Utilities to merge associate blocks and restrict depth of associate resolution #388
Conversation
This file contains 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
|
Documentation for this branch can be viewed at https://sites.ecmwf.int/docs/loki/388/index.html |
mlange05
force-pushed
the
naml-resolve-assoc-merge
branch
from
401c982
to
729e120
Compare
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #388 +/- ##
==========================================
+ Coverage 95.57% 95.96% +0.38%
==========================================
Files 201 198 -3
Lines 39833 39279 -554
==========================================
- Hits 38071 37694 -377
+ Misses 1762 1585 -177
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
mlange05
force-pushed
the
naml-resolve-assoc-merge
branch
2 times, most recently
from
02c5b7a
to
5f9779e
Compare
mlange05
force-pushed
the
naml-resolve-assoc-merge
branch
from
7b5df63
to
810c447
Compare
mlange05
force-pushed
the
naml-resolve-assoc-merge
branch
from
810c447
to
5b735e5
Compare
|
@reuterbal This one took a few little detours, but is now ready I think. Please have a look. |
reuterbal
approved these changes
Oct 17, 2024
| @@ -239,33 +239,41 @@ class CaseInsensitiveDict(OrderedDict): | |||
| https://stackoverflow.com/questions/2082152/case-insensitive-dictionary | |||
| """ | |||
| def __setitem__(self, key, value): | |||
| super().__setitem__(key.lower(), value) | |||
| key = key.lower() if isinstance(key, str) else key | |||
There was a problem hiding this comment.
Fantastic 😍 This has annoyed me so often but I haven't come around to fixing it...
loki/ir/tests/test_ir_nodes.py
Outdated
Comment on lines
316
to
319
| assert 'B%a' in assoc.association_map and assoc.association_map['B%a'] == a | ||
| assert b_a in assoc.association_map and assoc.association_map[b_a] == a | ||
| assert 'a' in assoc.inverse_map and assoc.inverse_map['a'] == b_a | ||
| assert a in assoc.inverse_map and assoc.inverse_map[a] == b_a |
There was a problem hiding this comment.
Could probably be stricter here and mandate is relationship, to establish that there shouldn't be object copies at play?
Suggested change
| assert 'B%a' in assoc.association_map and assoc.association_map['B%a'] == a | |
| assert b_a in assoc.association_map and assoc.association_map[b_a] == a | |
| assert 'a' in assoc.inverse_map and assoc.inverse_map['a'] == b_a | |
| assert a in assoc.inverse_map and assoc.inverse_map[a] == b_a | |
| assert 'B%a' in assoc.association_map and assoc.association_map['B%a'] is a | |
| assert b_a in assoc.association_map and assoc.association_map[b_a] is a | |
| assert 'a' in assoc.inverse_map and assoc.inverse_map['a'] is b_a | |
| assert a in assoc.inverse_map and assoc.inverse_map[a] is b_a |
loki/ir/tests/test_ir_nodes.py
Outdated
Comment on lines
322
to
326
| assert assign.lhs.scope == scope | ||
| assert assign2.lhs.scope == scope | ||
| assoc.rescope_symbols() | ||
| assert assign.lhs.scope == assoc | ||
| assert assign2.lhs.scope == scope |
There was a problem hiding this comment.
Same here? (is instead of ==)
reuterbal
added
the
ready for merge
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Note: This sits on top of PR #387 and requires this to be merged first.This PR adds a new utility to merge
Associateblocks and adds a feature to the existingresolve_associationsmethod to restrict the starting depth of the association resolution. Combining the two allows for a neat look and feel, where all except the first layer of (non-nested) associations get resolved after all non-nested ones have been merged into the outermost associate block. Tests have been added for both features.Edit: This one took on a life of its own after the interior mechanics of
ResolveAssociateswere changed ( PR #387 ). As the new resolver relies very strictly on identifying the "depth" via finding the defining scope, the new merge mechanism needs to carefully update the changes scopes to re-build the local symbol table of theAssociatenodes. This is done by moving the derivation mechanics from the frontend into the IR node and triggering it specifically (comment and suggestions welcome).I've also added a rudimentary node-level test for
Associatenodes and improved the associate mappings to allow symbols as keys, as well as strings.