[PATCH] D116638: [clang-format] Fix ignoring JavaScriptWrapImport when ColumnWidth: 0

[Martin Probst via Phabricator via cfe-commits]

mprobst added inline comments.


================
Comment at: clang/docs/ClangFormatStyleOptions.rst:2827
 
-**JavaScriptWrapImports** (``Boolean``) :versionbadge:`clang-format 3.9`
-  Whether to wrap JavaScript import/export statements.
+  * If ``ColumnWidth`` is 0 (no limit on the number of columns), then import
+    statements will keep the number of lines they start with.
----------------
andmis wrote:
> HazardyKnusperkeks wrote:
> > mprobst wrote:
> > > this seems odd to me: my understanding is that clang-format always reflows the entire document, there's no logic to ever "keep" whitespace.
> > > 
> > > Are you sure you are seeing this behaviour? The logic change below sounds more as if the ColumnWidth: 0, import lines might not break (mustBreak returns false), but might still break?
> > From what I can tell with `ColumnLimit` (and not `ColumnWidth`) 0, then `clang-format` does many things differently. If by design or by accident I'm not sure. I can see that often, since I format my code with a limit, but my tests without.
> I just double-checked and yes, as-implemented, with `ColumnLimit: 0` and `JavaScriptWrapImports: true`, import statements retain the number of lines they started with.
> 
> It is explicit in the code that with `ColumnLimit: 0`, we decide whether to break on a token based on whether there was a preexisting line break. (Check out `NoColumnLimitLineFormatter`.) I think you're right that usually clang-format reflows everything and I was also confused about this point.
> 
> There are other options that do similar things, for example `EmptyLineAfterAccessModifier`.
Can you check with code that would normally cause reformatting? E.g.

```
import {
  A, B,
  C,
} from 'url';
```



CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D116638/new/

https://reviews.llvm.org/D116638