1. I’m thinking the time might have come to write a wrapper around DOMDocument which actually makes it usable. Thoughts:

    • automatic conversion of various encodings to HTML entities to scoot round encoding issues
    • XPath queries still work but querySelector and querySelectorAll are implemented for both the document and individual elements via Symfony XPath → CSS converter and relative XPath queries
    • A DOMNodeList which actually implements ArrayAccess instead of acting like a fake array
    • Perhaps some javascript-inspired property names like innerText, innerHTML for consistency
    • Maybe some jQuery-influenced shortcut goodness for doing things like removing/replacing elements
  2. I’m rather impressed with guzzlephp.org’s HTTP Link header abstraction. Parsing Link headers and providing a simple API to check for the existence of/fetch links by rel is welcome attention to detail and cements it’s position as the only PHP HTTP client I will likely ever need.

  3. Need to use require.js to load a bunch of scripts compiled via assetic into a PHP file, annoyed by auto-append of .js, don’t want to set up irritating routing? Add a ? to the URL, require.js will add a .js to the query string, loading the file correctly.

  4. Turns out that performing a GET request on a data URI from PHP works if file_get_contents is used, not if cURL is used. I wonder what support is like for other server side languages — using data URIs in with could be the basis of some interesting .

  5. New CRUD fetcher/saver design coming on well. Decided on per-semantic-indexing, with abstract indexes and APIs so I never have to think about the SQL under the logic (which I do with ORMs like Doctrine).

    Implementing the whole lot using traits, too — mixins FTW.

  6. Finally decided that symfony Security component is way too complicated for my little , so ditching it — but I’ve learnt a lot from digging through it and my further efforts will try to provide some of the amazing flexibility it gives whilst being more performant and easier to understand #php

  7. Made some updates to my note autolinking flow — instead of lots of unstable regexes, only one runs now (courtesy of cassis) and the rest manipulate the XML–compatible HTML which results. Much more robust, much more extendable

  8. I love that now has shiny namespacing and a thriving code sharing community, but I think the heavily hierarchical namespacing practises used by some of the community (e.g. symfony components) are unhealthy.

    They are difficult to memorise, relying on (often slow) IDE autocomplete, and encourage a use statement for each class. That’s pretty much a scoped equivalent of from x import * in python — not a good practise! It’s still namespace pollution, it just takes longer to write.

    I am trying to use a more python–like, package-centred approach with much fewer subnamespaces. The outcome of this should be that you use the package name:

    use BarnabyWalters

    …and then using all the classes/subnamespaces from that root, e.g: $t = Posse\Helpers::convertHtmlToTwitterFormat($s);