Fix/delegate __lt__ to PackageVersionLike
#448
Conversation
| v3 = PackageVersion("latest") | ||
| v3 = PackageVersion("0.12.0") | ||
| assert v1 < v2 < v3 |
There was a problem hiding this comment.
this won't work otherwise, test was not proper since relied on string comparison ("0" < "l" == True)
There was a problem hiding this comment.
NOTE: we do not delegate to semver.VersionInfo to maintain the ability to use "latest" and "any", however we cannot compare PackageVersion("1.0.0") < PackageVersion("latest"), which will raise ValueError: latest is not valid SemVer string due to the deflection. See Version.__lt__ and Version.compare
There was a problem hiding this comment.
Having this change does not make sense at all, the latest version should always be the superior when comparing to other versions.
There was a problem hiding this comment.
Au contraire mon ami.
- This change would have made the original implementation fail.
- Comparing "latest" should not always return
True, in case latest is equal to current version being compared to. In this sense "latest" is a property. - We also have this "any" which poses another issue.
Hence, additional tests are required to cover these cases, for which I can stack a PR later this week. Prior to that, however, we need to decide how to deal with these. As I stated in my base PR (#428) we should reconsider whether we want to maintain "any" and "latest".
put a #428 (comment) in base PR to address later so we can merge #445, #447 and #448 into it.
| sorted_expected = list(map(str, sorted(self.expected_dependencies))) | ||
| sorted_missing = list(map(str, sorted(self.missing_dependencies))) | ||
| expected = list(map(str, self.expected_dependencies)) | ||
| missing = list(map(str, self.missing_dependencies)) | ||
| print("=" * 50) | ||
| print(f"Package {self.configuration_file}:") | ||
| print(f"Expected: {pprint.pformat(sorted_expected)}") | ||
| print(f"Missing: {pprint.pformat(sorted_missing)}") | ||
| print(f"Expected: {pprint.pformat(expected)}") | ||
| print(f"Missing: {pprint.pformat(missing)}") | ||
| print("=" * 50) |
There was a problem hiding this comment.
not sortable when author and name differ
| flat_tree.append( | ||
| sorted( | ||
| {package for package in tree_level if package not in dirty_packages} | ||
| ) | ||
| ) | ||
| flat_tree.append([]) | ||
| for package in tree_level: | ||
| if package not in dirty_packages and package not in flat_tree[-1]: | ||
| flat_tree[-1].append(package) |
There was a problem hiding this comment.
not sortable, this way we obtain a unique set that maintains a deterministic ordering
__lt__ to PackageVersionLike__lt__ to PackageVersionLike
__lt__ to PackageVersionLike__lt__ to PackageVersionLike
__lt__ to PackageVersionLike__lt__ to PackageVersionLike
2ec503d to
4784ac1
Compare
f6889df to
779c951
Compare
Codecov ReportBase: 97.31% // Head: 97.30% // Decreases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## fix/data_types__lt__ #448 +/- ##
========================================================
- Coverage 97.31% 97.30% -0.01%
========================================================
Files 221 221
Lines 19543 19548 +5
========================================================
+ Hits 19018 19022 +4
- Misses 525 526 +1
Flags with carried forward coverage won't be shown. Click here to find out more.
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
| queue.extendleft(sorted(roots)) # type: ignore | ||
| queue.extendleft(sorted(roots, key=str)) # type: ignore |
There was a problem hiding this comment.
this guarantees the ordering of root nodes is deterministic, and hence the returned List[T] of find_topological_order will be as well.
__lt__ to PackageVersionLike__lt__ to PackageVersionLike
Proposed changes
Delegate comparison operations to
PackageVersionLike._versionin custom__lt__methods on custom data types:PackageVersion,PublicIdandPackageId.Fixes
If it fixes a bug or resolves a feature request, be sure to link to that issue.
Types of changes
What types of changes does your code introduce to agents-aea?
Put an
xin the boxes that applyChecklist
Put an
xin the boxes that apply.developbranch (left side). Also you should start your branch off ourdevelop.Further comments
If this is a relatively large or complex change, kick off the discussion by explaining why you chose the solution you did and what alternatives you considered, etc...