Bug 204330 – Web Inspector: Local Overrides: the placeholder for the MIME type, status code, and status text is the same as the placeholder URL

• Status: RESOLVED
• Resolution: FIXED
• Component: Web Inspector

[Devin Rousso]

# STEPS TO REPRODUCE:
1. inspect any page
2. click the + in the bottom of the navigation sidebar of the Sources Tab
3. select "Local Override..."
4. delete the contents of the MIME type, status code, or status text inputs
 => URL is shown, instead of the default value for those inputs

[Devin Rousso]

Created attachment 383807 [details]
Patch

[Devin Rousso]

Created attachment 383808 [details]
[Image] After Patch is applied

[Joseph Pecoraro]

Comment on attachment 383807 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=383807&action=review

Some questions.

> Source/WebInspectorUI/ChangeLog:14
> +        (.popover .local-resource-override-popover-content .data-grid tr.header-content-type > :matches(.name-column, .value-column)): Added.
> +        Replace the hardcoded `placeholder` with an optional `options` object that can include a

The user experience of only placeholders seems weird. I'd prefer it include values where it has values.

> Source/WebInspectorUI/UserInterface/Views/LocalResourceOverridePopover.js:52
> -        let url = this._urlCodeMirror.getValue();
> +        let url = this._urlCodeMirror.getValue() || this._urlCodeMirror.getOption("placeholder");
>          if (!url)
>              return null;

Not allowing a path and having it fallback to the placeholder seems bad. In the other cases it seems fine because those are reasonable default values, but this one ending in "/path" seems dumb.

I think we should continue to disallow an empty url.

> Source/WebInspectorUI/UserInterface/Views/LocalResourceOverridePopover.js:119
>          if (!mimeType)
> -            mimeType = "text/javascript";
> +            mimeType = "text/plain";

I'd rather we prefer the default be "text/javascript" since text files are very rare on web pages.

> Source/WebInspectorUI/UserInterface/Views/LocalResourceOverridePopover.js:336
> -            value,
> +            ...options,

Why did we get rid of the initial value?

    value: value

Unless I'm missing something, this doesn't seem preferable to setting this value because you can't edit / copy the placeholder.

If you want, maybe the URL field could start with an empty value but have a placeholder to force the user to type in a URL, but that seems weird because it requires typing a full URL path when we could have pre-generated part of it.

[Devin Rousso]

Comment on attachment 383807 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=383807&action=review

>> Source/WebInspectorUI/ChangeLog:14
>> +        Replace the hardcoded `placeholder` with an optional `options` object that can include a
> 
> The user experience of only placeholders seems weird. I'd prefer it include values where it has values.

I think we should do this if we're given a resource to copy/override, but not if we're creating an "empty" local override.

>> Source/WebInspectorUI/UserInterface/Views/LocalResourceOverridePopover.js:52
>>              return null;
> 
> Not allowing a path and having it fallback to the placeholder seems bad. In the other cases it seems fine because those are reasonable default values, but this one ending in "/path" seems dumb.
> 
> I think we should continue to disallow an empty url.

That makes sense.  Agreed.

>> Source/WebInspectorUI/UserInterface/Views/LocalResourceOverridePopover.js:119
>> +            mimeType = "text/plain";
> 
> I'd rather we prefer the default be "text/javascript" since text files are very rare on web pages.

I feel like that's a bit "biased" towards users mostly using this to override JavaScript vs other types (e.g. CSS, HTML, etc.), but I don't particularly care that much.

>> Source/WebInspectorUI/UserInterface/Views/LocalResourceOverridePopover.js:336
>> +            ...options,
> 
> Why did we get rid of the initial value?
> 
>     value: value
> 
> Unless I'm missing something, this doesn't seem preferable to setting this value because you can't edit / copy the placeholder.
> 
> If you want, maybe the URL field could start with an empty value but have a placeholder to force the user to type in a URL, but that seems weird because it requires typing a full URL path when we could have pre-generated part of it.

We didn't so much "get rid of it" as we allow the caller to decide whether they want to supply it or not.

As I wrote above, I think it makes a lot of sense to set a value when we know there is a value (e.g. overriding an existing resource), but if we're creating a brand new local override I think having them be placeholders is more "inviting" in that it encourages the user to fill in their own value.

[Devin Rousso]

Created attachment 383833 [details]
Patch

[Joseph Pecoraro]

Comment on attachment 383833 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=383833&action=review

rs=me

> Source/WebInspectorUI/UserInterface/Views/DataGridNode.js:37
> +        this._selectable = selectable !== undefined ? selectable : true;
> +        this._copyable = copyable !== undefined ? copyable : true;
> +        this._editable = editable !== undefined ? editable : true;

Can we use nullish operators now?

    selectable ?? true

[Devin Rousso]

Comment on attachment 383833 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=383833&action=review

>> Source/WebInspectorUI/UserInterface/Views/DataGridNode.js:37
>> +        this._editable = editable !== undefined ? editable : true;
> 
> Can we use nullish operators now?
> 
>     selectable ?? true

I'd rather wait for nullish coalescing to reach Stage 4 before starting to use it.  This would be a perfect use case though :)

[WebKit Commit Bot]

Comment on attachment 383833 [details]
Patch

Clearing flags on attachment: 383833

Committed r252652: <https://trac.webkit.org/changeset/252652>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/57338064>