Fix inconsistency in Regex infinite timeout usages #77560
Conversation
In the Regex source generator, there is an inconsistency in which "timeout constant" we are comparing against vs. which constant we are setting when the timeout isn't specified. Fixing it to always use Regex.InfiniteMatchTimeout.
|
Tagging subscribers to this area: @dotnet/area-system-text-regularexpressions Issue DetailsIn the Regex source generator, there is an inconsistency in which "timeout constant" we are comparing against vs. which constant we are setting when the timeout isn't specified. Fixing it to always use Regex.InfiniteMatchTimeout. I could have gone either way on this - there are 2 references to each in the generated code. On one hand, if we always use On the other hand, if we use
|
ghost
locked as resolved and limited conversation to collaborators
In the Regex source generator, there is an inconsistency in which "timeout constant" we are comparing against vs. which constant we are setting when the timeout isn't specified.
Fixing it to always use Regex.InfiniteMatchTimeout.
I could have gone either way on this - there are 2 references to each in the generated code.
On one hand, if we always use
Timeout.InfiniteTimeSpanand just leaveRegex.InfiniteMatchTimeoutas a public API for back-compat, this one unused field could be trimmed away. (an extremely minor deal)On the other hand, if we use
Timeout.InfiniteTimeSpanhere, we should change all the places internally to use it as well, for consistency. Since this was a larger change, and doesn't really have that much benefit, I chose the simpler change. But I can easily be convinced the other way as well.