f3-factory/f3com-data

Content for the CMS

F3com-data

Content for the F3 CMS on fatfreeframework.com

Local Installation

• To get a local instance up and running that includes this documentation repo, please visit https://github.com/F3Community/F3com for instructions.

Participate

• Fork the repository
• Edit your file
• Send a pull request on GitHub
• After approval, your changes will be visible on the website

Hint: you can also modify files directly on GitHub which does the rest for you automatically.

Compatibility

With commit e56ac9b5f0623dc18b33d207b6d831ca546a1878, this repo became incompatible with the old F3com Wiki due to the new folder structure. If you want to use the old way, you have to checkout the latest working commit via

git checkout e82879cec1084b01ec6e4b3758673c2637328628

f3com-data's Issues

DEBUG Range

The Documentation:

https://github.com/F3Community/F3com-data/blob/3b0c48fe7c58bcf8f2f5ba843f541cb54408647c/framework-variables/main.md

says:

While developing your application, it's best to set the debug level to maximum so you can trace all errors to their root cause:
$f3->set('DEBUG',3);

and later

DEBUG can have values ranging from 0 (stack trace suppressed) to 2 (most verbose with classes & functions calls logs).

This is unclear shouldn´t either first be two or second quote say 3?

– anatol

8x Errors 404

Hi all,
Starting a new project served by F3

hence 1st thing 1st, reading the doc again

While browsing the doc found a couple of 404 as per:

https://www.drlinkcheck.com/results/4ef435/1/broken

Nothing serious, most old plugins and the like

Thanks a lot for F3

Milestones for the Website

Can be found at the base repository at
https://github.com/ikkez/F3com/issues

An argument seems missing in the Web mime method description

Here: https://github.com/f3-factory/F3com-data/blob/master/3.7/web/main.md#mime

There seems to be a second argument called $inspect that is not listed. I might be wrong, but doesn't it need to be explicitly set to true to get the benefits of the "lightweight alternative to Fileinfo"?

$f3->exists undocumented second parameter

There are no examples or explanations (or any other documentations) for the second parameter ($val) of $f3->exists( string $key [, mixed &$val=NULL] ) )

going to https://fatfreeframework.com/ in vivaldi allows me to download the php page

I am unable to duplicate this in chrome, however when I try to access from vivaldi browser, I received the php file as shown below.

the missing v3.4.x docs

• ONREROUTE variable in quick-references bcosca/fatfree#661
• Base->recursive is missing
• Base->mock details
• custom config sections @xfra35
• new ALIAS route building in template bcosca/fatfree#720
• View->afterrender hook bcosca/fatfree#663
• SMTP->attach has a new $alias parameter https://github.com/bcosca/fatfree/pull/617/files
• new SQL->pdo
• FRAGMENT hive var bcosca/fatfree@69b5a60
• EXCEPTION var
• make lists not war
• new abort and until method f3-factory/fatfree-core@c7ba6b4#commitcomment-10202651

Extend unit testing documentation

A user had questions about unit testing but the documentation has no complete overview. I gathered some ideas and snippets:

• Copy/Move information from bcosca/fatfree#unit-testing to F3Community/F3com-data
• Extend https://fatfreeframework.com/3.6/test where necessary
• Explain how to run the F3's unit tests (dev branch of bcosca/fatfree)
• It could also be useful to document how F3 can be combined with other testing tools, e.g. see my answer at Unit Testing with Fat-Free-Framework which utilizes PHPUnit.

Docs versioning

Hi guys,

3.6 has been released and there are quite some changes to bring to the docs.

Maybe it's time to decide how the website content should follow the framework history:

1. should it match the latest release only? (that's how it's been working so far)
2. or should it keep a history of changes? (that could be helpful for migrations and/or supporting older releases)

If we stick to the first strategy, we'd better create a dev branch that we could feed until the next release.

If we go for the second strategy, we need to decide how to display the docs history. It could be a small changelog for each modified API method, like on php.net, or a general version selector, like on dev.mysql.com.

Incorrect number of args in Session anti-csrf example

Apologies if I'm not reporting this correctly.

In 3.6 Session documentation.
Section Methods

NB: as an alternative, you can instantiate the Session class with the 4th parameter set to a hive key name, which will hold the CSRF token. E.g:
new DB\SQL\Session($db,'sessions',NULL,'CSRF');// now $f3->CSRF holds the token

For an SQL backed session, it should be the 5th argument, since there is an extra $force argument.
I.e:

function __construct(\DB\SQL $db,$table='sessions',$force=TRUE,$onsuspect=NULL,$key=NULL) {

letter wrong in docs

On page https://fatfreeframework.com/3.6/plug-ins#GrabbingDatafromAnotherSite

'headers'=>array( 'Accept: text/html,application/xhtml+xml,application/xml', 'Accept-Language: en-us' ),

need to be to work
'header'=>array( 'Accept: text/html,application/xhtml+xml,application/xml', 'Accept-Language: en-us' ),

Tag the GitHub repository

I propose to add the following tags to the repository:

• fat-free-framework
• documentation

and perhaps

• php because it's the documentation for a PHP framework

I would have added the tags if I had the permission to do so.

The tags could also be added to fatfree and fatfree-core. I'm not sure if it's necessary to tag F3com.

Remove links to Slack in favor of Matrix

See bcosca/fatfree#1272

git fix

@sn0opy can you force-pull the latest commits on the server once again please?
something went wrong again, see http://fatfreeframework.com/sql-mapper#changed

Routing: Catch-all

I feel a sub-section within routing should be devoted to displaying the correct way to provide a catch-all... Fat-Free doesn't provide a direct method for catch-all, so it may confuse new users. It's also one of the first things I do within a project to provide custom logic instead of a default 405.

Now personally I have two ways of achieving this, but I figured I'd ask everyone else to see if you guys have your own ways.

Method 1:

// known routes should be added before catch-all...

$f3->route('GET|HEAD|POST /*/*', function($f3) {
    //
    // add custom logic here....
    //

    // call F3 error handler for cases not matching custom logic
    $f3->error(405);
});

In my tests, "//" proved to work but I didn't use a full test suite with edge cases, so I'm unsure if it provides true catch-all support.
Using "//" also infers that a method has already caught all single parameter routes. E.g. /@action. With this said the wildcard may need to be at a proportional depth compared with existing routes.

Method 2:

I'm going to be lazy here, but it involves adding a custom error handler to catch 405 errors... I'll just provide an example from my error handler:

\Error::instance()->addHandler(405, function() use($f3) {
    // custom logic
    if (!$f3->get('AJAX') && in_array($f3->get('VERB'), array('GET', 'HEAD', 'POST'))) {
        $f3->reroute('/');
    }

    // call default F3 error handler (will pass on 405 error)
    call_user_func_array(array($f3, 'error'), $f3->get('ERROR'));
});

That's about it. Let me know your thoughts, or if you have an alternative method (we can pick which one is best suited) :)

Change tutorial to use `fatfree-core` instead of `fatfree`

When changing in the tutorial from fatfree to fatfree-core we are loosing the lib folder. The tutorial explains that we can use it to add additional libraries there. Since most projects are using Composer anyway, this is no loss IMHO. It would be nice to mention that it it's easy to configure the built-in autoloader or to inform the reader about the autoloading section.

Document system variable SEED

• New framework variable: SEED
• Usage of SEED in CSRF tokens

OAuth2 documentation missing in API Reference

Original issue: f3-factory/fatfree-core#175

Documentation for the RFC6455 WebSocket Server (since 3.6.0)

The WebSocket Server isn't documented yet.

It could be possible that I am remembering it wrong, but didn't @bcosca post an example somewhere?