.png){kind=logo}
Issue with HTTP API Enrichment: Failed to Parse Body in Clay
I think there is something wrong with the http api enrichment, its saying failed to parse body in clay, but works perfectly fine on serper playground, please help, thanks.
I think there is something wrong with the http api enrichment, its saying failed to parse body in clay, but works perfectly fine on serper playground, please help, thanks.
👋 Hey there! Our support team has got your message - we'll be back in touch within 24 hours (often sooner!). If you haven't already, please include the URL of your table in this thread so that we can help you as quickly as possible!
Hi Silver, thanks for reaching out. Taking a look at this now!
We recently made some changes to the HTTP API enrichment column where you'll now have to add quotation marks around your dynamic tokens! I made the change of adding these to the serper column which was producing errors and it now kicked off successfully. I hope this was helpful. If there's anything else I can assist you with, please let me know!
not my support thread here, but is this change going to affect existing HTTP API columns that don't have quotation marks or were those automatically updated and formatted upon making this change?
Hi Daniel, thanks for following up on this thread - this update should not impact past HTTP API setup - please reach out to us in a separate thread if it does.
heres an input i am still not getting results for, but i am on serper and google. could this be a clay problem ?
Hey there thanks for following up on this - I'm double checking this with our engineering team. It looks like the HTTP API column ran correctly but I see the issue your flagging here with the output discrepancies between Serper and the enrichment column. We'll let you know as soon as we know more!
ok please let me know
i think i figured it out
if my search query contains quotes it wont run
but when i take the quotes out, it runs
im guessing this has to do with the new update
Hello there! Happy to report that we heard back from the team and this issue should be fixed. Thank you for your patience, feel free to try again and let us know if the problem has been resolved on your end.
Thanks! We've reopened this thread. You can continue to add more detail directly in this thread.
This problem is still not solved. If I have quotes in my search query, it wont work
Hey there, thanks for following up on this! It seems as if all issues which are occurring in this table are related to stale data. You set up your workflow to obtain certain outputs but then edited some of the inputs within the workflow, causing stale data. To solve this, you will have to update your upstream cells by selecting "update stale cell" as follows. https://downloads.intercomcdn.com/i/o/w28k1kwz/1218480112/69353c3068c8509829d426f0c6e8/CleanShot+2024-10-16+at+14_41_17.gif?expires=1729106100&signature=f317b2aedb46d6c4e5dc4ac564c12baefb51d885035969ffaa4af7dea6400958&req=dSImHs12nYBeW%2FMW1HO4zXktEjlzquaxFKIHTD9N5UrCob6F28U3hdZftQ67%0ANNGp%0A You can see that rows run successfully once the upstream inputs have been updated correctly. If you're having trouble with the search results instead, can you show me the discrepancy in results between our API call and the serper.dev playground as you did for "taivara.com"? The search results seem aligned to what I'm seeing in the playground here. I hope this was helpful. If there's anything else I can assist you with, please let me know!
hey i understand the stale data part, but its only working because i took out the quotes
could you please check out this request where I can deeper into the problem : https://clayrunhq.slack.com/archives/C025KSBLPGX/p1729103113828799
Hey there, thanks for following up on this, I recorded a quick video with some thoughts on this issue: https://www.loom.com/share/382ef9460bc74420a776a8ffe711847c?sid=5b1e361b-2e53-4152-adb6-e74bc9df058d This appears to be a case of search parameter exhaustion. For many rows, the parameters generated by the AI formula simply do not yield any search results in google. The tests I demonstrated in the video also highlight this behavior.
thanks for the video bruno, but would you mind taking a look into this case specfically : https://clayrunhq.slack.com/archives/C025KSBLPGX/p1729103113828799
just need to know why its working with quotes on google but not in clay
Can you share the table you're running this search in? I can't find this search input in the table we've been working in so far.
this one
if you see serper doesnt work
but the google search works
but if you try to do the search query without quotes on serper it will work
Thank you for your patience! I was able to resolve this by setting up an additional formula in the table called "Quotation Mark Removal." A duplicate of your "Trial 1 HTTP API" column is now running successfully alongside the original. https://downloads.intercomcdn.com/i/o/w28k1kwz/1218551003/bd40fd7d8d484badf12aa6afdeae/CleanShot+2024-10-16+at+15_46_07.gif?expires=1729109700&signature=8770ab97daa19f02c82f9cb4b17ee4441779d7aa7bab016afced9a489f28721d&req=dSImHsx7nIFfWvMW1HO4zR8kYWcYrAX48t4Uby6i2AuAioqjNBzJDUKXuJds%0A5VRf%0A I reviewed our debugging logs, and it appears this behavior is a quirk of the Serper API. The request is being submitted in the correct format following the fix applied yesterday. I hope this was helpful. If there's anything else I can assist you with, please let me know!
hey bruno, I understand that it works without quotations. But the purpose of using quotations is so I can get more accurate results. thats why i need it to work with the quotations specifically.
it would always use to work before with quotations or without
but ever since the new update to add quotations inside of the body of the http api, it has not been working
hopefully this makes sense
Hey there, sorry about the delay here. I flagged this issue to our engineering team and will let you know as soon as I have news on this!
I’ve heard back from our engineering team and can confirm that we’re adding this issue to our backlog. For greater context, our product team prioritizes bugs based on the severity of the issue and how widespread of an experience it is with customers. To set expectations, items like this may not be addressed for some time. The timeline is based on the team's current existing set of work and upcoming feature releases, all subject to change. So what happens now? Now that your issue has been logged, I will tentatively mark this as closed. Please note that I will reach out to you as soon as I receive any new information about this bug. If you don’t hear from me, then the engineering team is still working on it. As soon as there's an update, this conversation will automatically re-open on my end and I will follow up with you. Do you have any further questions before I step away from this conversation until I have an update for you?