gulp-htmlcpr

gulp-htmlcpr is a Gulp plugin that copies HTML files with all the local assets (images, scripts, style files, etc.) that it needs. Gulp is a streaming build system utilizing node.js.

Install

npm install --save-dev gulp-htmlcpr

Usage

var htmlcpr = require('gulp-htmlcpr');

gulp.src('./src/index.html')
  .pipe(htmlcpr())
  .pipe(gulp.dest('./dist/'))

API

htmlcpr(options)

options.noRec

Type: string

Default value: ''

Files in the noRec directory will be copied, but the links in these files won't be followed.

options.blacklistFn

Type: function (url: string, filepath: string): bool

Default value: null

The function is called for every URL detected by the plugin. First parameter url is the detected URL and the second parameter src is the path of the currently processed file.

If the function returns truthy value the link is replaced with '' and the file referenced by the link is not copied to the destination dir.

options.skipFn

Type: function (url: string, filepath: string): bool

Default value: null

The function is called for every URL detected by the plugin. First parameter url is the detected URL and the second parameter src is the path of the currently processed file.

If the function returns truthy value, the link will be skipped. This runs prior to blacklistFn, so in the case of a url being matched by both, it will still be skipped.

options.schemelessUrlFix

Type: string|function (url: string, src: string): bool

Default value: null

The function is called when a schemeless url is detected (e.g. //cdn.superfast.com/myself.png). If the value of the option is a string, that string is considered to be the protocol and is appended to the URL. If the value is the function, that function is being called and the URL is replaced with its return value.

options.overwritePath

Type: function (filePathFromBase: string): string

Default value: undefined

The function is used to move some files to new location. For example, if all HTML page assets have to moved to a single sub-directory assets/, then overwritePath should be defined as follows:

  var path = require('path');
  ...
  overwritePath: function (src) {
    return path.join('assets', src);
  }
  ...

The emitted files will have the updated path, and the links to these files will be updated as well.

Usage Examples

Basic Use Case

Let's say you have a directory with the following structure:

├── app
│   ├── index.html
│   ├── css
│   │   ├── ... CSS files ...
│   │   └──
│   ├── images
│   │   ├── ... images ...
│   │   └──
│   ├── js
│   │   ├── ... javascript ...
│   │   └──
│   ├── fonts
│   │   ├── ... fonts ...
│   │   └──
│   └── vendor
│       ├── ... vendor files ...
│       └──
└── build

You want to copy index.html and any local files required by this file (images, scripts, etc.). You need to copy only the files required by index.html and no more.

htmlcpr comes to the rescue!

The following configuration would do exactly what you want:

gulp.src(['index.html', 'fonts/**'])
  .pipe(htmlcpr({
    norecDir: 'vendor',
  })
  .pipe(gulp.dest('build'))

If the only files included from index.html are css/main.css, js/main.js and vendor/fancy.css; in thier turns,css/main.css turn includes images/logo.png and vendor/fancy.css includes vendor/fancy-spinner.png. Then the contents of build directory after an execution of grun htmlcpr would be:

└── build
    ├── index.html
    ├── css
    │   └── main.css          # included by index.html
    ├── images
    │   └── logo.png          # included by main.css
    ├── js
    │   └── main.js           # included by index.html
    ├── fonts
    │   ├── ... all fonts ... # the fonts are explicitly specified
    │   ├── ... from app/ ... # in task object, hence they were
    │   └── ...    dir    ... # copied as is.
    └── vendor
        └── fancy.css

NOTE: since vendor was set as norecDir, any links in the vendor/fancy.css file were not followed and hence vendor/fancy-spinner.png was not copied.