From 109e1070237fa6cf460f345345bce8c7b811aa9a Mon Sep 17 00:00:00 2001 From: Tobias Grieger Date: Wed, 11 May 2022 05:48:43 -0400 Subject: [PATCH] Avoid unicode filename This is similar to 53e9d3ee100e0cf54d18eeae09dda6e09a29d090. Over at https://github.com/cockroachdb/cockroach, we use bazel and it chokes on this file: ``` java.io.IOException: Error extracting [...]/com_github_maruel_panicparse_v2-v2.2.1.zip to [...]: [...]/com_github_maruel_panicparse_v2/cmd/panic/internal/utf8/?tf8.go (Illegal byte sequence) ``` which is why we're on a fork so far. It would be nice to avoid this. There is an (eternal, it seems) [upstream discussion], so there is little home of bazel starting to support non-unicode filenames. I wonder if the loss in test coverage (if any) is acceptable in return for avoiding build problems like the ones we've been seeing. [upstream discussion]: https://github.com/bazelbuild/bazel/issues/374 --- .../utf8/\303\271tf8.go" => cmd/panic/internal/utf8/utf8.go | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename "cmd/panic/internal/utf8/\303\271tf8.go" => cmd/panic/internal/utf8/utf8.go (100%) diff --git "a/cmd/panic/internal/utf8/\303\271tf8.go" b/cmd/panic/internal/utf8/utf8.go similarity index 100% rename from "cmd/panic/internal/utf8/\303\271tf8.go" rename to cmd/panic/internal/utf8/utf8.go