Optimize & reduce lib folder for WebViewer

WebViewer comes bundled with the files required for all use cases. In many situations, some of these files are not needed.

A script comes packaged with Webviewer that can be used to delete any files that you will not be using. It is available in the scripts/ folder under the @pdftron/webviewer package. You can run this script by (at the root of your WebViewer folder) executing:

1npm install
2npm run optimize

The script will ask you a series of questions and delete files based on your answers:

• Do you want us to back up your files before optimizing?
  • Answer no to this question if you don't need to back up your files first.
  • Answer yes to this question if you do need to back up your files first.
• Will you be using WebViewer Server?
  • Answer yes to this question if you will be using WebViewer Server.
  • Answer no if you will be using the default client only mode.
• Will you be converting all your documents to XOD?
  • Answer yes to this question if you will be using a server to convert all your documents to the XOD format before viewing.
  • Answer no if you will be opening any format other than XOD.
• Do you need client side office support (docx, pptx, xlsx)?
  • Answer yes to this question if you want to support opening or converting Microsoft Office files (such as .docx, .pptx, xlsx).
  • Answer no if you do not need Microsoft office support.
• Do you need client side office support for legacy office files (doc, ppt, xls)?
  • Answer yes to this question if you want to support opening or converting Microsoft Office legacy files (such as .doc, .ppt, xls).
  • Answer no if you do not need Microsoft office support for legacy files.
• Do you need the full PDF API?
  • We provide a full API that can process documents outside of the viewer.
  • Answer yes to this question if you need to access to the full API.
  • In most cases, you do not need the full API and can answer no to this question.
• Do you want to use the product version of PDFnet.js?
  • Answer yes to this question if you don't need to access to type checking or console messages and would like a smaller version than the development version.
  • Answer no if you do need access to type checking or console messages and want the larger development version.
• Do you need to use the content editing feature?
  • Answer yes to this question if you want to use the content editing feature.
  • Answer no if you do not want to use the content editing feature.
• Do you need to use the office editing feature?
  • Answer yes to this question if you want to use the office editing feature.
  • Answer no if you do not want to use the office editing feature.
• Do you need to use the spreadsheet editing feature?
  • Answer yes to this question if you want to use the spreadsheet editing feature.
  • Answer no if you do not want to use the spreadsheet editing feature.
• Do you need to deploy to Salesforce?
  • To be able to upload WebViewer source code to Salesforce as static resource files of 5 MB in size, we need to optimize and split up WebViewer code into separate chunks.
  • Answer yes to this question if you are planning to deploy to Salesforce platform and need to optimize WebViewer.
  • Answer no if you are not deploying to Salesforce platform.
• Do you want to exclude the use of optimized worker files? If you exclude them make sure that you pass enableOptimizedWorkers: false to the WebViewer constructor.
  • Answer no to this question if you want to use optimized workers, which are faster, but may not be supported on some older browsers.

Note that the lib/core/pdf folder contains multiple versions of the same PDF SDK logic that are optimized for different browsers. The versions are:

• Threaded Wasm (Threaded WebAssembly): Currently only supported by Chrome but more browsers will support it in the future.
• Wasm: WebAssembly version for Edge, FireFox, iOS 11+ and Android Chrome 59+
• asm.js: For IE11, iOS10-, Android Chrome 58-. This is the slowest version.

Only 1 of the 3 is actually loaded at runtime, so while your server would have all the files, each user would only download a subset of them.