Prior to this patch, llvm-objcopy's error messages for archives with unsupported members only mentioned the archive name, not the member name, making them unhelpful. This change improves it by approximately following GNU objcopy's error message syntax of "<archive name>(<member name>): <problem>".
Details
- Reviewers
jakehehrlich rupprecht grimar seiya MaskRay • espindola alexander-shaposhnikov - Commits
- rZORGf53cc5999dc9: [llvm-objcopy] Improve error message for unrecognised archive member
rZORG1405c433db75: [llvm-objcopy] Improve error message for unrecognised archive member
rGf53cc5999dc9: [llvm-objcopy] Improve error message for unrecognised archive member
rG1405c433db75: [llvm-objcopy] Improve error message for unrecognised archive member
rG5a2b5ca7d2f6: [llvm-objcopy] Improve error message for unrecognised archive member
rL360251: [llvm-objcopy] Improve error message for unrecognised archive member
Diff Detail
- Repository
- rL LLVM
Event Timeline
test/tools/llvm-objcopy/ELF/archive-unknown-members.test | ||
---|---|---|
2 | unrecognised -> unrecognized I think? | |
26 | You have %t.thin1.a %t2.a for llvm-objcopy call and just %t.thin2.a for llvm-strip. | |
tools/llvm-objcopy/llvm-objcopy.cpp | ||
165 | nit: while you are here: I would perhaps remove this variable since it is used only once. |
test/tools/llvm-objcopy/ELF/archive-unknown-members.test | ||
---|---|---|
32 | recognised->recognized :] |
Sorry, I didn't see your latest comment before committing. I've fixed it in a follow up change.
test/tools/llvm-objcopy/ELF/archive-unknown-members.test | ||
---|---|---|
32 | recognised is not wrong.. |
test/tools/llvm-objcopy/ELF/archive-unknown-members.test | ||
---|---|---|
32 | I thought I read somewhere that the standard was American English in LLVM, but I now can't find it, so maybe I imagined it? |
test/tools/llvm-objcopy/ELF/archive-unknown-members.test | ||
---|---|---|
32 | Honestly saying I suggested to change unrecognised -> unrecognized initially because my spellchecker complained about this word. I realized that it is a British style only after your first reply :) I do not know what is the LLVM standard (I would not expect to see such a requirement probably.) Having consistency in spelling is not that bad still I think. I.e. in this test error messages spelling wasn't consistent with the comments before and now it is. But perhaps it is really was too subtle to care about? |
unrecognised -> unrecognized I think?