How to programmatically retrieve configuration of the specific rule?
I tried
$settings = rules_config_load('RULE_NAME');
It returns very basic info (name, ID etc) and empty "settings" array.
Cant also get it directly from DB .
It is stored in serialized array that can not be fully processed with the
unserialize() function
I got the same issue and end up here, here is a solution i found, you need to call actions() on your rule to access settings:
$rule = rules_config_load('RULE_NAME');
foreach ($rule->actions() as $action) {
$settings[] = $action->settings;
}
Related
I'm new to Drupal. I looked here and on google for a while before asking, but I'm sure I can't find the answer because I don't know how to ask the question.
Here is what's going on. I'm using a custom module to load certain entities and then output them in a specific format for an application to access. The problem is that the NODE BODY contains special information and media files that should be converted. My goal is to obtain the HTML output that would normally be used on this field.
// Execute an EntityFieldQuery
$result = $query->execute();
if (isset($result['node'])) {
$article_items_nids = array_keys($result['node']);
$article_items = entity_load('node', $news_items_nids);
}
// Loop through each article
foreach ($article_items as $article) {
return $article->body[LANGUAGE_NONE]['0']['value'];
}
All of this works great. The only problem is that I get things like this in the output:
[[{"type":"media","view_mode":"media_original","fid":"283","attributes":{"alt":"","class":"media-image","data-thmr":"thmr_32","height":"400","width":"580"}}]]
or
*protoss_icon*
My goal is to find a way that these items are converted just like they are when these articles are viewed normally.
I've tried doing things such as:
render(field_view_field('node', $article, 'body'));
or
render($article->body[LANGUAGE_NONE]['0']['value']);
without success. Thanks for any help, I'm learning so I don't have a complete grasp of the process drupal uses to build output.
You can try something like this (this works only with nodes not with other custom entity types):
$node = node_load($nid);
$field = field_get_items('node', $node, 'your_field_name');
$output = field_view_value('node', $node, 'your_field_name', $field[$delta]);
the field_view_value returns a renderable array for a single field value. (from drupal api documentation)
In my symfony2 application I need to display some totals at the top of all pages, ie "Already 200,154,555 users registered".
I don't want to run the query to come up with that count on every page load. The solution I've come up with is to create a "variable" entity that would have two columns, name and value. Then I would set up a console command that runs on cron which would update these variable entities (eg "totalPeople") with a query that counted all the rows of people, etc.
This feels a little heavy handed... Is there a better solution to this problem?
You could set global parameters and add a service to rewrite them. Then call the service from your Command.
Or directly set up a service to read/write a file (as a json array for example).
Or set up a option table with a row storing the data. It's not going to be a resource intensive query that way.
Here is what I'm using to store RSS feeds (after I parsed them)
public function checkCache($data=array(), $path = '')
{
foreach ($data as $service => $feed)
{
$service = strtolower($service);
$service = str_replace(' ', '-', $service);
$path = $path.'web/bundles/citation/cache/rss/' . $service . '.cache';
if ((!file_exists($path) || time() - filemtime($path) > 900) && $cache = fopen($path, 'w'))
{
$rss_contents = $this->getFeed($feed); //fetch feed content & returns array
fwrite($cache, serialize($rss_contents));
fclose($cache);
return $rss_contents;
}
else
{
$cache = fopen($path, 'r');
return unserialize(file_get_contents($path));
fclose($cache);
}
}
}
You can implement that on your backend for example so every time an admin logs it'll check for cache and refresh only if it's too old. Although I'm quite fond of the 4AM cron job solution too.
You could use the pagination feature of doctrine (if you use doctrine). That will leverage the "limit" part of your queries (even with joins) and will give you a total count of rows (via a count query).
I have a Drupal content type which contains a number of computed fields. Some (but not all) items are being added to this content type via a cron-triggered RSS feed importer. I'm trying to trigger computed field generation for new items in hook_cron. The following code grabs all items that haven't been tagged as 'submitted', loads and re-saves the node, and then marks the node as 'submitted'.
$query = db_select('node', 'n');
$query->fields('n', array('nid'));
$table_alias = $query->join('field_data_field_submitted', 'r', 'n.nid = r.entity_id AND r.field_submitted_value = 0');
$result = $query->execute();
foreach ($result as $record){
$q = $record->nid;
$n = node_load($q);
node_save($n);
$query = db_update('field_data_field_submitted')
->fields(array('field_submitted_value' => 1))
->condition('entity_id', $q)
->execute();
}
This code works the way I expect it to if I call it from a module-generated page (created using hook_menu with a page callback function). Nodes are resubmitted, and the computed field data is generated. When I put this code in my hook_cron function, the query works, it loops through the records and updates the 'submitted' value, but the computed fields are not computed. I'm confused as to why this would not get triggered in cron. Any help?
Doh! Finally realized that this was completely my own doing. Due to the nature of this content type, where we allow anonymous users to create new content, but explicitly do not trigger the computed fields when they create the content (long story, but short form is that authenticated users then verify & enhance this content, which is where the computed fields come in). So, as I was setting up the initial content, I disabled the computed fields for anonymous users (if $user->uid > 0), and completely forgot about that. Once I tweaked that logic to allow computed fields to be processed on import (triggering it with a field that has a value for the imported content, but not for other content), the problem was solved.
The cron run has access to the full bootstrap so there's no logical reason why your code would produce different results in that context.
That said, you're only updating the field_data_field_submitted table when you also need to update the field_revision_field_submitted table, so that might somehow account for the discrepancy.
Drupal provides an API for the field system so that these sorts of problems can be avoided completely. The same code you've used, rewritten the 'Drupal' way, would be:
$query = new EntityFieldQuery;
$query->entityCondition('entity_type', 'node')
->fieldCondition('field_submitted', 'value', 0);
$results = $query->execute();
if (!empty($results['node'])) {
$nodes = node_load_multiple(array_keys($results['node']));
foreach ($nodes as $node) {
$node->field_submitted[$node->language][0]['value'] = 1;
node_save($node);
}
}
I can't think of a good reason why the above code would fail on cron either so it might be worth giving it a whirl.
I have an issue with triming a field before it is saved. I wanted to use substr(), or regex() with preg_match(). I have built a Drupal 7 module, but it can't work at all. I have tried using the trim plugin in feeds tamper module, but it doesn't seem to work. The data I am using is from a feed from Google Alerts. I have posted this issue here.
This is what I have done so far, and I know my regular expression is wrong; I was trying to get it do anything, just to see if I could get it to work, but I am pretty lost on how to add this type of function to a Drupal module.
function sub_node_save() {
$url = $node->field_web_screenhot['und'][0]['url'];
$url = preg_match('~^(http|ftp)(s)?\:\/\/((([a-z0-9\-]*)(\.))+[a-z0-9]*)($|/.*$)~i',$url );
$node->field_web_screenhot['und'][0]['url'] =$url;
return ;
}
I used the Devel module to get the field.
If there's an easy way to use substr(), I would consider that or something else.
Basically, I just want to take the Google redirect off the URL, so it is just the basic URL to the web site.
Depending on your question and later comments, I'd suggesting using node_presave hook (http://api.drupal.org/api/drupal/modules!node!node.api.php/function/hook_node_presave/7) for this.
It's called before both insert (new) and update ops so you will need extra validations to prevent it from executing on node updates if you want.
<?php
function MYMODULE_node_presave($node) {
// check if nodetype is "mytype"
if ($node->type == 'mytype'){
// PHP's parse_url to get params set to an array.
$parts = parse_url($node->field_web_screenhot['und'][0]['url']);
// Now we explode the params by "&" to get the URL.
$queryParts = explode('&', $parts['query']);
$params = array();
foreach ($queryParts as $param) {
$item = explode('=', $param);
$params[$item[0]] = $item[1];
}
//valid_url validates the URL (duh!), urldecode() makes the URL an actual one with fixing "//" in http, q is from the URL you provided.
if (valid_url(urldecode($parms['q']))){
$node->field_web_screenhot['und'][0]['url'] = urldecode($parms['q']);
}
}
}
I have Pathauto configured to generate an alias based on the title of a node, for a specific content type. The problem is that I want to make small changes in this title before Pathauto uses it to generate the alias.
The first comment in this post suggests the use of hook_token_values, but I couldn't really understand how to use it, even after reading the docs. In my tests, when I implement this hook, the alias generated is always "array", which means I'm missing something.
Any help? Thanks.
It might be that you missed to implement hook_token_list as well. Providing a new token is a two step process:
Implement hook_token_list to declare the tokens you are going to provide. This will just be the name of the tokens, along with a short explanation, and the information to what type of objects the tokens will apply (e.g. node, user, taxonomy, ...)
Implement hook_token_value to actually generate the content of the tokens. This will be called when the tokens are to be replaced with the content they should stand for.
As you just want to provide an alternative version of the title token already provided by the token module, it is probably best to just copy the relevant portions from token_node.inc, stripped down to the relevant cases and adjusted to be used in another module:
/**
* Implementation of hook_token_list().
*/
function yourModule_token_list($type = 'all') {
if ($type == 'node' || $type == 'all') {
$tokens['node']['yourModule-title'] = t('Node title (customized version by yourModule)');
return $tokens;
}
}
This simply says that yourModule provides a token for node objects, named yourModule-title, along with a short description. The main work gets done in the other hook:
/**
* Implementation of hook_token_values().
*/
function yourModule_token_values($type, $object = NULL, $options = array()) {
$values = array();
switch ($type) {
case 'node':
$node = $object;
// TODO: Replace the check_plain() call with your own token value creation logic!
$values['yourModule-title'] = check_plain($node->title);
break;
}
return $values;
}
This will be called whenever the tokens for node objects are needed, with the node in question being passed as the $object parameter (for a user token, the $type would be 'user', and $object would be the user object, and so on for other types). What it does is creating an array of values, keyed by the token name, with the replacement for that token as the value. The original code from token_node.inc just runs the title through check_plain(), so this would be the place to insert your own logic.
In Drupal 7, the token functionality has been moved to core. Tokens are implemented by the hook_tokens and hook_token_info methods. For usage examples, follow the links provided, and look for links to functions that implement hook_tokens and hook_token_info… I found the statistics_tokens and statistics_token_info functions helpful in understanding how this hook works.
It's probably also worth noting that this hook needs to be implemented by a module… my first attempt I dropped my test functions into the theme's template.php, only to have nothing happen at all :-p