I am trying to develop a solution to following problem. I need to store in db information about time when user logged in and is on page. Currently I am writing to db when user login and logout with WCF service, but how to deal with situation when user closes window or goes to other webpage.
I am wondering if threaded function which calls every user every minute to check if he's alive is a good solution. Any help will be nice. Thanks.
If You can wait for data a bit(depending on Your aplication usage), You could save data to IsolatedStorage, and send it when user starts application again. It's pretty simple solution, but You will have to wait for data and some data will be lost, if user don't open application again(Again, depends on Your app).
Other solution would be sending data from JavaScript (How to call WCF from JS) during OnUnload or OnBeforeUnload event. Or even doing a simple HttpRequest from JS to some aspx site, passing time in query.
EDIT: Another thread is a nice idea(I have solution like this in my current project) but running it too often can clog IIS (depend on number of users, bandwidth etc). It also will prevent Session from timing out, even if user does nothing (that's main purpose for using this solution in my project).
Related
I am a bit dumbfounded on how to achieve this properly. So you have an app that is getting regularly updated, and you show those new updates to the user (something like discord update popup and whats new and such). How do you handle whether or not the user has seen it? A more high level question that "local storage", I am well aware of that or storing it in a DB, but this doesn't seem 1) Scalable, 2) Easy to work with in the long run. Are there any JS libraries that can simplify the process, eg or do I have to manually track each update? If there are more updates or the user hasn't logged in a while
I hope I conveyed what I am working on properly. Would love to hear how it's usually done.
Q: have an app that is getting regularly updated, and you show those new updates to the user
A: Service Worker can help you .That support reacting to push message .And they provide the Push API gives web applications the ability to receive messages pushed to them from a server, whether or not the web app is in the foreground, or even currently loaded, on a user agent.
More detail about start up a service worker to send users a message . https://developer.mozilla.org/en-US/docs/Web/API/Push_API .
Finally I suggest you try Workbox , that is a library that bakes in a set of best practices when working with service workers.
I'm working on an AngularJS 1.6 web application.
There are a number of AngularJS services in the web application. I'm beginning to implement features that use the services to persist state between views.
However, this is becoming problematic when the user logs off and another user logs on. When the new user logs on, the services persist state from the previously user. I know that I can theoretically write code to put the application in a specific state when a user logs in, but it seems to me like it would be easier to simply dispose the stateful services and instantiate new ones when a new user logs on.
Is there an expedient way to do this?
The expedient way to do this would be when a user logs out, do a hard refresh and send them to the login page. This should wipe data that is stored in memory across the whole application.
If you dont want to perform a hard refresh, you're just going to have to clear out the data manually for each service, which depending on the amount, could be a fair bit of work and most definitely not expedient.
I am curious to know about the real-world solution for dealing with forms that have large amount of data/fields or a wizard like interface (may be using AngularJS kind of GUI framework), especially if we want to take care of a scenario where the data persistence on back-end fails.
My questions are:
Is the form data saved in session, in the interim, (or may be on Browser itself using the JS libraries) till the user saves the final "Save" button?
Or is the data saved each time (i.e. on the user traversing from one screen to another using "Previous" or "next" buttons) on to a back-end database?
What happens if the form data has to be sent to an external web-service (instead of a database) and the call fails (due to timeout or any error)?
There is a strong chance that we will lose all the user entered data (unless we save it in a local database and re-try the web-service call later).
Do any of the caching f/w have any role to play here (including any AngularJS caching f/w)?
Thanks for sharing your knowledge.
I am building an sample application that lets user store comments.
I've created the registration and login process. When the user registers, his details are stored in a MySQL database and a token is returned to the browser. Now he can access the Profile page.
When an existing user logs in he is redirected to the profile page. The profile page is accessible only when a user registers or logs in.
After logging in, I want to show all his comments if he has already added them.
My frontend is in Angular and backend use Laravel. For authentication I use Satellizer.
I want to know, what is the best approach while playing with data, considering the fact that the user will add, edit his comments. Should I use localstorage and store data in a key value pair or should I create a json file which gets updated everytime the user adds a comment or makes a change.
I wanted to know what is the most efficient way to deal with data from server so that the application is fast even when it scales to a 10000 users and lot of data for each user.
Thanks
You should be updating it on the server when changes are made rather than only relying on localstorage. You can use localstorage to cache, but it should only be for immutable data, it shouldn't really be used for data that is going to change.
So in this case you'll be adding and updating new comments via your API (ideally a RESTful one!). Once you've made a change, you could store the comments locally and only update them when the user makes a new comment, however you'll quickly run into issues where the data is invalid on different clients. (i.e. if you update the comments on a different computer, the other computer won't be aware).
Alternatively, you could cache the comments and then simply ping the server to find out if new comments have been added. This could be using a HEAD request for example to check the last modified date on your comments resource.
You can store comments data locally on user browser, but you should properly manage it.
I don't how much load your server will have and if the time invested now worths it.
You can fetch comments and store them locally
User adds a comment, then you update locally and send a request to the server
You need to track the request response, if requests fail so notify user and remove comments from local.
if request was successful so you can continue on your way.
** facebook uses this "success first" approach
user does an action, and he see it happens instantly, in the background it could take few seconds, only if it fails they will notify you.
** look at their commenting process, when you comment, it appears instantly, no loading... but in the BG the load happens.
I need my (python) google app to perform an action (submit a form) if the user logs out. This is simple enough to do if they use the logout links in my app, but if they log out from a gmail page or something, I don't know how to handle it.
Another possible source of error would be if the user closed the browser window, shut down their computer, etc. resulting in a log-off. Is this scenario is equivalent to what I describe in the previous paragraph, or are they different somehow?
To expand this question since it seems the above is not at all trivial: if I set a cleanup function on a timeout, will the python session in fact continue to run in the GAE cloud after the cookie expires, and actually execute the timeout function?
Close browser window and shut down computer result log out because of session expired (cookie). It is slightly different from user click log out manually.
In both case, I don't think GAE can track these behaviors.
The best thing that I can think about is to develop a browser extension.
Or just don't design the service based on detecting user's log out.