dropping observation with a con

[Nadav Dvir]

Hello all,
I would like to drop certain observations from my database and I am not sure how to do so.
My data consists of stock' ticker, dates, additional date variable, and a binary variable which equals 1 if the additional date variable is equal to the date under each ticker

for example:
Name Date date1 binary
apple 01jan2010 . 0
apple 02jan2010 02jan2010 1
apple 03jan2010 . 0
google 01jan2010 . 0
google 02jan2010 . 0
google 03jan2010 03jan2010 1
tesla 01jan2010 . 0
tesla 02jan2010 . 0
tesla 03jan2010 . 0

I would like to drop every stock that has no date1 (or binary=1) observations, so in the case of my example, that would mean drop all the observations of tesla.
Note: I would like to keep the observations of the other stocks, even though some of them do not coincide with date1.
additional note: 1 observation is enough to keep the stock in the dataset.

result:
Name Date date1 binary
apple 01jan2010 . 0
apple 02jan2010 02jan2010 1
apple 03jan2010 . 0
google 01jan2010 . 0
google 02jan2010 . 0
google 03jan2010 03jan2010 1

Thank you all!

[Ali Atia]

bysort name (binary): drop if binary[_N]==0

[Andrew Musau]

bys Name: drop if sum(binary) <1

[Nadav Dvir]

bysort name (binary): drop if binary[_N]==0

Thank you for the reply, for some reason, the command drops additional observations (for example a row of apple)

The same goes for Andrew's command.
also, each of the two commands drops a different number of observations (which should not happen I believe?)

[Andrew Musau]

You probably have some leading or trailing spaces in your names so that Stata does not recognize that the dropped observation is part of a particular group which should not be dropped. Also, there is an assumption that you do not have missing values, although this should not cause a difference between my code and Ali's. In any case, you need to present a data example using dataex that replicates your issue.