I am writing logic hook and I need to update 1 module fields by using other module elements.
<?php
class logic_hooks_class {
function after_save_method($bean, $event, $arguments) {
if (!isset($bean->ignore_update_c)||$bean->ignore_update_c === false {
//here I need to get module's reservations element: amount_reserved
//should I load relationship like $bean->load_relationship('reservations'); ??
//need to set total_reserved = amount_reserved;
//by the way amount_reserved might have several values for one reserved
$bean->goods = $bean->amount-$bean->total_reserved;
$bean->ignore_update_c = true;
$bean->save();
}
}
}
?>
You can load the beans for the relationship like this:
$bean->load_relationship('reservations');
$reservations = $bean->reservations->getBeans();
Now one can loop the $reservations and fetch and sum up the wanted value.
Just to clarify how load_relationship works. The parameter for load_relationship should be the link vardef pointing to the name of the relationship. A logical name is the plural name for the module, like reservations.
Related
This is a really trivial problem. I am just curious on how to deal with this in a "professional" manner.
I am trying to stick to variable naming convention. For NodeJs I am doing camelCasing. For database, I am using PostgreSQL and using underscore_casing.
Now the problem arises when I query data from PostgreSQL. I'll get a user object with following format,
{user_id: 1, account_type : "Admin"}
I can pass this object directly to server side-render and will have to use underscore casing to access account_type. Of course, I can manually create a new user JSON object with property userId and accountType but that is unnecessary work.
Is it possible to follow variable naming convention for both language and avoid having mixed variable names casing in some files? What is a good way to stay organized?
The are two good ways to approach this issue. The simplest one - do no conversion, use the exact database names. And the second one is to camel-case columns automatically.
Either way, you should always follow the underscore notation for all PostgreSQL declarations, as it will give you the option to activate camel-casing in your app at a later time, if it becomes necessary. Never use camel-case inside the database, or you will end up in a lot of pain later.
If you want the best of both worlds, follow the underscore notation for all PostgreSQL declarations, and convert to camel-case as you read data.
Below is an example of how to do it properly with pg-promise, copied from event receive example:
// Example below shows the fastest way to camelize column names:
const options = {
receive(e) {
camelizeColumns(e.data);
}
};
function camelizeColumns(data) {
const template = data[0];
for (var prop in template) {
const camel = pgp.utils.camelize(prop);
if (!(camel in template)) {
for (var i = 0; i < data.length; i++) {
const d = data[i];
d[camel] = d[prop];
delete d[prop];
}
}
}
}
Also see the following article: Pg-promise and case sensitivity in column names.
UPDATE
The code above has been updated for use of pg-promise v11 or later.
I've struggled with this too, and I've concluded that there's really no way to avoid this kind of ugliness unless you rewrite the objects that come from the database. Fortunately, that's not too difficult in Javascript:
const fromDBtoJS = (obj) => {
// declare a variable to hold the result
const result = {};
// iterate over the keys on the object
Object.keys(obj).forEach((key) => {
// adjust the key
const newKey = key.replace(/_[a-z]/g, (x) => x[1].toUpperCase());
// add the value from the old object with the new key
result[newKey] = obj[key];
});
// return the result
return result;
};
Here's a JSFiddle. The "replace" code above was found here
If you wanted to use classes for models in your application, you could incorporate this code into the constructor or database load method so it's all handled more-or-less automatically.
I want to override the item listing template file core/themes/classy/templates/dataset/item-list.html.twig for listing the fields field_slider_images as well as field_blog_tags respectively of their's multiple values of the field.
I have selected "Unordered List" in the view.
Please do check the attached image.
I have created following files :
item-list--field-blog-tags.html.twig
item-list--field-slider-images.html.twig
But, this is not rendered for the listing of the fields.
When I have created item-list.html.twig then only it will access.
However, both fields have different data to style and I am not able to get the current field name which is loading it's data in item-list.html.twig.
Had a brief look at this and it doesn't seem that 'item-list' to have suggestions, which is quite unfortunate.
In this situation there are two options:
Create your own suggestion which would accomplish exactly what you need.
You'll have to do something like this:
/
/*add new variable to theme with suggestion name*/
function hook_theme_registry_alter(&$theme_registry) {
$theme_registry['item_list']['variables']['suggestion'] = '';
}
//send a value to newly added variable to use it build the suggestion
function hook_ENTITY_TYPE_view(array &$build, $entity, $display, $view_mode) {
//add condition here if field exists or whatever, do the same for other field
$build['field_slider_images']['#suggestion'] = 'field_slider_images';
}
//use newly added variable to build suggestion
function hook_theme_suggestions_THEME_HOOK(array $variables) {//THEME_HOOK=item_list
$suggestions = array();
if(isset($variables['suggestion'])){
$suggestions[] = 'item_list__' . $variables['suggestion'];
}
return $suggestions;
}
Now you should be able to use item-list--field-slider-images.html.twig
Second option is to do what others in core did: use a new theme
function hook_ENTITY_TYPE_view(array &$build, $entity, $display, $view_mode) {
//add condition here if field exists or whatever, do the same for other field
$build['field_slider_images']['#theme'] = array(
'item_list',
'item_list__field_slider_images',
);
}
In CiviCRM webform, you can 'enable tag and groups'. Configuring those allows you to create option elements in the webform.
This creates one 'widget', one dropdown or set of checkboxes. I have two field instances where I want the user to select a group - say for example
which mailing lists do you want to receive (a,b,c)
what food are you interested in (d,e,f)
a,b,c,d,e and f are all groups. I can not change that.
How could I do that ?
A technical suggestion below, but first, I'd suggest that your real solution is to not use groups for the second question. Groups are set up nicely to handle mailing lists, but if it's to track interests, you'd be better off setting those up as custom fields. It'll solve this immediate issue, and it'll make it easier to deal with tandem searches and so forth (on list b and likes food d).
Now if you must have them as groups, you can create a fake field and move checkboxes into it using jQuery. Create the fake field with one option that you don't care about, but label it "What food are you interested in", or equivalent. Then, edit the Groups field that CiviCRM generated: label it more specifically as "which mailing lists...", and choose Static Options so it doesn't start offering up just any group for someone to choose.
Now, add the following javascript:
// first remove the dummy checkboxes in your fake field
$('#yourdummyfield .form-item').each( function() { $(this).remove(); });
// now move things into the dummy field
$('#yourdummyfield').append( $('#groupsfield .form-item-d');
$('#yourdummyfield').append( $('#groupsfield .form-item-e');
$('#yourdummyfield').append( $('#groupsfield .form-item-f');
From the form processing perspective, they'll all be evaluated as the "groups" field. However, they'll look separate. For better or worse, this will have to be adjusted as you add new groups fields.
After using Andrew Hunts suggestion for a while, I finally solved this on the server side, in a custom module, using webform logic as described here
http://www.webomelette.com/drupal-webform-submission-presave-hook
Basicly, on presave, I look for 2 custom fields containing group ids (mailing and food in the example). Then I add these to the CiviCRM groups field.
I'll add the code below, which has some more logic:
to make it flexible, I use one hidden field to contain the fieldkey
of the civicrm groups selector to add the other fields in. that
field is called 'the_groups_element' (but its not the groups element, it contains the key of the groups element)
there is only one foods group allowed, so before it adds you to a food group, it removes all other food groups from the groups selector.
You could probably make it even more generic, but since I had different logic for the different groups, this was suitable for me.
function getFoodGroups() {
// return foodgroups
}
function getMailGroups() {
// return mailgroups
}
function MYMODULE_webform_submission_presave($node, &$submission) {
$groupselm = '';
$groups_cid = false;
$foods_cid = false;
$mailings_cid = false;
// http://www.webomelette.com/drupal-webform-submission-presave-hook
foreach($node->webform['components'] as $cid=>$comp) {
if ($comp['form_key']=='the_groups_element') {
$groupselm = $comp['value'];
break;
}
}
if ($groupselm) {
foreach($node->webform['components'] as $cid=>$comp) {
if ($comp['form_key']==$groupselm) $groups_cid = $comp['cid'];
if ($comp['form_key']=='the_foods') $foods_cid = $comp['cid'];
if ($comp['form_key']=='the_mailings') $mailings_cid = $comp['cid'];
}
$group_gids = $submission->data[$groups_cid];
if (!$group_gids) $group_gids=array();
if ($foods_cid!==false && $submission->data[$foods_cid]) {
// remove all current foods
foreach ($group_gids as $gidx=>$group_gid) {
foreach (getFoodGroups() as $foodgroup) {
if ($group_gid==$foodgroup['gid']) {
if ($debug) drupal_set_message('removing foodgroup '.$foodgroup['gid']);
unset($group_gids[$gidx]);
}
}
}
// validate and add submitted regions
$foodsgids = $submission->data[$foods_cid];
if (!is_array($foodsgids)) $foodsgids = array($foodsgids);
foreach ($foodsgids as $foodsgid) {
foreach (getFoodGroups() as $foodgroup) {
if ($foodsgid==$foodgroup['gid']) {
$group_gids[]=$foodsgid;
break; // only one food allowed
}
}
}
}
if ($mailings_cid!==false && $submission->data[$mailings_cid]) {
// just add submitted mailings, dont remove any
$mailinggids = $submission->data[$mailings_cid];
if (!is_array($mailinggids)) $mailinggids = array($mailinggids);
foreach ($mailinggids as $mailinggid) {
foreach (getMailGroups() as $mailing) {
if ($mailinggid==$mailing['gid']) {
if ($debug) drupal_set_message('adding mailing '.$mailing['gid']);
$group_gids[]=$mailinggid;
}
}
}
}
$submission->data[$groups_cid] = array_unique($group_gids);
}
Using Grails (or hibernate), I was wanting to know if there is a specific design pattern or method we should be using when implementing a SEARCH of our domain.
For example, on my website, I want to be able to filter(or search) by multiple properties in the domain.
EG: For I have a page which displays a list of HOTELS. When I submit a search form, or if a user clicks "filter by name='blah'", when I enter the controller I get the following:
Domain
String name
String location
Controller
if(params.name && params.reference) {
// Find name/reference
} else if(params.name) {
// Find name
} else if(params.reference) {
// Find reference
} else {
// Find all
}
As you can understand, if there are more properties in the domain to search/filter, the longer the controller gets.
Any help. Please note, I do not want to use the 'searchable' plugin, as this is too complex for my needs.
I would embed these in a named query in the Domain class itself. For example:
Class Hotel {
String name
String city
String country
boolean isNice
static namedQueries = {
customSearch { p ->
if (p?.name) eq('name', p.name)
if (p?.city) eq('name', p.city)
if (p?.country) eq('name', p.country)
if (p?.isNice != null) eq('isNice', p.isNice)
}
}
}
Then later in a controller somewhere ...
def results = Hotel.customSearch(params)
Of course this is a very simple example, but you can expand on it using the same named query or even adding others and chaining them together.
I'm doin a simple query linq to retrieve a label from an optionSet. Looks like the formatted value for the option set is missing. Someone knows why is not getting generated?
Best Regards
Sorry for the unclear post. I discovered the problem, and the reason of the missing key as formattedvalue.
The issue is with the way you retrieve the property. With this query:
var invoiceDetails = from d in xrmService.InvoiceSet
where d.InvoiceId.Value.Equals(invId)
select new
{
name = d.Name,
paymenttermscode = d.PaymentTermsCode
}
I was retrieving the correct int value for the option set, but what i needed was only the text. I changed the query this way:
var invoiceDetails = from d in xrmService.InvoiceSet
where d.InvoiceId.Value.Equals(invId)
select new
{
name = d.Name,
paymenttermscode = d.FormattedValues["paymenttermscode"]
}
In this case I had an error stating that the key was not present. After many attempts, i tried to pass both the key value and the option set text, and that attempt worked just fine.
var invoiceDetails = from d in xrmService.InvoiceSet
where d.InvoiceId.Value.Equals(invId)
select new
{
name = d.Name,
paymenttermscode = d.PaymentTermsCode,
paymenttermscodeValue = d.FormattedValues["paymenttermscode"]
}
My guess is that to retrieve the correct text associated to that option set, in that specific entity, you need to retrieve the int value too.
I hope this will be helpful.
Best Regards
You're question is rather confusing for a couple reasons. I'm going to assume that what you mean when you say you're trying to "retrieve a label from an OptionSet" is that you're attempting to get the Text Value of a particular OptionSetValue and you're not querying the OptionSetMetadata directly to retrieve the actual LocalizedLabels text value. I'm also assuming "formatted value for the option set is missing" is referring to the FormattedValues collection. If these assumptions are correct, I refer you to this: CRM 2011 - Retrieving FormattedValues from joined entity
The option set metadata has to be queried.
Here is an extension method that I wrote:
public static class OrganizationServiceHelper
{
public static string GetOptionSetLabel(this IOrganizationService service, string optionSetName, int optionSetValue)
{
RetrieveOptionSetRequest retrieve = new RetrieveOptionSetRequest
{
Name = optionSetName
};
try
{
RetrieveOptionSetResponse response = (RetrieveOptionSetResponse)service.Execute(retrieve);
OptionSetMetadata metaData = (OptionSetMetadata)response.OptionSetMetadata;
return metaData.Options
.Where(o => o.Value == optionSetValue)
.Select(o => o.Label.UserLocalizedLabel.Label)
.FirstOrDefault();
}
catch { }
return null;
}
}
RetrieveOptionSetRequest and RetrieveOptionSetResponse are on Microsoft.Xrm.Sdk.Messages.
Call it like this:
string label = service.GetOptionSetLabel("wim_continent", 102730000);
If you are going to be querying the same option set multiple times, I recommend that you write a method that returns the OptionSetMetadata instead of the label; then query the OptionSetMetadata locally. Calling the above extension method multiple times will result in the same query being executed over and over.