL

Preload a document into WebViewer server prior to actually viewing it.

Parameters

url

string

The URL of the document to be preloaded, should be a fully qualified absolute path

Returns

Returns code 200 once the information has been received by the server. The response is asynchronous -- the return does not indicate that all preload work has been completed.

Example

For example, to pre-fetch http://domain/document.pdf, make a GET request to http://<docker_address>/blackbox/PreloadURL?url=http%3A%2F%2Fdomain%2Fdocument.pdf

Metrics

GET
myserver/blackbox/metrics

Returns metric data for Prometheus.

Returns

  • response code 200 - Returns the metric data in the Prometheus format

Example

You can use this API with Prometheus by setting a target in Prometheus to it. Prometheus will then scrape the APIs as defined in our configuration docs.

HealthCheck

GET
myserver/blackbox/HealthCheck

Checks whether your server is still able to process jobs. WebViewer Server tests at 30 second intervals as to whether it can process a document and updates the result of this check.

Returns

  • response code 200 - the server is up and working
  • response code 503 - the server is down

Example

For example, to check the state of your server, make a GET request to http://<docker_address>/blackbox/HealthCheck

GetPerfInfo

GET
myserver/blackbox/GetPerfInfo

GetPerfInfo returns the queue latency (how long it takes to start working on any new work item) averaged over different timescales.

If the queue is momentarily held up by work items during usage, this is normal, and would show up as fluctuating values at small timescales (with larger timescales staying pretty steady)

If the queue is blocked due to work items with very heavy compute load or other reasons, it would manifest as all timescales rising steadily.

Returns

Returns a JSON object containing the queue times for each timescale.

response code 200

JSON

1{"queueTime": [ 500ms timescale, 2s timescale, 5s timescale, 30s timescale, 60s timescale ]
2    "errorRate": [ unused ]
3    "systemInfo": {
4      "memoryData":{"total":value in mb,"available":value in mb},
5      "diskData":{"total":value in GB,"free":value in GB},
6      "heapSize":value in MB,
7      "heapMaxSize":value in MB,
8      "server_id":unique id,
9      "os_info": operating system,
10      "convert_queue_length": number of items queued for converting,
11      "fetch_queue_length": number of items queued for fetching,
12      "main_queue_length": number of items queued for all other types of work,
13      "cpu_load": value (percentage of total),
14      }
15    }

Example

For example, to check the state of your work queue, access http://domain/document.pdf, make a GET request to http://<docker_address>/blackbox/GetPerfInfo

GetSupportedFormats

GET
myserver/blackbox/GetSupportedFormats

Get the currently supported formats by the server. Will return the formats in a JSON array. Any formats not listed by this API will be rejected by the server. Requires 1.5.7 or greater.

Returns

  • response code 200 - will return the allowed server formats in a JSON array

Example

curl localhost:8090/blackbox/GetSupportedFormats

GetPDF

GET
myserver/blackbox/GetPDF

GET request to convert a document to PDF format, and return the result.

Parameters

uri

string

The URI of the document to be converted, should be a fully qualified absolute path. Required.

ext

string

For usage when the file you wish to use GetPDF with is not a PDF. Specify the format of the file you wish to convert to PDF.

fmt

string

The return format, if set to "data", will redirect to the actual pdf document, otherwise will return json data in the form [{"uri":"<location of the pdf>"}]

linearize

boolean

If set to 'true', will linearize the returned document.

xfdf

string

The raw annotation data you wish to merge in with the document when performing the GetPDF job. This is expected in the XFDF format. Requires a POST request instead of a GET request.

PDFTron-Custom

string

A HTTP header which can be set with a JSON object containing custom headers like so:

JavaScript

1request.setRequestHeader('PDFTron-Custom', JSON.stringify({
2    Cookie: "AUTHCOOKIE=123;OTHERCOOKIE=124", myHeader:"test"
3  }));

The custom headers will be used when the server fetches the document requested through the uri argument.

cacheKey

string

The key to cache the document against. This will force the server to always return the same cached item corresponding to the cacheKey passed.

Returns

  • If fmt is set to "data", returns the converted PDF data.
  • Otherwise returns a url to the converted result, relative to http://<docker_address>/blackbox/. This result is in json: [{"uri":"<location of the pdf>"}]
  • In the case where this functionality has not been licensed, will return code 403 (forbidden).

Example

To obtain http://domain/document.docx in PDF format, access , make a GET request to http://<docker_address>/blackbox/GetPDF?uri=http%3A%2F%2Fdomain%2Fdocument.docx&fmt=data&ext=docx

To instead obtain a json formated link to the conversion result, make a GET request to http://<docker_address>/blackbox/GetPDF?uri=http%3A%2F%2Fdomain%2Fdocument.docx

Changing the result URL download name

If you wish for the result from GetPDF to be saved with a different name than it has and downloaded directly, set the query parameters on the download link attach=true and dlname to the name you wish to give the file. The example below shows how this would be done:

http://localhost:8090/data/Converted/Fetched/myfile228288ccc.pdf?attach=true&dlname=mynewdocumentname.pdf

CacheCheck

GET
myserver/blackbox/CacheCheck

Check if a document was already cached by WebViewer Server.

The parameters used should match the options used by the document when it was originally cached. The parameter options are as follows. Parameters will effect whether the cache finds the document.

Parameters

uri

string

Required

The URL for the file. This can be a cid://url in the case of file uploads, or an internet url.

ext

string

The manually specified extension of the file.

password

string

The password used for the file.

cacheKey

string

The cacheKey used for the file.

officeOptions

string

The officeOptions used for the file.

cadOptions

string

The CAD options used for the file.

rastOptions

string

The rasterizer options used for the file.

Returns

Returns 200 if the document was cached, 404 if it was not.

Example

JavaScript

1const request = new XMLHttpRequest();
2  const docUrl = 'https://pdftron.s3.amazonaws.com/downloads/pl/webviewer-demo.pdf';
3  let url = joinPaths('https://demo.apryse.com/blackbox/CacheCHeck?uri=' + encodeURIComponent(docUrl) + "&password=mypassword");
4  request.open('GET', url);
5  request.onreadystatechange = () => {
6    if (request.readyState === 4) {
7      try {
8        if (request.status === 200) {
9          console.log("Success")
10        } else {
11          //error
12        }
13      } catch (e) {
14        //error
15      }
16    }
17  };
18  request.send();

GetThumb

GET
myserver/blackbox/GetThumb

Retrieve an image version of a PDF page. Allows the retrieving of files that can be used as thumbnails or previews.

Parameters

uri

string

The URI of the document to be converted, should be a fully qualified absolute path

page

string

The page to render, with page 0 representing the first page.

size

string

The size of the thumbnail expected. Accepts large (1280p) medium (640p) and small (320p)

ext

string

For usage when the file you wish to use GetThumb with is not a PDF. Specify the format of the file you wish to generate a thumbnail for.

renderAnnots

boolean

Defaults to false, if set to true includes annotations on thumbnail

PDFTron-Custom

string

A HTTP header which can be set with a JSON object containing custom headers like so:

JavaScript

1request.setRequestHeader('PDFTron-Custom', JSON.stringify({
2    customHeaders: { Cookie: "AUTHCOOKIE=123"}
3  }));

The custom headers will be used when the server fetches the document requested through the uri argument.

Returns

Returns code 200 if the image render was successful and returns json containing the URI to the image file in the format:

JavaScript

1{
2  [uri: 'https://pdftron.com/document.pdf']
3}

Example

JavaScript

1const request = new XMLHttpRequest();
2  const docUrl = 'https://pdftron.s3.amazonaws.com/downloads/pl/webviewer-demo.pdf';
3  const page = 0;
4  const size = 'large';
5  let url = joinPaths('https://demo.apryse.com/blackbox/GetThumb?uri=' + encodeURIComponent(docUrl) + "&page=" + page + "&size=" + size);
6  request.open('GET', url);
7  request.withCredentials = true;
8  request.onreadystatechange = () => {
9    if (request.readyState === 4) {
10      try {
11        if (request.status === 200) {
12          const data = JSON.parse(request.responseText);
13          for (let i = 0; i < data.length; i++) {
14            if(data[i].uri){
15              ///do something with thumb
16            }
17          }
18        } else {
19          //error
20        }
21      } catch (e) {
22        //error
23      }
24    }
25  };
26  request.send();

Did you find this helpful?

Trial setup questions?

Ask experts on Discord

Need other help?

Contact Support

Pricing or product questions?

Contact Sales