Clicking a link created after init after clicking a link resolves old data

• Click a deeplink (A) on the web
• You’ll be redirected
• branch.init
• Get data of A
• branch.link (B)
• Click the created link B
• You’ll be redirected
• branch.init
• Get data of … A (expected B)

Refresh keeping the ?_branch_match_id=XXX keeps resolving A

• Remove ?_branch_match_id=XXX and press enter
• branch.init
• Get data of B

[sweetlikepete]

I’ve tracked down what I think is the issue. When the SDK is deciding to cache or refetch init data there is a check which is resolving to true when (presumably) it should not.

link_identifier === sessionData[‘click_id’]

sessionData[‘click_id’] is the cached identifier of the last visited link - that part is correct. link_identifier gets initialized from the ‘r’ hash, i.e. http://127.0.0.1:5000/linkage-c/#r:172663601188894407

The problem seems to be all links created through branch.link (that look like https://bnc.lt/l/95Kcs0EAY7) redirect to different urls, but with the hash.

The reason ?_branch_match_id=XXX resolves the issue is because it trumps the value of the ‘r’ hash.

(utils.getParamValue(‘_branch_match_id’) || utils.hashValue(‘r’))

My conclusion is that the error lives in the redirect logic on bnc.lt. It should generate a unique hash per url.

[jsaleigh]

The code has been merged and made live for a long time, closing this one out. Thanks team!