[rrd-developers] What does the future hold

Tobias Oetiker tobi at oetiker.ch
Sat May 26 19:00:55 CEST 2007

Hi Alex,

> to add "flags" field to global context structure, and this way
> was able to pass "-q" flag from mode-specific to global code of
> rrdtool.
>  	This implementation seems ugly, and the right idea,
> I suppose, is this: Pass some structure like "request mode" from main()
> to all mode-specific code parts. It would allow to pass full status
> information back to main().

the idea behind this is, that a normal rrd_xxx function should not
output anything. It should just ready the data to be picked up by
whoever is going to call this function ... unfortunately the whole
'returning data' business is not done in a consistant manner across
rrdtool ... the implementation I like most, is the one for
rrd_info. it is quite generic and could be used in oterh functions
as well.

with this done, rrdtool (the frontend) could be written in a more
generic fashion, and could also take options as to what it should
do with the output of a command. Syntax could be:

  rrdtool [frontend-option] function [function parameters]


  rrdtool --silent graph bla.png ...

  rrdtool --return-key-value graph bla.png ...

the way to go about this is to first write some notes about your
vision for implementing something, so that you do not create code
that does end up in the bit-bucket because it does not fit my vision
for rrdtool.


Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten
http://it.oetiker.ch tobi at oetiker.ch ++41 62 213 9902

More information about the rrd-developers mailing list