Setup
I have a Master-Detail relationship, and on the master, I sum up a field from the Detail object.
Question:
I need to allow the user to check a box on the master object which will add the sum by 1 or decrement by 1. How do I do this?
Here's what I've tried:
In an after update trigger on the master increment/decrement within the trigger. Of course I get an error:
expected exception, contact your administrator: NonCashCompAfterUpdate: execution of
AfterUpdate caused by:
Created a new field, and used the existing field as a helper. Basically, I've hidden the old RFS while still using it. This works, but I wanted to see if there was a better way.
Before I went any further I wanted to check in for ideas.
Thanks!
There is no way to directly alter values on a roll-up summary field in salesforce, except by altering the underlying detail records. The second proposal, as I understand it, would involve three fields, the roll-up summary, the checkbox, and the displayed summary (which would be a formula field). The displayed summary field would look something like this:
Rollup_Summary_Field__c + IF(Checkbox_Field__c, 1, -1)
The read-only nature of a roll-up summary field is preserved in Apex as well, so you would not be able to alter it through a trigger. Presumably, that is the nature of the error that you got, though the excerpt that you have attached omits the actual error.
Related
Hope you're well. I'm currently building out a report, but despite my best efforts so far, I can't get some information to populate within the report. It does not appear to me that salesforce is recognizing the field "Agent Incoming Connecting Time" within the object "AC_Agent_Performance". However, I can pull in some other fields within the same object into the Agent Performance report, so I'm not clear on what is not taking place in the field that I wish to see within the report. Here are some of the things that I've tried:
I have checked the access to the field. The first photo (Photo 1) Shows an example of a working object, the the second one shows an example of one that does not.
The API name seems to work, and is consistent with other fields within the object that work.
I have checked the page layout for the object (even though I don't think this is the issue), and I have mirrored other fields to the best of my knowledge that ARE populating within the report.
I reviewed the CTI flows to see if there was something missing in there on a lark, but there was nothing in there that would lead me to believe that this was the source of the problem.
I have tried setting up a new field in the object (formula), that references the field that I'm trying. to pull in, but that just returns a result of 'zero' for all values.
One thing that I have done that appears to be working, is I have set up a joined report, which uses both "AC Agent Performance" object and "AC Historical Queue Metrics" object in the report. The result that is returning appears to be accurate (please see the picture (picture number 3)). However, I don't think that this is the right way to go about this, and I don't want to do it this way. I want to use the report with one object rather than with two.
I know that permissions are the most likely issue, so I've taken a close look at these. Please let me know if there is something wrong with how I have the permissions configured. The First image depicts the 'Field Level Security'. The second image depicts the'field accessibility'. They are both like this, the whole way down:
Please note one other thing, which is that the last picture depicts a different field within the object displaying in the report.
Does anyone have any ideas on how I can proceed so the field "Agent Incoming Connecting Time" will display within the report?
Please also note, that these are objects that contain data that is populated from AWS' Amazon Connect.
This last photo, shows that the object does not have any information in it within the report.
If the field isn't populated there's not much you can do on the reporting side of things. You already tried "joined report". You should check why the integration doesn't populate it, maybe read integration documentation, contact the managed package's support...
The tables are connected with lookup or master-detail, right? In a pinch you could try making formula field on "AC Agent Performance" looking "up" and pulling the value from related AC historical queue metrics. If the relationship is other way around (performance -> down to related list -> metrics) you could try to make-do with a master detail and rollup summary field. I don't know this package, no idea if you can pull it off when you don't have full control over the fields.
If you can't really use the relationships and absolutely need to report on single table - you could capture intermediate results of the report to a helper table and then report on that. It's called "reporting snapshots". Or write some nightly (hourly?) batch that recalculates stuff and writes homemade "rollup" to these fields?
I have a custom object. There are combination of fields like month_c,Project_c,contact_c and role_c which determine the record as unique.
I can write a trigger on before insert to check if there are any records with the same combination already existing. The question i wanted to ask was
how can i make the insertion stop. once i find there is already an record then it should just not insert the record. it doesnt need to throw / show an error.
Thanks
Prady
Although others have answered this with better solutions (generally non-code solutions are more flexible for the end users of the system), the answer to stop a particular record from being inserted is to add an error to a field on that record.
For instance, if I was inserting accounts I might have something like this:
for(Account sAcct : trigger.new)
{
if(sAcct.Name == 'DeniedAccount')
{
sAcct.Name.addError('This is a denied account.');
}
}
Another alternative is to create a class extending Exception, and then throw a new instance of that exception class, this will prevent all of the records passed to the trigger from being processed as opposed to specific records.
A few days ago I found an incredibly useful and simple solution (in a ForceTree.com article) that prevented me from having to write a trigger. It allows you to combine fields to check for uniqueness using a workflow rule and a custom field.
Here's a walk-through:
http://www.forcetree.com/2010/07/unique-field-combination-in-salesforce.html
I have a similar situation in Salesforce. It is handled by creating a field that contains a value composed of all the values needed to guarantee uniqueness (in your case, month_c,Project_c,contact_c and role_c ). You then select the "Unique" checkbox for that field. Duplicates will now end up in the trash.
In my case, this new field is filled in (and pushed into Salesforce) by an external program. In your case, you'll need to fill in the value in your trigger. I think this is more efficient than doing SOQL queries in a trigger, but I have not done any checking to confirm this.
I have a SQL database where one of my fields 'Status' has one of 10 entries in it (not from a dropdown list). I let some people have access to the data through an Access database, they can add and modify entries.
My question is, in Access, how can I make it so that the 'Status' field is a drop-down list from which the users can choose from (the 10 entries which are already in the SQL database)? It would be easier for them and also mean that mistakes cannot be made.
Many thanks
Scott
The usual way to do this is to use a combo box on a form with the row source taken from the look-up table and the bound column set to the field (column) of the table to be updated.
In Access you can add lookup information to a column. That will automatically display a dropdown list.
Step 1: Start the lookup wizard:
Step 2: After the wizard, the lookup settings should look like this:
Step 3: When your users open a table, they should see the dropdown box:
In addition to the solution described by Andomar you must not use another table as the source for your lookup. You can also provide the lookup-values in a list, which is hardcoded in the table-definition. This is fine for simple scenarios where the lookup is something that is not likely to be changed.
Several issues here:
table datasheets are not suitable user interface for users.
you can create a saved QueryDef and if you view the properties of a field, the second tab is just like the second tab in table design, and allows you to use a combo box as your display type in your query. I would generally recommend against this, as, like table datasheets, a QueryDef is not a proper UI element, and if you use the saved query in other queries, you can run into the same problems that cause lookups in table fields to be such as bad idea.
you're building a UI, so use the tools that Access provides you for building a UI. That means a form. Your form's recordsource would have the bare data, and you'd create a combo box on your form that is bound to the field in the table behind your form, and displays the values from the lookup tables. There's a wizard to step you through this. If you like the look of datasheets so much (column sizing, sorting, show/hiding are all features that are nice in datasheets), you can set your form to display as a datasheet.
My advice is that for building a user interface, use the tools Access provides for creating user interface. In my opinion, a dropdown list belongs in a form, and nowhere else. While I occasionally might add one to a query for quick-and-dirty editing, I would never do that in objects that users are going to use.
I am getting the following error in an MS Access ADP when trying to add a record on a form linked to a MS SQL Server 2000 table:
Run-time error '31004':
The value of an (AutoNumber) field
cannot be retrived prior to being
saved.
Please save the record that contains
the (AutoNumber) field prior to
performing this action.
note: retrieved is actually spelled wrong in the error.
Does anyone know what this means?
I've done a web search and was only able to find the answer at a certain site that only experts have access to.
First of all, if you are going to look at experts-exchange - do it in FireFox, you'll see the unblocked answers at the bottom of the page.
Second, do you have a subform on that form that's using the autonumber/key field on the master form? Do you require the data that's on that subform to be saved (i.e., having its own key) before the main form is saved. You could be into a deadlock of A and B requiring each other to be saved first.
Other than that, you must somehow be accessing that autonumber field whenyou are saving it. The best I can suggest is to step through the code line by line.
Are you trying to assign the value of an Identity field to a variable or something else before you have saved the record?
For whatever reason, your app is trying to read the value of the identity field before the record has been saved, which is what generates that identity field. In other words, no value exists for the Autonumber field until the row is saved.
I think we'd need to see more code or know more about the steps that lead up to this error to resolve it in more detail.
You should have add some lines of code to show us how you're managing your data and what you are doing exactly. But I am suspecting an issue related to a recordset update. can you identify when the autonumber value is created? Is it available in a control on a form? Can you add a control to display this value to check how it is generated when adding a new record? Is the underlying recordset properly updated? Can you add something like me.recordset.update on some form events: I would try the OnCurrent one ...
I asked about setting up an in-memory dataset with TClientDataset, and was told I had to use the "Create Data Set" command in the form designer. That works just fine until I try to create a master-detail database relation. If I have a field def of type ftDataSet, then running the Create Data Set command (or calling the method at runtime) gives an "Invalid Field Type" error. I did a few searches and found that other people have had this problem before, but nobody ever seems to come up with an answer. Does anyone know how to fix this?
Instead of putting your detail in a ftDataSet column, just use a unique integer or GUID for the key reference, and place the detail data in a different TClientDataSet and have that key. Like you would in a normal database. Each TClientDataSet representing a different table.
If you create a field of type ftDataSet, you need to define the dataset, else the type is not valid.
You can define the dataset, by:
select the field.
select the ChildDefs property and click the [...] button.
add the fields.
Now you can create the dataset. Unless one of the children is of type ftDataSet in which case you have to recursively repeat the process.
A nasty problem that cost me a lot of time already. Finally I found a workaround, described here (Thanks you guys!):
http://www.delphigroups.info/3/6/171869.html
Summary: The culprit are the field definitions of the master client dataset, i.e., the one that contains the ftDataset fields. As soon as the field definitions of the master client dataset are not empty, the command CreateDataSet fails with the above described error message ("Invalid Field Type").
Workaround:
Create master-detail tables as instructed, but make sure the field definitions are empty when doing CreateDataset:
a) Designtime: Before you do "master dataset | RightClick | Create DataSet" in the design form, you have to click on the master dataset, go to object inspector, click on FieldDefs, and delete all the fields in the FieldDefs window. (The very first time you built the master dataset there are none in there anyway, but later there are.) Only then do Create Dataset. (I tried this myself with Delphi 2007; it works.)
b) Runtime: Call [Name of Masterdataset].FieldDefs.Clear before you do [Name of Masterdataset].CreateDataSet. (I did not try this out myself, but it is credibly described in the above named link.)
If this was too complicated, please look in the above mentioned link.