You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Providing for either a string or a function that provides a dynamic
string is "cute" and as Ryan Dahl says in "10 Things I Regret About
Node" it's those cute things that you end up regretting in the end.
I already regret it. I had to reimplement all of fmt to support the
flattened versions. That was probably stupid. Then I can to implement MarshalJSON and MarshYAML (and family) because I'm not just dealing
with basic JSON and YAML types. That is unnecessary and just does not
warrant possible value of having dynamic documentation.
Plus, this inhibits the greater good of bundling multi-language
documentation with CmdBox composite commands, which will never have any
function definitions in them. Having that is far more important than
supporting the possible use of dynamic documentation detection.
Besides, if a Command author wants to provide some insight into the
system it is far better to create a command for that than to build it
into some form of dynamic docs.
This is gonna hurt. I put a lot of time into this. Wasted time. But
sunk-cost fallacy be damned.
The text was updated successfully, but these errors were encountered:
Providing for either a string or a function that provides a dynamic
string is "cute" and as Ryan Dahl says in "10 Things I Regret About
Node" it's those cute things that you end up regretting in the end.
I already regret it. I had to reimplement all of
fmt
to support theflattened versions. That was probably stupid. Then I can to implement
MarshalJSON
andMarshYAML
(and family) because I'm not just dealingwith basic JSON and YAML types. That is unnecessary and just does not
warrant possible value of having dynamic documentation.
Plus, this inhibits the greater good of bundling multi-language
documentation with CmdBox composite commands, which will never have any
function definitions in them. Having that is far more important than
supporting the possible use of dynamic documentation detection.
Besides, if a Command author wants to provide some insight into the
system it is far better to create a command for that than to build it
into some form of dynamic docs.
This is gonna hurt. I put a lot of time into this. Wasted time. But
sunk-cost fallacy be damned.
The text was updated successfully, but these errors were encountered: