I am developing a web based back-end application. In this back-end, I can View, Add, Edit, and Delete information. There is a form and a table positioned under the form.
First, the application will load information into the table. When I choose to edit a row, I click an "Edit" link in the row and the information will be loaded into the form. Then I can change the information and save it.
In my Activity Diagram for "Edit", should I start from the user clicking the "Edit" link or start from loading information to the table?
I am a little confused because this loading information step is already included in "View" Activity Diagram.
well i think it depends.if you draw all the roles in partitions then you can use "the loading action" only once and show transition from this to other activities like if you create partitions for view,edit and delete link then you can show loading of the table only once and then show transition from it directly to "edit row" activity in edit link partition and likewise for other activities too.needless to say that you have to use branching accordingly.you could have posted the rough diagram you created so that improvements could be made.
Related
Not sure if this is the correct way to represent the flow between the "Home" page button, "Culture" page button and the "Song" page button (as all these buttons are present in all of the pages, where the user is able to click any of these buttons and perform certain activities)
This seems very messy as the arrows are pointing in all direction, which is why I am stuck on how to represent this.
UI of the application:
Why did you choose an activity diagram for this purpose? A state machine diagram is more appropriate. Clicking on a menu option transfers the application to a different state. You can use a superstate to model the transitions that can be made in any application state.
I would use an activity diagram to visualize the paths through a complex use case, not all UI actions.
See my paper 'Which UML models should we make', pages 12 and 13.
I'm new to Domino Designer. I'm trying to modify an existing application which is running on a remote server. I create a blank xPage and view in browser and it works. I add a label and view in browser and it works. I add a combo box and setup a data source for my xPage. The data source is in the current application and I'm using a view. I setup my combo to point to that data source and specify BindTo value. I view it in browser and I get Error 500 "HTTP Web Server: Command Not Handled Exception" I can open the view that I've used and see all the data in the view. I can even add, edit, and delete from the view.
Ideas... Thnxs in advance.
Additional Info. I've added a "View" from "Container Controls" and used the same data source and the same view and it also works in the browser.
The error being printed is:
javax.faces.el.PropertyNotFoundException: Error getting property 'CompanyName' from bean of type lotus.domino.local.View
com.sun.faces.el.PropertyResolverImpl.getValue(PropertyResolverImpl.java:119)
com.ibm.xsp.el.PropertyResolverImpl.getValue(PropertyResolverImpl.java:144)
com.sun.faces.el.impl.ArraySuffix.evaluate(ArraySuffix.java:182)
com.sun.faces.el.impl.ComplexValue.evaluate(ComplexValue.java:163)
com.sun.faces.el.impl.ExpressionEvaluatorImpl.evaluate(ExpressionEvaluatorImpl.java:257)
com.sun.faces.el.ValueBindingImpl.getValue(ValueBindingImpl.java:150) ....
This was already answered in the comments, but I want to write a better answer.
The problem in this case is that a view data source is being used for editing a document. Directly, this is not a possibility, though there are a few great ways to get around this. The first way, and my favorite, is to use an inline form. I think the Extension Library Demo database offers a few great examples of how to go about doing this. One way is to use a tooltip dialog. Another way is an inline form. A repeater is used and for every row in the view, a hidden panel can be made which contains the document data source and all field which are required to be changeable. By clicking on a button or link, the panel is made visible. This is my favorite variant because it is simple, quick, and generally looks great.
Another way to go about this is to have a view data source only, but use scripts to retrieve the document and set the values in the background and then save and release. It all depends on what you are trying to achieve.
My main recommendation is to look at the ExtLib demo and take a look at what they are doing there.
Background: a third party web application with a requirement on the order page to explicitly guide staff when a certain category of product is ordered.
Normal process:
select product to add order line (click on product image)
New process:
select product to add order line (click on product image)
[add on] simulate click on edit order line
[add on] simulate clear of default additional information form fields
enter additional information and click save
Note that the two steps to be performed by the add-on could be performed by the user however when things get busy or new staff begin is often the case that the order gets processed with the default value.
Any pointers in the right direction would be appreciated.
A multi-browser solution is preferred but not essential.
I am an experienced developer, including web development, but have no experience in browser add on/extension development.
Thanks.
You can make an addon out of this very easily. You addEventListener to the gBrowser and if the event.target is the product element then preventDefault so the click doesnt go thru (you will have to do mousedown and mouseup prevent as well). Then rather get the id of the "edit order line" and do .click() if that doesnt work and you really need to simulate use MouseEvent:
MDN :: MouseEvent
MDN :: Initatite/Dispatch MouseEvent
To enter aditional information get the id of the fields, and set their value. Then get the "Save" button id and then do .submit() or .click()
i said do get id in this post but you can use querySelector and go by css selector
see this page here on how to interact with content in currently selected tab:
https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Tabbed_browser
basically just do gBrowser.contentDocument.getElementById
Background: Xpage with multiple data sources. Each row of tabbed table contains an edit panel bound to a different data source than the main page. Each page contains a table to enter new documents. These documents are children of the main document. There is a view control that shows the user created documents immediately. The button in the table saves the document in addition to adding some control fields. It does a partial refresh of the editPanel.
Recently I added validation to the fields in the entry table. I use a function, and prevent updating from anything but the button. I am using Tommy Valand's method from this blog post: http://dontpanic82.blogspot.com/2010/03/xpages-making-validation-behave.html. If I don't use Tommy's method, then the validators fire when switching tabs, and are useless.
All of this works sort of, but I am experiencing the following issues.
1) I have to push the button twice to get the partial refresh to fire. Not only that, but the validation messages to not fire until the button is pushed twice. The first part was still an issue prior to adding the validation.
2) The table values do not clear after the refresh does happen. This used to work prior to adding the validation. The only way to get the values to clear now, is to do a full refresh (F5).
I really just need partial refresh to behave like it should. Any help is greatly appreciated.
I am answering my own question because I think that I might have found out why the partial refresh was acting so flaky. Here is what I did that solved my issues: Each tab had an editPanel. If I make the edit panel do a partial refresh to itself then the buttons on that panel seems to behave as you would expect them to.
When I create a new core control panel, the default in No submission. I do not know whether this issue manifests itself only when using multiple data sources and/or tabbed tables. Hopefully this fix will be of use to someone down the road. It is worth a try if you are having issues with partial refresh working correctly.
this is not a technical question, but one for advice regarding the best practices in designing an Android tablet UI.
I've got my concept of an Android Phone app pinned down.
The first activity (master view) launched contains a tab bar with three fragments from which the user can launch detail view activities of different sorts.
Both the master-view activity and the detail-view activities have actions in their action bars. Different detail views have different action items.
My question is: How should I organize and display the action items on a tablet, where an activity combines both views side by side?
The problem is the unified action bar for both the master-view fragment and whatever kind of detail fragment is shown. I do not think it is a good idea to start messing with the contents of the action bar whenever a different kind of detail view is opened.
The Android Design Guide does not tell you much on that front. There is a sample of a Contacts app in the "Multi-pane Layouts" section, but it does not actually deal with the problem. It evades it, by putting the single relevant action as an icon inside the detail view fragment.
Any advice with regards to best practices and references are appreciated.
I would suggest leaving your master details icons in the action bar, whilst putting your details view icons in another view/area within the details fragment.
My reasoning would be that icons in the action bar affect / are associated with the whole app / view on screen. Whilst your details icons only affect the details view and therefor should not be in the action menu when showing multiple fragments.
I guess you will have to see how the designs look..
I am not a fan of the action bar icons being changed from within the same activity (even if it contains multiple fragments), however when you load a new activity (like in your phone design) then I say yeah throw them in the action bar.
If I understand your question correctly, which I think is basically a question of how multiple Fragments (i.e. when on a multi-pane layout such as on a tablet) should contribute to the single ActionBar, it's quite straightforward and it is actually briefly discussed in the documentation here. Essentially, you can have the multiple Fragments all contributing their own menu items / action items to the single action bar, via some simple API calls.