Modular library
From APIDesign
| Line 1: | Line 1: | ||
In some cases one needs to have multiple [[ProviderAPI|different implementations]] of some [[API]] being used at once. These implementations are being written in [[distributed development]] environment, they may not be written against the same version of the [[library]] at all. The [[proximity]] between the [[API]] and the implementations is very weak. As a result trying to mimic the principles used while designing [[vendor library]] will not work. | In some cases one needs to have multiple [[ProviderAPI|different implementations]] of some [[API]] being used at once. These implementations are being written in [[distributed development]] environment, they may not be written against the same version of the [[library]] at all. The [[proximity]] between the [[API]] and the implementations is very weak. As a result trying to mimic the principles used while designing [[vendor library]] will not work. | ||
| - | In case of [[modular library]] the [[proximity]] of the [[API]] designer and the implementers is almost the same as the [[ | + | In case of [[modular library]] the [[proximity]] of the [[API]] designer and the implementers is almost the same as the [[proximity]] of the [[API]] designer and the [[ClientAPI|users of the API]]. This requires the [[API]] vendor to prepare [[evolution]] plan for both - the [[ClientAPI]] as well as [[ProviderAPI]]. |
As my discussion at [[OSGiCon]] revealed, people facing this situation realize the difference between the [[Modular library|M-N]] and [[Vendor library|1-N]] case. However, without having [[experience]] with the [[Modular library|M-N]] case, they fallback to most obvious solution: Use ''abstract classes''. While extending ''abstract classes'' is definitely more compatible than [[final interface|extending interfaces]], it can never be 100% safe. The safest solution is to [[APIvsSPI|separate API for clients and providers]] (taken to extreme, while demonstrating all benefits, in the extensible [[visitor]] case). | As my discussion at [[OSGiCon]] revealed, people facing this situation realize the difference between the [[Modular library|M-N]] and [[Vendor library|1-N]] case. However, without having [[experience]] with the [[Modular library|M-N]] case, they fallback to most obvious solution: Use ''abstract classes''. While extending ''abstract classes'' is definitely more compatible than [[final interface|extending interfaces]], it can never be 100% safe. The safest solution is to [[APIvsSPI|separate API for clients and providers]] (taken to extreme, while demonstrating all benefits, in the extensible [[visitor]] case). | ||
| - | Weak [[proximity]] changes everything. An [[API]] designer can simplify its life by claiming all [[ProviderAPI|providers]] are [[proximity|close]]. However sometimes there is just no way around then to admit they are [[proximity|distinct]]. Then one has to accept design style suitable for [[modular | + | Weak [[proximity]] changes everything. An [[API]] designer can simplify its life by claiming all [[ProviderAPI|providers]] are [[proximity|close]]. However sometimes there is just no way around then to admit they are [[proximity|distinct]]. Then one has to accept design style suitable for [[modular library]]. As it is clear that the [[modular library|M-N]] case is a superset of [[vendor library|1-N]] case, I'd like to ask: Why not use the [[modular library|M-N]] style all the time? |
Revision as of 21:31, 8 May 2012
In some cases one needs to have multiple different implementations of some API being used at once. These implementations are being written in distributed development environment, they may not be written against the same version of the library at all. The proximity between the API and the implementations is very weak. As a result trying to mimic the principles used while designing vendor library will not work.
In case of modular library the proximity of the API designer and the implementers is almost the same as the proximity of the API designer and the users of the API. This requires the API vendor to prepare evolution plan for both - the ClientAPI as well as ProviderAPI.
As my discussion at OSGiCon revealed, people facing this situation realize the difference between the M-N and 1-N case. However, without having experience with the M-N case, they fallback to most obvious solution: Use abstract classes. While extending abstract classes is definitely more compatible than extending interfaces, it can never be 100% safe. The safest solution is to separate API for clients and providers (taken to extreme, while demonstrating all benefits, in the extensible visitor case).
Weak proximity changes everything. An API designer can simplify its life by claiming all providers are close. However sometimes there is just no way around then to admit they are distinct. Then one has to accept design style suitable for modular library. As it is clear that the M-N case is a superset of 1-N case, I'd like to ask: Why not use the M-N style all the time?
Follow
