Make generated test failures more useful #1498
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Currently, when a `sys` compilation fails, the only context given is the
command used to compile the code.
This is less than helpful. This PR adds all the compilation commands'
output to the error message.
It also prints the message. `unwrap` doesn't format properly error
messages, so we use `println!` as well. This results in the output being
duplicated.
I consider the duplicate output to be acceptable, it's already far
better than the current output, the fact that code is generated makes
testing and developing a lot harder and we shouldn't let perfect be
enemy of the better.
Here is a sample of the difference this commit makes:
BEFORE
```
thread 'cross_validate_layout_with_c' panicked at 'called `Result::unwrap()` on an `Err` value: "compilation command \"cc\" \"-Wno-deprecated-declarations\" \"-std=c11\" \"-D__USE_MINGW_ANSI_STDIO\" \"-I/usr/include/wpe-webkit-1.1\" \"-I/usr/include/glib-2.0\" \"-I/usr/lib/glib-2.0/include\" \"-I/usr/include/sysprof-4\" \"-I/usr/include/libsoup-3.0\" \"-I/usr/include/libmount\" \"-I/usr/include/blkid\" \"-pthread\" \"-I/usr/include/wpe-1.0\" \"-DWPE_ENABLE_XKB=1\" \"tests/layout.c\" \"-o\" \"/tmp/abiXdITDL/layout\" failed, exit status: 1"
```
AFTER
```
compilation command "cc" "-Wno-deprecated-declarations" "-std=c11" "-D__USE_MINGW_ANSI_STDIO" "-I/usr/include/wpe-webkit-1.1" "-I/usr/include/glib-2.0" "-I/usr/lib/glib-2.0/include" "-I/usr/include/sysprof-4" "-I/usr/include/libsoup-3.0" "-I/usr/include/libmount" "-I/usr/include/blkid" "-pthread" "-I/usr/include/wpe-1.0" "-DWPE_ENABLE_XKB=1" "tests/layout.c" "-o" "/tmp/abiXdITDL/layout" failed, exit status: 1
stdout:
stderr: In file included from tests/layout.c:7:
tests/manual.h:3:10: fatal error: cog/cog.h: No such file or directory
3 | #include <cog/cog.h>
| ^~~~~~~~~~~
compilation terminated.
thread 'cross_validate_layout_with_c' panicked at 'called `Result::unwrap()` on an `Err` value: "compilation command \"cc\" \"-Wno-deprecated-declarations\" \"-std=c11\" \"-D__USE_MINGW_ANSI_STDIO\" \"-I/usr/include/wpe-webkit-1.1\" \"-I/usr/include/glib-2.0\" \"-I/usr/lib/glib-2.0/include\" \"-I/usr/include/sysprof-4\" \"-I/usr/include/libsoup-3.0\" \"-I/usr/include/libmount\" \"-I/usr/include/blkid\" \"-pthread\" \"-I/usr/include/wpe-1.0\" \"-DWPE_ENABLE_XKB=1\" \"tests/layout.c\" \"-o\" \"/tmp/abiXdITDL/layout\" failed, exit status: 1\nstdout: \nstderr: In file included from tests/layout.c:7:\ntests/manual.h:3:10: fatal error: cog/cog.h: No such file or directory\n 3 | #include <cog/cog.h>\n | ^~~~~~~~~~~\ncompilation terminated.\n"', tests/abi.rs:198:31
```
| @@ -346,13 +346,22 @@ impl Compiler { | |||
| } | |||
|
|
|||
| pub fn compile(&self, src: &Path, out: &Path) -> Result<(), Box<dyn Error>> { | |||
| use std::str::fom_utf8; | |||
There was a problem hiding this comment.
Suggested change
| use std::str::fom_utf8; | |
| use std::str::from_utf8; |
Comment on lines
+358
to
+359
| let stdout = from_utf8(&output.stdout).unwrap_or(""); | ||
| let stderr = from_utf8(&output.stderr).unwrap_or(""); |
There was a problem hiding this comment.
Suggested change
| let stdout = from_utf8(&output.stdout).unwrap_or(""); | |
| let stderr = from_utf8(&output.stderr).unwrap_or(""); | |
| let stdout = from_utf8(&output.stdout).unwrap_or_default(); | |
| let stderr = from_utf8(&output.stderr).unwrap_or_default(); |
maybe?
|
I did use It seems stderr actually is printed and is visible in the terminal when running I'm not sure why in the first place I didn't see the compilation error output. I'm fairly certain it didn't show up. Maybe it depends on what calls the Since this is a regression I'm going to put the pull request in draft mode. |
|
I now know why the tests previously failed without showing any errors. I'm closing this in favor of a fix for the problem in question. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Currently, when a
syscompilation fails, the only context given is the command used to compile the code.This is less than helpful. This PR adds all the compilation commands' output to the error message.
It also prints the message.
unwrapdoesn't format properly error messages, so we useprintln!as well. This results in the output being duplicated.I consider the duplicate output to be acceptable, it's already far better than the current output, the fact that code is generated makes testing and developing a lot harder and we shouldn't let perfect be enemy of the better.
Here is a sample of the difference this commit makes:
BEFORE
AFTER