Post creation date

[WebdevWebi]

Hi!
is it possible to extract the exact date of creation of the post through the scraper?
For example: 10/04/2025 instead of 13 days ago?

Thanks for help!

[FaustRen]

@WebdevWebi
Hi! Thanks for your question.

A new version of the scraper has just been released — it significantly improves stability and now supports extracting thousands of posts in one go. 🎉

Unfortunately, due to Instagram’s limitations, it’s still not possible to specify an exact date range. The scraper starts from the current time and goes backward based on the number of days you set. However, for each post, the exact creation date (e.g., 10/04/2025) is included in the data.

Hope this update helps! Let me know if you run into any issues.

[WebdevWebi]

@FaustRen
Hii! thanks for the reply!
I tried to update my project with the latest changes you integrated, but in scraping, in the "ftime" data I always receive the date as 1 days ago.

I attach a screenshot of the output, maybe I'm doing something wrong

[FaustRen]

Hi @WebdevWebi, thanks for the update!

From your screenshot and description, it looks like you're using data.head(1).to_dict() — this only prints the first row of the result. So you're always seeing "1 day ago" simply because that’s the value of ftime in the first post.

To verify that you're getting multiple posts (from different days), you can try:

print(data[['ftime', 'time']])
print(len(data))  # To check how many posts were actually returned

[WebdevWebi]

Hi @FaustRen thanks for the reply!

Yes yes I know that in that way I only receive the first row, but the main problem is that I receive the "ftime" data with the formatting "1 day ago", "3 days ago" and not "29/04/2025" or "26/04/2025"

Also if the "time" data corresponds to the Unix timestamp for January 1, 1970, calculating the data from this number:
1745836231
it turns out to be 05/28/2025 10:50:31 a future date

am I doing something wrong?
In the screenshot u can see what i mean.
Now i used

print(data[['ftime', 'time']])
print(len(data))  # To check how many posts were actually returned

[FaustRen]

@WebdevWebi
Bro, convert this to date.

Just like this:

Sample - 1745836231:

[WebdevWebi]

@FaustRen Thanks again for reply!
Okok, i was probably making a mistake in the conversion!

Thanks!

[WebdevWebi]

@FaustRen
Sorry for the 1000 questions I ask you, but I have another problem.

Since this morning I've noticed that the scraper is not working well, that is, in the output it always tells me that the account is private, when in reality it is not.

Until a few days ago everything worked perfectly with the same public accounts.

Could you check what has changed?
Thanks again!

[FaustRen]

@WebdevWebi Hey yo! 👋

The target website has seriously leveled up its anti-scraping defenses lately 😭

I'm planning to push an update within the next two weeks.

If things go well, the next version should work — though scraping might be noticeably slower 😭😭😭

Thanks so much for reporting this 😭😭😭😭😭

[WebdevWebi]

@FaustRen Oh nooo! 😥

Nice, i'm waiting your updates!
Good work!

[FaustRen]

@WebdevWebi Hey, I've released an updated version

This update requires you to install the seleniumbase and requests packages (if you haven’t already).

A caching mechanism for cookies and headers has been added.
This means that on first run—or when the cache expires—the script will take about 5–10 minutes to perform the Cloudflare challenge and generate a new session.
Once the cache is created, everything will run quickly again during the valid period.

Due to stricter anti-bot measures from Cloudflare, the first 12 posts can no longer be retrieved via the old API.
Instead, they’re now extracted directly from the HTML, and you'll find them in the result["init_posts"].

Please give it a try and let me know if it works for you. Thanks

[WebdevWebi]

@FaustRen Ehii! Thanks for update!

I'm testing the new scraper update, I changed the code to extract data from ["init_posts"] as you can see from the screenshot, but in output it still says that the account is private, when in reality it is not.

Am I doing something wrong in the code?

Thanks again!

[FaustRen]

@WebdevWebi thanks for your reply, I've released an updated version again,
try it again, thank you so much.

[WebdevWebi]

@FaustRen Thanks for reply!
I'm trying with your updated, but this is the output that i received.

I can't understand why we can't get past Cloudflare...