You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm not sure what __lambda_3; in operator() is supposed to do, and __lambda_4{}da_3{} is nonsense.
I suspect you're replacing the substring at [4,7) with something longer, then doing another replacement on the same line, without adjusting the offsets to account for the previous replacement. This yields lots of strange output (another example: removing auto a = makes it emit lambdas in the output), but I think most or all of it is all the same root cause.
(Additionally, two lambdas on the same line expand to two classes with the same name. Append column number to subsequent ones, maybe?)
The text was updated successfully, but these errors were encountered:
Thanks for reporting. It looks like the matcher matches twice. Column number is a excellent idea. I like to keep the name stable, hence the line number. Adding the column number to it should make it more robust. I will look into that.
Found this on LLVM Weekly. Neat project.
Unfortunately, lambda handling seems quite buggy. For example:
I'm not sure what __lambda_3; in operator() is supposed to do, and __lambda_4{}da_3{} is nonsense.
I suspect you're replacing the substring at [4,7) with something longer, then doing another replacement on the same line, without adjusting the offsets to account for the previous replacement. This yields lots of strange output (another example: removing auto a = makes it emit lambdas in the output), but I think most or all of it is all the same root cause.
(Additionally, two lambdas on the same line expand to two classes with the same name. Append column number to subsequent ones, maybe?)
The text was updated successfully, but these errors were encountered: