Module: CarrierWave::Mount

Included in:
ActiveRecord
Defined in:
lib/carrierwave/mount.rb

Overview

If a Class is extended with this module, it gains the mount_uploader method, which is used for mapping attributes to uploaders and allowing easy assignment.

You can use mount_uploader with pretty much any class, however it is intended to be used with some kind of persistent storage, like an ORM. If you want to persist the uploaded files in a particular Class, it needs to implement a `read_uploader` and a `write_uploader` method.

Defined Under Namespace

Modules: Extension

Instance Method Summary collapse

Instance Method Details

#mount_uploader(column, uploader = nil, options = {}, &block) ⇒ Object

Mounts the given uploader on the given column. This means that assigning and reading from the column will upload and retrieve files. Supposing that a User class has an uploader mounted on image, you can assign and retrieve files like this:

@user.image # => <Uploader>
@user.image.store!(some_file_object)

@user.image.url # => '/some_url.png'

It is also possible (but not recommended) to omit the uploader, which will create an anonymous uploader class.

Passing a block makes it possible to customize the uploader. This can be convenient for brevity, but if there is any significatnt logic in the uploader, you should do the right thing and have it in its own file.

Added instance methods

Supposing a class has used mount_uploader to mount an uploader on a column named image, in that case the following methods will be added to the class:

image

Returns an instance of the uploader only if anything has been uploaded

image=

Caches the given file

image_url

Returns the url to the uploaded file

image_cache

Returns a string that identifies the cache location of the file

image_cache=

Retrieves the file from the cache based on the given cache name

remote_image_url

Returns previously cached remote url

remote_image_url=

Retrieve the file from the remote url

remove_image

An attribute reader that can be used with a checkbox to mark a file for removal

remove_image=

An attribute writer that can be used with a checkbox to mark a file for removal

remove_image?

Whether the file should be removed when store_image! is called.

store_image!

Stores a file that has been assigned with image=

remove_image!

Removes the uploaded file from the filesystem.

image_integrity_error

Returns an error object if the last file to be assigned caused an integrity error

image_processing_error

Returns an error object if the last file to be assigned caused a processing error

image_download_error

Returns an error object if the last file to be remotely assigned caused a download error

image_identifier

Reads out the identifier of the file

Parameters

column (Symbol)

the attribute to mount this uploader on

uploader (CarrierWave::Uploader)

the uploader class to mount

options (Hash=> Object)

a set of options

&block (Proc)

customize anonymous uploaders

Options

:mount_on => Symbol

if the name of the column to be serialized to differs you can override it using this option

:ignore_integrity_errors => Boolean

if set to true, integrity errors will result in caching failing silently

:ignore_processing_errors => Boolean

if set to true, processing errors will result in caching failing silently

Examples

Mounting uploaders on different columns.

class Song
  mount_uploader :lyrics, LyricsUploader
  mount_uploader :alternative_lyrics, LyricsUploader
  mount_uploader :file, SongUploader
end

This will add an anonymous uploader with only the default settings:

class Data
  mount_uploader :csv
end

this will add an anonymous uploader overriding the store_dir:

class Product
  mount_uploader :blueprint do
    def store_dir
      'blueprints'
    end
  end
end


136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
# File 'lib/carrierwave/mount.rb', line 136

def mount_uploader(column, uploader=nil, options={}, &block)
  mount_base(column, uploader, options, &block)

  mod = Module.new
  include mod
  mod.class_eval <<-RUBY, __FILE__, __LINE__+1

    def #{column}
      _mounter(:#{column}).uploaders[0] or _mounter(:#{column}).blank_uploader
    end

    def #{column}=(new_file)
      _mounter(:#{column}).cache([new_file])
    end

    def #{column}_url(*args)
      #{column}.url(*args)
    end

    def #{column}_cache
      _mounter(:#{column}).cache_names[0]
    end

    def #{column}_cache=(cache_name)
      _mounter(:#{column}).cache_names = [cache_name]
    end

    def remote_#{column}_url
      [_mounter(:#{column}).remote_urls].flatten[0]
    end

    def remote_#{column}_url=(url)
      _mounter(:#{column}).remote_urls = [url]
    end

    def write_#{column}_identifier
      return if frozen?
      mounter = _mounter(:#{column})

      if mounter.remove?
        write_uploader(mounter.serialization_column, nil)
      elsif mounter.identifiers.first
        write_uploader(mounter.serialization_column, mounter.identifiers.first)
      end
    end

    def #{column}_identifier
      _mounter(:#{column}).read_identifiers[0]
    end
    
    def store_previous_changes_for_#{column}
      @_previous_changes_for_#{column} = changes[_mounter(:#{column}).serialization_column]
    end

    def remove_previously_stored_#{column}
      before, after = @_previous_changes_for_#{column}
      _mounter(:#{column}).remove_previous([before], [after])
    end
  RUBY
end

#mount_uploaders(column, uploader = nil, options = {}, &block) ⇒ Object

Mounts the given uploader on the given array column. This means that assigning and reading from the array column will upload and retrieve multiple files. Supposing that a User class has an uploader mounted on images, and that images can store an array, for example it could be a PostgreSQL JSON column. You can assign and retrieve files like this:

@user.images # => []
@user.images = [some_file_object]
@user.images # => [<Uploader>]

@user.images[0].url # => '/some_url.png'

It is also possible (but not recommended) to omit the uploader, which will create an anonymous uploader class.

Passing a block makes it possible to customize the uploader. This can be convenient for brevity, but if there is any significatnt logic in the uploader, you should do the right thing and have it in its own file.

Added instance methods

Supposing a class has used mount_uploaders to mount an uploader on a column named images, in that case the following methods will be added to the class:

images

Returns an array of uploaders for each uploaded file

images=

Caches the given files

images_urls

Returns the urls to the uploaded files

images_cache

Returns a string that identifies the cache location of the files

images_cache=

Retrieves the files from the cache based on the given cache name

remote_image_urls

Returns previously cached remote urls

remote_image_urls=

Retrieve files from the given remote urls

remove_images

An attribute reader that can be used with a checkbox to mark the files for removal

remove_images=

An attribute writer that can be used with a checkbox to mark the files for removal

remove_images?

Whether the files should be removed when store_image! is called.

store_images!

Stores all files that have been assigned with images=

remove_images!

Removes the uploaded file from the filesystem.

images_integrity_error

Returns an error object if the last files to be assigned caused an integrity error

images_processing_error

Returns an error object if the last files to be assigned caused a processing error

images_download_error

Returns an error object if the last files to be remotely assigned caused a download error

image_identifiers

Reads out the identifiers of the files

Parameters

column (Symbol)

the attribute to mount this uploader on

uploader (CarrierWave::Uploader)

the uploader class to mount

options (Hash=> Object)

a set of options

&block (Proc)

customize anonymous uploaders

Options

:mount_on => Symbol

if the name of the column to be serialized to differs you can override it using this option

:ignore_integrity_errors => Boolean

if set to true, integrity errors will result in caching failing silently

:ignore_processing_errors => Boolean

if set to true, processing errors will result in caching failing silently

Examples

Mounting uploaders on different columns.

class Song
  mount_uploaders :lyrics, LyricsUploader
  mount_uploaders :alternative_lyrics, LyricsUploader
  mount_uploaders :files, SongUploader
end

This will add an anonymous uploader with only the default settings:

class Data
  mount_uploaders :csv_files
end

this will add an anonymous uploader overriding the store_dir:

class Product
  mount_uploaders :blueprints do
    def store_dir
      'blueprints'
    end
  end
end


285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
# File 'lib/carrierwave/mount.rb', line 285

def mount_uploaders(column, uploader=nil, options={}, &block)
  mount_base(column, uploader, options, &block)

  mod = Module.new
  include mod
  mod.class_eval <<-RUBY, __FILE__, __LINE__+1

    def #{column}
      _mounter(:#{column}).uploaders
    end

    def #{column}=(new_files)
      _mounter(:#{column}).cache(new_files)
    end

    def #{column}_urls(*args)
      _mounter(:#{column}).urls(*args)
    end

    def #{column}_cache
      names = _mounter(:#{column}).cache_names
      names.to_json if names.present?
    end

    def #{column}_cache=(cache_name)
      _mounter(:#{column}).cache_names = JSON.parse(cache_name) if cache_name.present?
    end

    def remote_#{column}_urls
      _mounter(:#{column}).remote_urls
    end

    def remote_#{column}_urls=(urls)
      _mounter(:#{column}).remote_urls = urls
    end

    def write_#{column}_identifier
      return if frozen?
      mounter = _mounter(:#{column})

      if mounter.remove?
        write_uploader(mounter.serialization_column, nil)
      elsif mounter.identifiers.any?
        write_uploader(mounter.serialization_column, mounter.identifiers)
      end
    end

    def #{column}_identifiers
      _mounter(:#{column}).read_identifiers
    end
    
    def store_previous_changes_for_#{column}
      @_previous_changes_for_#{column} = changes[_mounter(:#{column}).serialization_column]
    end

    def remove_previously_stored_#{column}
      _mounter(:#{column}).remove_previous(*@_previous_changes_for_#{column})
    end
  RUBY
end

#uploader_option(column, option) ⇒ Object

Return a particular option for a particular uploader

Parameters

column (Symbol)

The column the uploader is mounted at

option (Symbol)

The option, e.g. validate_integrity

Returns

Object

The option value



42
43
44
45
46
47
48
# File 'lib/carrierwave/mount.rb', line 42

def uploader_option(column, option)
  if uploader_options[column].has_key?(option)
    uploader_options[column][option]
  else
    uploaders[column].send(option)
  end
end

#uploader_optionsObject



26
27
28
# File 'lib/carrierwave/mount.rb', line 26

def uploader_options
  @uploader_options ||= superclass.respond_to?(:uploader_options) ? superclass.uploader_options.dup : {}
end

#uploadersObject

Returns

Hash=> CarrierWave

what uploaders are mounted on which columns



22
23
24
# File 'lib/carrierwave/mount.rb', line 22

def uploaders
  @uploaders ||= superclass.respond_to?(:uploaders) ? superclass.uploaders.dup : {}
end