.TH "doc_decisions_0b_rejected_cmake_spec_md" 3elektra "Thu Sep 7 2023" "Version 0.11.0" "Elektra" \" -*- nroff -*-
.ad l
.nh
.SH NAME
doc_decisions_0b_rejected_cmake_spec_md \- CMake Spec 
 
.SH "Problem"
.PP
The compilation variants of plugins blow up the number of plugins\&. Additionally there is the concept of default storage + resolver that is needed for bootstrapping plugins\&.
.SH "Constraints"
.PP
.IP "\(bu" 2
full default resolver need to be different from other default resolver for testing
.IP "\(bu" 2
there is no standard resolver, they always should state their configuration
.PP
.SH "Assumptions"
.PP
.IP "\(bu" 2
keep it not too difficult to configure, even though most people will go for the defaults
.PP
.SH "Considered Alternatives"
.PP
.IP "\(bu" 2
many CMake variables for every case KDB_DEFAULT_STORAGE, KDB_DEFAULT_RESOLVER for dynamic and static cases
.IP "\(bu" 2
Have a PLUGINS field that state what is the default dynamic and static plugin, e\&.g\&. ! for storage ? for resolver ds?resolver_b_u_b;ds!dump
.PP
.SH "Decision"
.PP
Rejected: keep default plugins as-is
.SH "Rationale"
.PP
.SH "Implications"
.PP
.SH "Related Decisions"
.PP
.SH "Notes"
.PP