Fixed bug in dyn_var implementation where returns from [] and * cause bug #73
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
…ables initialized from such expressions don' t get removed
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Outdated suggestions cannot be applied.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.
A recent change (#65) simplified the implementation of pointer types esp. struct types so that the overloaded * and [] operators returned dyn_var instead of builder. This allowed accessing members from these easier. However this causes a major bug where if the user writes code snippets like -
the copy a is completely removed. This is because the operators construct the returned object using (builder::cast). Due to copy elision, a itself is constructed with (cast) causing errors. If a is used multiple times, each use will have ptr[i]. Bugs arise when ptr[i] has side effects.
This bug was discovered while implementing a toy language interpreter with @tekknolagi
To fix this issue, we introduce a clone of dyn_var called dyn_var_mimic which is exactly like dyn_var (inherits from dyn_var), but IS NOT dyn_var. We return this type from the operator. Thus is the user directly accesses the members, the behavior is the same like before, but if a dyn_var is constructed from such an expression, copy elision doesn't happen and a copy is created.
Minor change where dyn_var_parent_selector now strips of references from types before inheriting so dyn_var<T&> also have members. This is unrelated to above.
Sample38 is updated to test this.