I'm using SmartGwt's RestDataSource for interacting with a non java, non smart gwt back end. I have a a ListGrid which fetches and updates it's data using GET. This sometimes fails because of the size of the URL. How can I change it to use POST?
Related
Background:
I'm using DataTables in conjunction with a JS library called "Turbolinks", which basically turns your application into a Single Page Application (SPA) without all the overhead of using a true client-side framework. It is extremely useful for Ruby on Rails application performance.
There's a couple of headaches it introduces though - one is compatibility with DataTables. I've got it working pretty well by basically destroying any DataTable on a Turbolinks navigation, and then re-initializing it on turbolinks page load again. This method works well and seems to be the all-around accepted answer as to the best practice to get DataTables to work with Turbolinks.
Question:
On of the last features / finishing touches I'm trying to add to some of my applications is DataTable state saving. The issue I'm facing is that every time a table is destroyed/re-initialized on a page navigation, the...I'm actually not quite sure what to call it, but it looks like from inspecting the settings object on the stateSaveCallback - it looks like its the sInstance and/or the sTableId:
DataTables_Table_0
Then the localStorage key gets set as:
DataTables_DataTables_Table_0_/current_path: "{data: data}"
where current_path is whatever path/page you're on.
Then when it get re-initialized upon returning to the page, it gets set as DataTables_Table_1, and so on and so forth - so the state never gets correctly loaded.
Is there a way to override that ID (or some way to set the name of it in the stateSaveCallback / stateLoadCallback) so that it doesn't increase the last '0', '1', etc at the end of it? That way when the table is re-initialized, it will pull the saved state from just DataTables_Table/current_path?
The answer is to simply give the table an ID! Then DataTables won't assign it its own ID with the incrementing number and the saveState option just works.
Also, the destroy/re-init actually causes the server to get hit twice in the case of an AJAX table.
The better way to do it is to disable the turbolinks cache for any index pages with datatables. If not, you'll end up doing two requests to the server when only one is needed.
I have a large amount of text information which I'll be loading in a table.
I want the user to search through it, using datatables search.
I wasn't doing lazy loading earlier, but now I'm thinking of using it now.
However, if I lazy load and the user searches for data, he/she won't be able to see everything, since the data isn't completely loaded.
Am I guess this correctly or does datatables work around this somehow ?
There are two processing modes: client-side and server-side, see Processing modes for more information.
Client-side processing - the full data set is loaded up-front and searching/filtering/pagination is done in the browser.
Server-side processing - an Ajax request is made for every table redraw, with only the data required for each display returned. Searching/filtering/pagination is performed on the server.
There is also Scroller extension, it's a virtual rendering plug-in for DataTables which allows large datasets to be drawn on screen very quickly.
Along with server-side processing it could be used to perform lazy loading of large amount of data. When searching, the request will be made to the server to search the whole dataset and only subset needed to be displayed at the moment will be returned.
See Server-side processing for more information on request and response in server-side processing mode.
I have this requirement : My WL application have a set of static pages that might be updated any time. Originally the source of all static content is a desktop page that will be transformed by xsl to a mobile friendly content. The problem that I don't want to do that on each request (HA requirement).
I want to get some inspiration on how to architect that without using direct update mechanism (don't want the end user to get notified of these updates).
I should note that pages will change rarely every few month maybe.
I'm thinking about 2 ways of doing that :
1- Making the transformation on adapter side and rely on WL caching so that transformation is not made each time (does that exist ?). But how the adapter will get notified of page change and flush the cache ? Should I program some advanced java based adapter ? (Storing in the cache and having a kind of a job that scans every day for content changes ?)
2- Doing it mobile side but I don't know how to get notified of changes !
Is your only problem with Worklight's Direct Update that the user is being notified and is required to explicitly approve the transfer?
In this case why not use the option of Silent Direct Update?
The property you're looking for is updateSliently set to true in initOptions.js.
For this to work it is required, obviously, that connectOnStartup will be set to true as well.
perhaps what is doable is to use an adapter to fetch the HTML (or whatever it is) and save it to the device's local storage and then have the app display this content, this way you do not alter the app's web resources and not trigger Direct Update.
i have an action which will invoke a service (not database)to get some data for display,and i want to do paging on these data.however,every time a second page is clicked,it will invoke this action and of course invoke the service again,actually when i click the first page link,it already generate the whole data including what the second page needs. i just want to invoke the service once and get all the data,and later when paging,i don't need to invoke the service again,how can i deal with that?hope someone could give me a hint~
There are several ways to address this. If it's practical and a limited amount of data, it's ok to return the entire data set in the first request.
If that's your case I would consider returning a pure JSON object when you load the page initially. You can then deserialize this into a JS object variable on the web page that you can perform your paging operations against. This is an example of client side paging where all the data exists client side.
Another approach is to do Ajax based paging where you request the data for the next page as needed. I would still recommend returning JSON in this scenario as well.
The two approaches differ in that the first one returns all the data upfront, whereas the second one only returns what you need to render any given page.
So, I've read a lot about using ExtJS's fileuploadfield to submit a form via an IFRAME. I understand that I'm supposed to reply with a JSON object indicating success or failure; fine. What I want to know is, how can I get more information back to the calling code? I don't want to simple send a file and say "yup, that worked fine" -- I want to submit a document, act on it, and return a result.
Say I have the user upload an XML document -- I might want to do a lookup or conversion based on it and update the contents of a form on my page accordingly. Is this even possible? I'd strongly prefer to avoid involving Flash or embedded applets if at all possible. If need be, I could even restrict this behavior to HTML5-compliant browsers...
I honestly thought I wasn't seeing the response I sent, but it was a server-side error. My success callback is now firing, with the full text of my server's response available as f.responseText (where f is the first argument to the success callback). My mistake!