Feature: CallerArgumentExpressionAttribute #51952
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -6600,4 +6600,43 @@ To remove the warning, you can use /reference instead (set the Embed Interop Typ | |
| <data name="ERR_FunctionPointerTypesInAttributeNotSupported" xml:space="preserve"> | ||
| <value>Using a function pointer type in a 'typeof' in an attribute is not supported.</value> | ||
| </data> | ||
| <data name="ERR_BadCallerArgumentExpressionParamWithoutDefaultValue" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute may only be applied to parameters with default values</value> | ||
| </data> | ||
| <data name="ERR_NoConversionForCallerArgumentExpressionParam" xml:space="preserve"> | ||
| <value>CallerArgumentExpressionAttribute cannot be applied because there are no standard conversions from type '{0}' to type '{1}'</value> | ||
|
There was a problem hiding this comment.
Should conversio be also implicit? #Closed There was a problem hiding this comment. I copied the exact same text from existing diagnostics. I'm not entirely sure about the correct terminology. Will need to go through the Standard conversions in spec. There was a problem hiding this comment. @AlekseyTs The following code compiles (standard explicit conversion from using System.Runtime.CompilerServices;
public class C
{
public void M([CallerMemberName] object name = null)
{
}
}So it seems standard conversion is correct There was a problem hiding this comment.
Doesn't conversion in this scenario go from string to object instead? In reply to: 600250422 [](ancestors = 600250422) There was a problem hiding this comment. @AlekseyTs Consider the following code: using System.Runtime.CompilerServices;
public class C
{
public static void M([CallerMemberName] C name = null)
{
}
}The error message is:
Changing There was a problem hiding this comment.
The question is whether the standard conversion should also be implicit. In reply to: 600628537 [](ancestors = 600628537) There was a problem hiding this comment. @AlekseyTs, Well, looking at the spec of CallerMemberName:
There was a problem hiding this comment. Ops I was swapping the source and destination for conversion in mind 😄 |
||
| </data> | ||
| <data name="WRN_CallerArgumentExpressionParamForUnconsumedLocation" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute applied to parameter '{0}' will have no effect because it applies to a member that is used in contexts that do not allow optional arguments</value> | ||
| </data> | ||
| <data name="WRN_CallerArgumentExpressionParamForUnconsumedLocation_Title" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute will have no effect because it applies to a member that is used in contexts that do not allow optional arguments</value> | ||
| </data> | ||
| <data name="WRN_CallerFilePathPreferredOverCallerArgumentExpression" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute applied to parameter '{0}' will have no effect. It is overridden by the CallerFilePathAttribute.</value> | ||
| </data> | ||
| <data name="WRN_CallerFilePathPreferredOverCallerArgumentExpression_Title" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute will have no effect; it is overridden by the CallerFilePathAttribute</value> | ||
| </data> | ||
| <data name="WRN_CallerLineNumberPreferredOverCallerArgumentExpression" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute applied to parameter '{0}' will have no effect. It is overridden by the CallerLineNumberAttribute.</value> | ||
| </data> | ||
| <data name="WRN_CallerLineNumberPreferredOverCallerArgumentExpression_Title" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute will have no effect; it is overridden by the CallerLineNumberAttribute</value> | ||
| </data> | ||
| <data name="WRN_CallerMemberNamePreferredOverCallerArgumentExpression" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute applied to parameter '{0}' will have no effect. It is overriden by the CallerMemberNameAttribute.</value> | ||
|
There was a problem hiding this comment.
English nit: remove There was a problem hiding this comment. There was a problem hiding this comment. Should I do the same with existing warnings for consistency? e.g: or only remove "The" from the new warnings? There was a problem hiding this comment. Of course it is... I guess just leave it as is. There was a problem hiding this comment. What we should do, though, is add a There was a problem hiding this comment.
FWIW, I am not aware of a requirement to add comments like that. Obviously, a developer can add remarks if he or she would like to do so. In this particular case, however, I believe that the suggested remark would state the obvious - type names are not translated and I personally wouldn't add it. It is quite obvious that "CallerArgumentExpressionAttribute" is not a word. In reply to: 605055933 [](ancestors = 605055933) There was a problem hiding this comment. I think there was some issues with translation for things that looks "trivial" that shouldn't be translated. See also #48364 There was a problem hiding this comment. Also an example where "unmanaged" keyword was incorrectly translated, despite it's surrounded with single quotes which is a high indication it shouldn't be localized. roslyn/src/Compilers/CSharp/Portable/xlf/CSharpResources.ja.xlf Lines 170 to 174 in 02428e3 There was a problem hiding this comment.
This is your PR, feel free to add remarks that you feel appropriate. In reply to: 605664888 [](ancestors = 605664888) There was a problem hiding this comment. Well, let's check with the loc team. @cristianosuzuki77 Is a "Locked" comment required for <trans-unit id="WRN_CallerMemberNamePreferredOverCallerArgumentExpression">
<source>The CallerArgumentExpressionAttribute applied to parameter '{0}' will have no effect. It is overriden by the CallerMemberNameAttribute.</source>
<note />
</trans-unit> |
||
| </data> | ||
| <data name="WRN_CallerMemberNamePreferredOverCallerArgumentExpression_Title" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute will have no effect; it is overridden by the CallerMemberNameAttribute</value> | ||
| </data> | ||
| <data name="WRN_CallerArgumentExpressionAttributeHasInvalidParameterName" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute applied to parameter '{0}' will have no effect. It is applied with an invalid parameter name.</value> | ||
| </data> | ||
| <data name="WRN_CallerArgumentExpressionAttributeHasInvalidParameterName_Title" xml:space="preserve"> | ||
| <value>The CallerArgumentExpressionAttribute is applied with an invalid parameter name.</value> | ||
| </data> | ||
| <data name="IDS_FeatureCallerArgumentExpression" xml:space="preserve"> | ||
| <value>caller argument expression</value> | ||
| </data> | ||
| </root> | ||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this condition ever true? #Closed
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@AlekseyTs Yes. When the arguments order matches the parameters order. I don't know the reason, only figured that by testing.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just to confirm. We have omitted arguments and argsToParamsOpt.IsDefault can still be true?
In reply to: 600234291 [](ancestors = 600234291)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not sure what you mean by "We have omitted arguments". But yes I confirmed it can be true.
For example, for the following test: