I want to make a table that gives me the quantity of purchases for each product for the last 3 days.
I start by making this search:
host="L81341" sourcetype="csv" earliest=-2d@d | chart count by Product , _time span=1d | convert timeformat="%Y/%m/%d" ctime(_time)
but the problem is the date is not being converted to a human readable format. I have something like this:
Product1 / 1439676000 /1439762400 /1439848800 Product2 / 50 / 50 / 35 Product3 / 0 / 1 / 85
I want something like that:
ProductID / 2015-08-15/2015-08-16/2015-08-17
Product1 / 50 / 50 / 35
Product2 / 0 / 1 / 85
host="L81341" sourcetype="csv" earliest=-2d@d | bin span=1d _time | convert timeformat="%F" ctime(_time) | chart count by Product,_time
In this way, we bin by day and make the conversion of the time value upfront, (since the _time field would no longer exist after the chart in your search, instead you have fields for each numeric time), Then we perform the chart with the converted _time field.
Edit: Updated the timeformat as it looks like in your example that you're looking for an ISO 8601 formatted date.