Various features for example pipeline #31

HEL · 2026-07-07T15:30:12Z

HEL commented

2026-07-07 15:30:12 +00:00

This rather large PR regroups many small additions and fixes that were useful to demonstrate some strengths of the system and are generally interesting / useful

This includes, but is not limited to:

import statements now introduce variables in the environment, with unknown types
Column can now be used in type expression, e.g. Column[int]
copy, info and sort_values methods on frames
not operator and proper handling of unary operations in Midas and Python
properly check parameter default values (ensure type matches annotation)
correctly parse int literals as integers in Midas (previously, all numbers were parsed as floats)
handle UnknownType more gracefully in some places
handle scalar operations on frames and columns (e.g. df * 2.5 - 3.2)
use formulae to compute the return type of some aggregation methods (e.g. mean = (Type + Type) / int)

This rather large PR regroups many small additions and fixes that were useful to demonstrate some strengths of the system and are generally interesting / useful This includes, but is not limited to: - import statements now introduce variables in the environment, with unknown types - `Column` can now be used in type expression, e.g. `Column[int]` - `copy`, `info` and `sort_values` methods on frames - `not` operator and proper handling of unary operations in Midas and Python - properly check parameter default values (ensure type matches annotation) - correctly parse int literals as integers in Midas (previously, all numbers were parsed as floats) - handle `UnknownType` more gracefully in some places - handle scalar operations on frames and columns (e.g. `df * 2.5 - 3.2`) - use formulae to compute the return type of some aggregation methods (e.g. `mean = (Type + Type) / int`)

HEL self-assigned this 2026-07-07 15:30:12 +00:00

HEL added 26 commits 2026-07-07 15:30:12 +00:00

feat: handle import statements 3de36e5bd5

import statements now introduce variables with unknown types to avoid "Unknown variable" warnings

fix(checker): handle application of column type 16c19cca75

fix: add __name__ to preamble 8bc35f7754

feat(checker): add copy and info methods a640b8b3dd

feat: handle not operator 164648e8df

fix(checker): properly check parameter defaults aab4641584

fix(checker): handle all unary ops in evaluator af6056a83f

fix(parser): parse int literals as integers 56987f6cff

avoid casting to float when the literal is an integer

fix(checker): handle setting unknown as column 4eb133ce17

fix(checker): report predicate argument errors 5e87ba2cd1

fix(gen): generate assertion definitions 9277bd2cd0

feat(checker): handle scalar ops on frames and columns 1098e33d07

fix(gen): only generate length assertion for non-scalar ops 5051e155c0

feat(checker): add unsupported parameter flag f48ebd49d1

feat(checker): add sort_values on frames and columns c6b8c0a145

docs: add docstrings to generator submodule ed07b01563

fix: make Frame/Column inherit from pandas classes c2f41eb392

tests: update with integer literals fe34c77e4c

tests: update with unsupported flag 5311307a6f

chore: add weather pipeline example 83eecd612e

chore: add comments to weather example 5958e3612b

feat(checker): add formula to compute aggregation type 48be2d454c

fix(cli): tweak highlight diagnostic tooltip f3dec414cc

fix(checker): compute aggregation of column groupby 25c11c3a53

docs: update docstring with new formula 40bda81c32

tests: update with new aggregate return types 094554cb72

HEL merged commit e855a09a6b into main

2026-07-07 15:30:22 +00:00

HEL deleted branch feat/complete-example

2026-07-07 15:30:22 +00:00

HEL referenced this issue from a commit

2026-07-07 15:30:22 +00:00

Merge pull request 'Various features for example pipeline' (#31) from feat/complete-example into main

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: HEL/midas#31