Join the AMA (Ask Me Anything) with the Firefox leadership team to celebrate Firefox 20th anniversary and discuss Firefox’s future on Mozilla Connect. Mark your calendar on Thursday, November 14, 18:00 - 20:00 UTC!

This site will have limited functionality while we undergo maintenance to improve your experience. If an article doesn't solve your issue and you want to ask a question, we have our support community waiting to help you at @FirefoxSupport on Twitter and/r/firefox on Reddit.

Search Support

Avoid support scams. We will never ask you to call or text a phone number or share personal information. Please report suspicious activity using the “Report Abuse” option.

Rohkem teavet

When reading JSON formated backup my bookmarks, In what format are the lastModifed and dateAdded proporties?

  • 6 vastust
  • 1 on selline probleem
  • 9 views
  • Viimati vastas cor-el

more options

I would like to analyze my bookmarks using python. I can't figure out the format of the lastModifed and dateAdded fields. There are too many digits for a unix time stamp. When I truncate the number to the first 10 digits and use pythons datetime.fromtimestamp I get dates that don't make any sense. Any help would be greatly appreciated.

I would like to analyze my bookmarks using python. I can't figure out the format of the lastModifed and dateAdded fields. There are too many digits for a unix time stamp. When I truncate the number to the first 10 digits and use pythons datetime.fromtimestamp I get dates that don't make any sense. Any help would be greatly appreciated.

Valitud lahendus

this is what I'm doing that seems to work. from JSON bookmark file: "dateAdded": 1285530321809000,

python 3 code I'm using after truncating the last 6 digits: datetime.datetime.fromtimestamp(1285530321).strftime('%Y-%m-%d %H:%M:%S')

Result: >>> '2010-09-26 15:45:21'

Loe vastust kontekstis 👍 0

All Replies (6)

more options

Those date stamps are in microseconds, so you would have to remove the last three digits or divide the number by 1000 if your software expects milliseconds or six digits for seconds.

The date and time is saved in 64 bit EPOCH format in micro seconds.

Muudetud cor-el poolt

more options

I'm not certain where the discrepancy is but all the numbers in the fields are 16 digit numbers. Dividing by 1000 or dropping the last 3 digits still leaves an integer which is too large. Before I ask the question I tried converting the timestamps using what I thought was the first 10 digits but it turned out I used 11 and it was giving me dates far in the future. So, if I use the first 10 digits it gives me dates that make sense. It seems to me that the last 3 digits (all zeros) must be padding to fill 64 bits. Anyway the problem seems to be solved. Thank you for the help and the info.

more options

Does this converter work on those dates: http://www.esqsoft.com/javascript_examples/date-to-epoch.htm (Option 2)

more options

You didn't answer that time format your software is using.

JavaScript uses millisecond resolution.

If dividing by 1E3 still gives a date in the future for you then it is likely that seconds are used in your case and you would have to divide by 1E6 instead of 1E3

more options

Valitud lahendus

this is what I'm doing that seems to work. from JSON bookmark file: "dateAdded": 1285530321809000,

python 3 code I'm using after truncating the last 6 digits: datetime.datetime.fromtimestamp(1285530321).strftime('%Y-%m-%d %H:%M:%S')

Result: >>> '2010-09-26 15:45:21'

more options

Yes, that is the date-time in Unix Epoch format in seconds.