[Kst] how to filter "live" data

[Dan Field]

Ben:

Thanks for the reply.  I am aware of the comment character, but this
isn't going to work for my requirements.  Basically, I'm looking at a
big log file with different record types and I want to pick out just the
record type that I want to view.  I suppose I can use a pipe like

tail -f <datafile> | grep --line-buffered "<pattern>" > filteredfile

and then use kst to read the "filteredfile" output from above.

Best,
    Dan


On 6/29/2015 4:21 AM, Ben Lewis wrote:
> Hi Dan,
>
> There is a provision in the ASCII data wizard to nominate a
> character(s) as a comment indicator. I presume this has the effect of
> filtering out the remaining characters on the line (I've never used
> it). It's a long shot but if the lines you want to filter out all
> begin with the same character and this character does not appear in
> the lines you want to keep this might work for you.
>
> Regards, Ben
>
> On 28/06/2015 2:50 AM, Dan Field wrote:
>> Hi folks:
>>
>> I would like to generate plots from a constantly changing log file.  The
>> problem is that there are different record types in this file and I'm
>> only interested in graphing a particular type.  The records in which I'm
>> interested can be selected based upon the first N characters of the
>> record so a simple grep expression like ^<prefix> would suffice.
>>
>> I know I can grep the log file and send it to a different log file which
>> kst reads, but then updates only occur when the grep program decides to
>> flush its output.  Is there some sort of filter within kst that allows
>> me to exclude records from the data?
>>
>> Best,
>>      Dan
>>
>> _______________________________________________
>> Kst mailing list
>> Kst at kde.org
>> https://mail.kde.org/mailman/listinfo/kst
>
>
>