Transform external interface to a modular and extensible design - #579
Draft
Krzmbrzl wants to merge 61 commits into
Draft
Transform external interface to a modular and extensible design#579Krzmbrzl wants to merge 61 commits into
Krzmbrzl wants to merge 61 commits into
Conversation
This ensures that the registration is only performed once instead of every time the corresponding header file is included somewhere.
This handler takes care of looping over data inputs and converting them to the appropriate data types. Furthermore, this wrapper takes care of propagating user-created aliases from one step to another. This allows for a much more streamlined implementation of specific processing steps, provided they process one data input at a time.
This assertion is doubly wrong in case the symmetrizer is not particle-number conserving. First, because calling rank() on a tensor that has unequal number of bra and ket indices will throw an exception. And second, the code flow up to this point guarantees rank > 1 only in the particle-number conserving case anyway. Hence, even if calling rank() wasn't ill-defined the logic would be flawed.
This gets rid of the unnecessary lambda function (that is only called exactly once and immediately after its declaration) in order to change the control flow to eliminate redundant simplify steps as well as to move the expression expansion up to be the very first step in processing.
The verification code can't properly deal with "partial" symmetrizers (where the number of bra and ket indices differ) as in those cases the symmetrizer doesn't actually contain all external indices but the affected code will think so. Since messing up the external indices in a ResultExpr is somewhat less likely than when specifying the indices on a per-function-call basis, we just disable the verification code in this case.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
All processing is now performed in terms of individual processing steps and the flow across those is completely defined in the JSON driver. This replaces the C++ code that drove the processing and only took options from the JSON file.
This changes makes it much simpler to extend the external interface to add new types of processing steps or to adapt existing ones as the required changes will be quite localized to the associated step implementations.