Created attachment 505228 [details] [diff] [review] fix v1.0 windres (mingw rc compiler) doesn't like UTF-16 files like nptest.rc. It was converted by test for bug 607121. The same effect can be achieved using UTF-8 file. It's also nicer for developers. Currently diffs for this file show up as binary diffs. For that reason I'm attaching the converted version instead of diff. It was generated using iconv and the only manual change was for codepage pragma.
Attachment #505228 - Flags: review?(jfkthame)
Comment on attachment 505228 [details] [diff] [review] fix v1.0 If you can figure out how to make this work, I'd be delighted - I'd much rather have UTF-8 source. But unfortunately, this does not work as it stands, at least with the MS resource compiler; the DLL ends up with a string where each UTF-8 byte has been "converted" to unicode through codepage 1252, so the Unicode text we're trying to test gets destroyed.
Attachment #505228 - Flags: review?(jfkthame) → review-
Summary: notest.rc doesn't compile with windres → nptest.rc doesn't compile with windres
Created attachment 505404 [details] [diff] [review] fix v1.1 The previous patch showed in hexdump the char U2122, so I assumed it would be fine. Now I've found better way to achieve it, we don't even need UTF-8 for that. All we need is using L"" instead of "" for quoting strings together with \x escaper. The attached nptest.rc gives me byte-to-byte identical .res file on VS10 to current m-c version and works fine with windres.
Comment on attachment 505404 [details] [diff] [review] fix v1.1 Great, thanks!
Attachment #505404 - Flags: review?(jfkthame) → review+
Status: NEW → RESOLVED
Last Resolved: 7 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.