The current json output for --mode lint is not human-readable. It would be great to be able to print a diff instead of json.
It can be a new --diff flag, or --format json/diff, or --mode diff.
For example, instead of
[
{
"path": "separator.in",
"lines": {
"start": 4,
"end": 6
},
"message": "These lines are out of order.",
"fixes": [
{
"replacements": [
{
"lines": {
"start": 4,
"end": 6
},
"new_content": " \"Bar\",\n \"Baz\",\n \"Foo\"\n"
}
]
}
]
},
{
"path": "separator.in",
"lines": {
"start": 12,
"end": 16
},
"message": "These lines are out of order.",
"fixes": [
{
"replacements": [
{
"lines": {
"start": 12,
"end": 16
},
"new_content": "// Sticky comment\n\"Bar\",\n\"Baz\",\n\"Foo\"\n// Trailing comments\n"
}
]
}
]
}
]
I'd like to see
--- goldens/separator.in
+++ goldens/separator.in
@@ -1,17 +1,17 @@
Comma separator:
static class Foo {
// keep-sorted-test start
- "Foo",
+ "Bar",
- "Bar",
+ "Baz",
- "Baz"
+ "Foo"
// keep-sorted-test end
}
Comma separator with comments:
-// keep-sorted-test start sticky_comments=yes
-"Foo",
+// keep-sorted-test start sticky_comments=yes
// Sticky comment
"Bar",
-"Baz"
+"Baz",
+"Foo"
// Trailing comments
// keep-sorted-test end
The current json output for
--mode lintis not human-readable. It would be great to be able to print a diff instead of json.It can be a new
--diffflag, or--format json/diff, or--mode diff.For example, instead of
[ { "path": "separator.in", "lines": { "start": 4, "end": 6 }, "message": "These lines are out of order.", "fixes": [ { "replacements": [ { "lines": { "start": 4, "end": 6 }, "new_content": " \"Bar\",\n \"Baz\",\n \"Foo\"\n" } ] } ] }, { "path": "separator.in", "lines": { "start": 12, "end": 16 }, "message": "These lines are out of order.", "fixes": [ { "replacements": [ { "lines": { "start": 12, "end": 16 }, "new_content": "// Sticky comment\n\"Bar\",\n\"Baz\",\n\"Foo\"\n// Trailing comments\n" } ] } ] } ]I'd like to see