refactoring of the constant?

[tehnerd]

do you consider to refactor this part of the code

const (
    // maxOids is the maximum number of oids allowed in a Get()
    maxOids = 60

    // Base OID for MIB-2 defined SNMP variables
    baseOid = ".1.3.6.1.2.1"

    // Java SNMP uses 50, snmp-net uses 10
    defaultMaxRepetitions = 50

    // TODO comment
    defaultNonRepeaters = 0
)

by moving it to the Default for example.
the reason is:
a lot of networking hardware uses oids from the private space (.1.3.6.1.4...)
(.1.3.6.1.2 is IETF only)
as well as i have found a problem with defaultMaxRepetitions equals to 50 size on some of the networking hardware (so IMO it's better to be configurable instead of hardcoding)

[virtuallynathan]

Hey @tehnerd,

I think these were set by the original author as sensible defaults, they can of course be changed. If you, and the community at large feel these aren't actually sensible, then I suppose it makes sense to change them. Thoughts @soniah & @codedance?

[soniah]

I'll have to look at the code in detail...

• Nathan Owens notifications@github.com [2014-07-16 07:27:57 -0700]:

Hey @tehnerd,

I think these were set by the original author as sensible defaults, they can of course be changed. If you, and the community at large feel these aren't actually sensible, then I suppose it makes sense to change them. Thoughts @soniah & @codedance?

[codedance]

The defaults are there to support the Walk* methods.

I agree that MaxRepetitions should be configurable. I'll change this issue status to "enhancement" and submit a pull request for review.

Regarding OID: You can of course supply your own base OID (and the API encourages this). The default base is there to simply make it easier to not misuse! e.g.

mysnmp.Walk(".1")  // Walk from a very higher level

mysnmp.Walk("1.3.6.1.2.1") // Walk a specific table - e.g.  RFC 1213 system table

mysnmp.Walk("")  // Walk ???? - OK, let's default to the default public space as this should work on every standards compliant device.  Better than throwing a panic :-)

The design is analogous to net-snmp... from the man page:

If no OID argument is present, snmpwalk will search the subtree rooted at SNMPv2-SMI::mib-2

[codedance]

I've created a pull request of Sonia's master under #18 that should address requirements and close this issue. It's a low risk change but does hit public API surface so a review would be good.

Are we happy with the names?

As an aside:
My only other thinking would be to nest them at a deeper level such as

GoSNMP.Walk.MaxRepetitions
GoSNMP.Walk.NonRepeaters

but I feel the extra code complexity would offset any public surface area gain.

[virtuallynathan]

This was merged, I think we can close this unless @tehnerd has any other comments?

[tehnerd]

sounds good for me. thank you