27 Jun 15 04:52, you wrote to me:
KR>>> The random system chooses between two given origin definitions
KR>>> randomly. What would happen if the is one origin definition only?
KR>>> What would happen if there is one username definition only?
ml>> see? the name "random" just implies some automatic choice using a
ml>> selector to choose from more than one option...
KR> That matches the function exactly. But "automatic choice" "selector"
KR> is the group match in the list of groups. ;)
ahhh... and i suppose having only one option to choose from is just as valid as
having several... i'm getting old...
KR>>> group pdq
KR>>> username "another name"
KR>>> endgroup
KR>>> would do your wish. xyz and pdq must be exact echotags and be before
KR>>> other wildcarded echotags. In the random system the first hit
KR>>> counts.
ml>> echotags? not groups??
KR> The random system is very powerful and can handle "groups" in many ways.
this makes more sense now that you explain it like you have... it didn't make
much sense to me before...
KR> Like Nick posted it can be groups by letters, as used in fidoconfig
KR> definitions (i didn't knew that before, ty). This affects all echos
KR> which are part of the group as defined in fidoconfig.
you're welcome... i do use fidoconfig definitions and my golded reads
fidoconfig directly for all areas...
KR> group can be one single echotag, which matches your example in the
KR> easiest way. group can be a wildcarded echotag, to match all REQUEST.*
KR> echos or NODES.* or ALLFIX* or BBS* and so on.
interesting... i don't know why i would want to do that but it is cool...
KR> group can be a so far unknown unique groupname and echos can be
KR> included into this new group with the MEMBER keyword. As far as i know
KR> "member" can use wildcards too. The trick is that it's valid to use
KR> more than one member keyword. You could build a group with same
KR> defaults for all ALLFIX*, BBS* and OS2* echos for example.
wowow... it could get complicated and twisted at the same time :)
ml>> why is it called "group" then?
KR> Because the group and endgroup keywords are used to build groups in
KR> many ways.
yeah, i guess at some point, something has to be chosen as the delinieator...
ml>> i have my echotags grouped for similar items... eg: group z is my
ml>> automatically added areas... it is at the bottom of the list... group
ml>> a is my local areas... group b is fidonet backbone carried areas... i
ml>> also break out administrative areas into their own groups... these
ml>> groups are the same as the groups the tosser uses...
KR> If it works, that's great! ;-)
it does for me... all areas get a default description of "auto added area" and
then i just go into my fidoconfig and give it the proper description from the
NA file or from the echolist if the NA file doesn't have one... at that time, i
also adjust the path to place the area's files in the proper directory and
change the defined group the area belongs to... the only other thing is to
manually move the files to the proper path... mc (midnight commander) helps
with that in a quick pinch ;)
i found that going with a simple alphabetical ordering was best and easiest...
especially if one is in multiple networks...
eg:
~/ftn/msgbases
~/ftn/msgbases/fidonet
~/ftn/msgbases/fidonet/backbone
~/ftn/msgbases/fidonet/backbone/a
~/ftn/msgbases/fidonet/backbone/b
~/ftn/msgbases/fidonet/backbone/c
[...]
~/ftn/msgbases/fidonet/zone1only
~/ftn/msgbases/fidonet/region18only
~/ftn/msgbases/fidonet/net3634only
[...]
~/ftn/msgbases/anothernet/backbone
~/ftn/msgbases/anothernet/backbone/a
~/ftn/msgbases/anothernet/backbone/b
~/ftn/msgbases/anothernet/backbone/c
keeps things pretty orderly and clean... all my newly added areas are placed in
the base directory for the network they originate in...
KR> My next step should be international templates and origin lines, but
KR> i'm to lazy at the moment.
yeah, i don't think i'm going to go that far... now, if there was a way to
select different tagline files based on the group, that might be interesting ;)
)\/(ark
... I am positive that a definite maybe is probably in order.
---
* Origin: (1:3634/12.73)
|