Skip to content
PHP library to allow caching and compressing of javascript and css
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
build
src
static
tests
.gitignore
LICENSE
README.md
composer.json
composer.lock
implementation_idea.md
new_register_crushing_logic.md

README.md

cache-n-crush

Please note: This is a custom caching framework designed both to play around with using uglify js and uglify css, and to be used for a custom website not suited for standard caching systems. In production you probably will want to use other caching frameworks such as assetic.

Version 0.1 was a proof of concept to test if it worked. Since this does indeed appear to work im going to rework it to fill a number of potential objectives:

  • Ability to state which files you want to crunch and load
  • One file will be produced per set of crunch files. If you crunch A, B, and A+B they will exist as 3 separate crunch files
  • The output of crunch A and B will not be used for A and B, and will be completely separate
  • If the combined crunch is not available on the filesystem an exception will be thrown
  • In debug mode the raw files will be loaded instead of the crunched file
  • In debug mode, every time the raw files are loaded they are crunched
    • These crunched files will be saved as a new revision and will allow easy deployment calling the specific crunch method
    • When crunching files manually it will check for these new cached files and use them if present to speed up cache deployment
    • Some way to know which crunched files have new revisions you could deploy
  • CSS files will be supported to merge css into a single file
  • Uglify css will need to be ported over similarly to uglify js
    • maybe add some kind of similar interface, etc?
    • maybe allow any kind of crunching software to be plugged in?
  • Store a list of previously crunched files, and their current file somewhere "nice"
    • Allow a user to recrunch everything if its changed using above store
    • Allow marking a crunch as "stale" after a file has been removed from the filesystem
    • Some way to view this potentially?
  • Have some basic versioning of crunched files

Current Issues

Possible implementation

  • JSON store to store all the data about the files
  • Then export this to a basic php array file used for the main cache
  • JSON file stores all the details such as new files, etc
  • Each time the cache is modified json file is written out as a php file that works as the bootstrap

TODO:

  • Remove all comments but allow custom comments at the top of the file configurable
  • Allow getting the details of all previously crushed files and their constitutent paths
  • Allow crushing by identifier (using the details of the constituent parts)
  • Allow checking of identifiers to find what hasnt yet been updated (has changes in constituent files)
  • Comment all code
  • Comment all tests
  • Write some more tests
  • Create an example implementation repo?
  • Run speedtests on importing a json file and including the php file? (only used the PHP file including as I have made the assumptions that it will be faster, possibly not true?)

Usage

No documentation available currently

Installation

No documentation available currently

Tests

No documentation available currently

License

This is licensed under the MIT license. For more information see the LICENSE file.

You can’t perform that action at this time.