BACKGROUND
I have an app that was set up using a default branch domain (ex: my.app.link).
Links were created using my.app.link. Many of these links needed to mirror existing links I have in production (with another linking service that we are migrating off of) that use a format not supported in Branch.io, so I had to work around the issue by using a '-' in place of '.' in the link's path. This is a solution recommended by Branch with the assurance that the '.' would be translated to a '-' on the backend at Branch.io (this appears to work fine).
I recently had a custom subdomain CNAMEd over to custom.bnc.lt (ex: myapp.mydomain.com).
The link settings in the Branch.io dashboard have been updated to use the custom subdomain (change is valid: "Domain is correctly set up and SSL is working.").
ISSUES
All pre-existing quick links still have the old default domain, no clicks register even though they 'work' with the new subdomain.
Duplicating one of these links picks up the new subdomain, but the url that gets created is http, not https (which is the default for any other link we created and there doesn't seem to be a way to choose one protocol or the other, nor can it be updated in the dashboard). This also seems to be causing link clicks to not track in the dashboard.
The fact that we mirrored links from another platform using a workaround (replacing '.' for a '-'), also seems to cause link clicks to not track in the dashboard.
QUESTIONS
How does Branch.io handle link clicks for a quick link that has been set up for an app using a custom subdomain? Does the configured link need to match 1:1 to the link that is clicked on the device?
How can I fix the original links, especially if I can't duplicate them (to pick up the custom subdomain) in a way that gives me a 1:1 match to the previous link I created (since the new link is http and the others are https, as would be distributed in a live/production environment)?
Why isn't Branch tracking link clicks for the x.y.z links that I set up as x-y-z? This aliasing works fine, and seemed to be tracking clicks fine in test.
Amruta from Branch.io here:
1. How does Branch.io handle link clicks for a quick link that has been set up for an app using a custom subdomain?
A: For all Branch links, every time a Branch link is clicked the click is is registered by Branch. Using a custom domain (if configured properly) should not cause any problems in registering clicks on the Branch dashboard neither should using Quick links cause any issues. Also Branch links with either schemes http:// or https:// register link clicks. Having Branch links with http:// scheme should not cause any issues with registering link clicks
2.All pre-existing quick links still have the old default domain
A: The old links created on the app.link will not be updated with your custom domain. If you wish to use links with your link domain, you can create new links with the same link parameters. (PS: both links on the old app.link domain and links on your custom domain should work as before if you still have your App configured to support the app.link domain links.)
3. Duplicating one of these links picks up the new subdomain, but the url that gets created is http, not https
A: At present, the Branch API creates Branch links for custom domains with the 'http://'scheme. This is because we are not sure if custom domains have the SSL certificate and creating links with the https:// scheme would break for such domains. Hence by default, all custom domain links are created using the http:// scheme. There is no direct way to create links using the https:// scheme using the Branch API, the only way you can do it is by manually changing the scheme from 'http' to 'https://'.
4. Why isn't Branch tracking link clicks for the x.y.z links that I set up as x-y-z? This aliasing works fine, and seemed to be tracking clicks fine in test.
I believe, the issue here has nothing to do with your custom domain but is maybe related to a key mismatch. You mention that the link clicks were tracked in the test version of the App, which leads me to believe that your app is still using the Branch key from the test version (key format: key_test_XXXXXXXXXXXXXXX). Please ensure that your App is using the correct key. The simple test for this: When you click on the Branch link, you get the link parameters in your app and the link parameters contain +clicked_branch_link:true. If so, the link click will be tracked on Branch's end.
If you still continue to face issues, you could write to integrations#branch.io and our team would be more than happy to help you resolve the issue.
Related
Can't navigate a specific page by url
Hello everyone, I have a little problem, I had a "boilerplate" project with Vue3 + Laravel. And I wanted to use link so people can navigate specific page, for example here's a link you can navigate
https://klim2020.github.io/
After that you are able to open portfolio page by clicking on Portfolio link at the right top of the page. As you see you have that
/cv/en
adress at the address bar
But if you will try to navigate this link from other source you will get Error
Cannot GET /cv/en
try to navigate from here https://klim2020.github.io/cv/en
Why is this happening can you tell me please?
First off, I do recommend hosting on Netlify or Vercel, those will fix the issue that you do have here.
GH pages are just not worth it. And the other 2 services are more flexible, as free and quite powerful.
The issue here being that if you go to a specific path with a "hard navigation", not a client-side one (via vue-router), you SPA will not be in that specific endpoint. Hence why, you will get a 404 (your initial entry point is available a /).
You also need to use the history mode rather than the current # that you do have in your path.
Best thing? The part explaining how to solve your current issue with Netlify/Vercel is also explained in a section below (mainly providing a redirects/rewrites configuration).
I have the following requirement and based on the branch docs I cannot get myself a straight answer.
We want to enable deep links that if the user has the app installed that it uses them and if not redirects to download then redirects.
If the user is on a desktop they get redirected to a page to download app.
We want to control all the actual deep linking using react navigation config and wanted to know how we can pair the two!? The guides are very poorly written and makes understanding the flow a nightmare.
Some guidance and suggestions on how best to accomplish this would be great.
P.s. are the links generated that when app found it takes the prefix and replaces with appname:// ??
Your requirement for deep linking can be handled with Branch, you can set the URI schemes of the apps under the link settings of your Branch dashboard and also mention your app on PlayStore/Appstore for the user to download the app when clicked on the link.Similarly you can add the redirection URLs for the scenarios when your link is clicked on a desktop.
Alternatively you can use our React SDK to create links with all these link properties and even more. More details here.
The handling of the deep link data and redirecting the user to the desired page can be found here.
If you wish to know more about how Branch passes data through to the app and attributes app sessions, check this.
We already set up a branch project, with branch links and it works fine, is it possible to have 2 different behaviors on 2 links on branch?
i am talking about the behavior of the desktop click:
for example, i want that one link will open site A, and another will open site B (not under the same domain)
or to have one link redirect to a web page, and another to be redirected to the Branch-hosted SMS Landing Page
couldn't find the option on branch.io link settings
thanks.
Full disclosure, I work at Branch. You can definitely have each of your Branch links redirect to different sites.
For a Quick link, you need to update the Desktop redirect to the web address you want the link to redirect to.
For the Branch links you create in the Mobile SDK or using the HTTP API, you can add the $desktop_url:"http://www.example.com" parameter to the links.
To learn more about Deep links, you can refer to the documentation here.
I'm trying to authenticate a desktop application with OAuth. I already asked a question, but I deleted it as I discovered that the method was obsolete (integrated web view).
The idea is to open the browser and then have the page call back my application with a custom:// URI scheme.
Of course it doesn't work, Google asks that the scheme contains a dot (like com.org.myapp), but the native app can only have a fmp:// uri scheme, which I can't change.
The documentation doesn't mention anything, and I can't create a local temporary server.
1- Can I use the fmp:// scheme as a callback in any way?
2- If not, is it possible to redirect the call to another centralised web server? Everything I try generates a 400 error on the google page, saying "redirect_uri_mismatch" or similar stuff.
As stated in the comments above, the only solution is to create an intermediate web server that serves a page which then opens the URL. OR creates a record in the database with the authentication code.
This issue began last week. Prior to that, I was not having any problems, and I am not aware of any changes to my site or platform.
Now, when I want to share a new blog or article to Google+, it fails. Typically I would use the embedded +1 buttons on a specific post and use the expanded box to create my post. The box comes up, but instead of a nice title and image, it now uses the URL only for the page, and sometimes adds extra characters to the end of the URL. If I take the URL from the address bar and try to compose a new update directly on Google+, I get the same issue. If I enter the URL into the Link field on a status update, it usually comes back "could not load website."
Here's a sample
You'll find Google+ sharing buttons above and below the article. Sharing to every other network works as expected.
My site is a Drupal site that has been operating for 10 months. I am a Drupal developer, but have never encountered an issue like this.
It appears that requests from the Googler crawler are being rejected by your server. I tried testing the microdata with the structured data testing tool and it runs into problems connecting to your site. Other sites work fine.
If you have access to your site's Apache access logs, I would check those for problems coming to that URL. You can narrow your search down by looking for the user agent of the crawler: Google (+https://developers.google.com/+/web/snippet/)
My guess is that something changed in your server's configuration that is the cause. Start with the logs and see what's there.