Trigraphs are evaluated both before and after line splicing

[PanoramixDeDruide]

Just discovered that the Wave Driver (not sure about other setups) preprocesses the following code in a way that doesn't conform to the standard:

???/
?=define X 2
X

Is evaluated as 2, while the standard (at least from what I know, would love a source on this) says trigraphs aren't supposed to be evaluated after line splicing happens. Therefore, I believe the correct way to process the above snippet should be to convert ??/ to \, see there are no more trigraphs to convert, splice the first two lines into ??=define X 2, and then not to convert the ??= trigraph to a # because we've already spliced lines. Expected output is as follows:

??=define X 2
X

[jefftrull]

Sorry about the delay! Let me see if I have this right. You expect:

???/
?=define X 2

to become via trigraph substitution of ??/:

?\
?=define X 2

and then, via line splicing:

??=define X 2

without any further trigraph substitution occurring? And that Wave behaves differently?

Based on experimenting with clang and gcc, and comparing the output to the lexed_tokens sample, I agree. Wave seems to be excessively aggressive, performing at least two (conceptual) passes here.

That said, trigraphs seem fairly obscure - so much so that they were eliminated in C++17. The gcc docs say:

Trigraphs are not popular and many compilers implement them incorrectly. Portable code should not rely on trigraphs being either converted or ignored.

As a result, I think this is so low priority as to become WONTFIX in Wave. In fact I think you could argue that supporting them past C++14 should count as a bug :)