$generate lhs problem. Manual needs to be updated.

[Kevin Darcy]

Mark Andrews wrote:
> In message <49AF42F8.9070103 at chrysler.com>, Kevin Darcy writes:
>   
>> Jeremy,
>> I don't think the definitions of "rhs" and "lhs" are at issue. What 
>> apparently led the original poster to the wrong solution initially was 
>> the verbiage in the manual stating "Any single *$* symbols within the 
>> *lhs* side are replaced by the iterator value", which implies that "$" 
>> replacement _only_ occurs within the LHS. As Mark confirmed, "$" can 
>> also occur in the RHS, and in fact that's what was required for the 
>> correct solution.
>>
>> Personally, I wouldn't remove "within the LHS" from the verbiage 
>> completely, otherwise someone will undoubtedly complain about not being 
>> able to perform a "$" replacement in the class, type or TTL fields 
>> (users being users :-)
>>
>> But, maybe it could be amended to "within the LHS or RHS"...
>>     
>
> 	The quoted text was taken from a table describing all the
> 	elements of a $GENERATE.  I don't see how anyone reading
> 	the table could say that $ only is valid on the left hand
> 	side especially when there are examples above the table
> 	showing it on both sides.
>
> 	Mark
>
> range
> 	This can be one of two forms: start-stop or start-stop/step.
> 	If the first form is used, then step is set to 1. All of
> 	start, stop and step must be positive.
> lhs
> 	This describes the owner name of the resource records to be
> 	created.  Any single $ (dollar sign) symbols within the lhs
> 	side are replaced by the iterator value. To get a $ in the
> 	output, you need to escape the $ using a backslash \, e.g.
> 	\$. The $ may optionally be followed by modifiers which
> 	change the offset from the iterator, field width and base.
> 	Modifiers are introduced by a { (left brace) immediately
> 	following the $ as ${offset[,width[,base]]}. For example,
> 	${-20,3,d} subtracts 20 from the current value, prints the
> 	result as a decimal in a zeropadded field of width 3.
> 	Available output forms are decimal (d), octal (o) and
> 	hexadecimal (x or X for uppercase). The default modifier
> 	is ${0,0,d}. If the lhs is not absolute, the current $ORIGIN
> 	is appended to the name.  For compatibility with earlier
> 	versions, $$ is still recognized as indicating a literal $
> 	in the output.
> ttl
> 	Specifies the time-to-live of the generated records. If not
> 	specified this will be inherited using the normal ttl
> 	inheritance rules.  class and ttl can be entered in either
> 	order.
> class
> 	Specifies the class of the generated records. This must
> 	match the zone class if it is specified.  class and ttl can
> 	be entered in either order.
> type
> 	At present the only supported types are PTR, CNAME, DNAME,
> 	A, AAAA and NS.
> rhs
> 	rhs is a domain name. It is processed similarly to lhs.
>
>  
>   
>> Jeremy C. Reed wrote:
>>     
>>> On Wed, 4 Mar 2009, Takahiro Masuda wrote:
>>>
>>>   
>>>       
>>>> The tricky part is ${3,0,d} waas not working. I bumped into a site
>>>> that stated $GENERATE range rhs type lhs
>>>>     
>>>>         
>>> That is wrong.
>>>
>>>   
>>>       
>>>> I then tried $GENERATE 11-30  $ PTR ${3,0,d}.COOL.COM. and this
>>>> worked.
>>>>     
>>>>         
>>>   
>>>       
>>>> Anybody here have the ability to update the manual?
>>>>     
>>>>         
>>> Yes.
>>>
>>> But it appears your second try is correct.
>>>
>>> I can improve the documentation to make sure that it explains the two 
>>> abbreviations:
>>>
>>> lhs is left hand side (the label).
>>>
>>> rhs is the right hand side (the RDATA).
>>>
>>> Will that work for you?
>>>       
You're right, I wasn't looking at the documentation in full context, 
only the text that the original poster showed.

Seems pretty clear to me, the example even shows "$" replacements on the 
right-hand side.

Perhaps the original poster will clarify what they think needs changing 
in the documentation.

Or, perhaps, were they trying to navigate the .html file without using a 
browser, e.g. in a text editor? That can give a pretty distorted view, 
but it's hardly ISC's fault if people do that...

                                                                         
               - Kevin