Skip Menu |

Preferred bug tracker

Please visit the preferred bug tracker to report your issue.

This queue is for tickets about the CASCM-Wrapper CPAN distribution.

Report information
The Basics
Id: 71590
Status: rejected
Priority: 0/
Queue: CASCM-Wrapper

People
Owner: MITHUN [...] cpan.org
Requestors: john.oyler [...] ttu.edu
Cc:
AdminCc:

Bug Information
Severity: Wishlist
Broken in: (no value)
Fixed in: (no value)



Subject: The hsql command is a special case that requires different behavior.
The SCM command "hsql" needs to be treated differently than other commands. It allows sending arbitrary select queries, the important ones being anything in the harvest schema. It allows you to check which packages exist, which package groups exist, etc etc. Basically access to all the metadata. The scripter would supply a query string to that particular method, which should return values in a similar fashion to the DBI module. It could even de-stupidify the <nl> and <tab> substrings that CA thinks is an intelligent way to escape embedded newlines and tabs in records. I'm willing to help with the coding, just would like some input on all the details.
Show quoted text
>It could even de-stupidify the > <nl> and <tab> substrings > that CA thinks is an intelligent way to escape embedded newlines and > tabs in records.
Hehe I agree. I guess the same can be said for commands like 'hlv', 'hsv' and 'hsigget' which returns some sort of delimited output. However, the main intention of this module was to provide a wrapper on running commands and not parsing the output - which can sometimes make it version dependent. I am open to creating _helper_ modules which parses output into perl structures. Probably a CASCM::Wrapper::ParseOutput or similar where output parsing can be explicitly requested (similar to log parsing). On Tue Oct 11 01:33:14 2011, http://j-oyler.myopenid.com/ wrote: Show quoted text
> The SCM command "hsql" needs to be treated differently than other > commands. It allows > sending arbitrary select queries, the important ones being anything in > the harvest schema. It > allows you to check which packages exist, which package groups exist, > etc etc. Basically access > to all the metadata. > > The scripter would supply a query string to that particular method, > which should return values in > a similar fashion to the DBI module. It could even de-stupidify the > <nl> and <tab> substrings > that CA thinks is an intelligent way to escape embedded newlines and > tabs in records. > > I'm willing to help with the coding, just would like some input on all > the details. > >
The requested functionality is not in the scope of this module.