This repository has been archived by the owner on Dec 15, 2022. It is now read-only.
Provide additional stop conditions for 'anonymous-classes-and-new' pattern #214
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
Vigilans
changed the title
Vigilans
changed the title
sadikovi
reviewed
Nov 27, 2019
spec/java-spec.coffee
Outdated
| @@ -1755,6 +1768,41 @@ describe 'Java grammar', -> | |||
| expect(lines[7][19]).toEqual value: 'start', scopes: expected | |||
| expect(lines[8][19]).toEqual value: 'start', scopes: expected | |||
| expect(lines[9][19]).toEqual value: 'start', scopes: expected | |||
| expect(lines[10][19]).toEqual value: 'start', scopes: expected | |||
|
|
|||
| # See issue https://github.com/atom/language-java/issues/180 | |||
| expect(lines[8][1]).toEqual value: 'g', scopes: expected | ||
| expect(lines[9][1]).toEqual value: 'g', scopes: expected | ||
| expect(lines[10][1]).toEqual value: 'g', scopes: expected | ||
| expect(lines[15][3]).toEqual value: 'a', scopes: ['source.java', 'meta.class.java', 'meta.class.body.java', 'meta.method.java', 'meta.method.body.java', 'meta.definition.variable.java', 'variable.other.definition.java'] |
There was a problem hiding this comment.
Does this test fail without your change?
There was a problem hiding this comment.
See #180, it will produce a nested scope like this for line[15][3]:
variable.other.definition.java
meta.definition.variable.java
meta.function-call.java
meta.function-call.java
meta.function-call.java
meta.function-call.java
meta.function-call.java
meta.function-call.java
meta.function-call.java
meta.function-call.java
meta.method.body.java
meta.method.java
meta.class.body.java
meta.class.java
source.java
There was a problem hiding this comment.
I am well aware of the issue. I was only asking if you verified that the test fails without your changes.
There was a problem hiding this comment.
I see. I test it by removing \\[ in the end regex, and it produces such failures:
|
Please, enable edits from maintainers on your pull request. |
sadikovi
changed the title
|
Merging to master, thanks! |
sadikovi
pushed a commit
that referenced
this pull request
Mar 25, 2020
### Description of the Change Similar to #212, it uses a two-scope pattern to distinguish `inner-class` block and `array-initializer` block. it is based on #214 to provide correct end regex for `inner-class` and `array-initializer`. #### Detailed description Conside how a syntax scope terminates: ``` new Class() <end of new expression> ^~~~~ terminate here new Class[] <end of new expression> ^~~~~ terminate here new Class() {} <end of new expression> ^~~~~ terminate here new Class[] {} <end of new expression> ^~~~~ terminate here ``` Previous new-expr grammar handles following ways of termination: * inner-class: `(?<!\\])\\s*({)`...`(})` * parens (square): `[`...`]` * parens (round): `(`...`)` * parens (square + curly): `[`...`]`...`{`...`}` The usage of `\s*` suffers from new-line and comment-in-between pattern, and this should be relaxed using a two-scope model. Besides, `inner-class` and `array-initializer` do not share same syntax (`class-body` and `code`), so they should fall in different scope. So, this PR: * Differentiate `object-new` and `array-declaration` using two different scopes. * Put `inner-class` and `array-initializer` in different scopes, with similar names. ### Alternate Designs This PR repetitively uses the end regex of new expression: ``` 'end': '(?=;|\\)|\\]|\\.|,|\\?|:|}|\\+|\\-|\\*|\\/(?!\\/|\\*)|%|!|&|\\||\\^|=)' ``` to determine when a scope terminates. It would be better if we could find a way to reduce its usage. One possible alternative is: match when inner end token reached (`)`, `]` or `}`), not the outer token that ends it (the complex end regex). But just like #214, it is hard to tell when `)` or `]` is met, whether there will be a `{}` block next to it. I (maybe) tried such silly regexes (cannot remember them correctly): * outer - `(?<=\\)`...`}`, inner: `{`...`(?=}` The problem with it: if there's no `}` after `)`, it won't ends, causing the whole `new-expr` scope won't ends. e.g. `A[] arr = new A[]{ new A(), new A() };` - the comma won't ends first `new A()` because inner scope eats until `}` is met. ### Benefits * Can correctly highlight `array-initializer` when there's any distance of spaces between ']' and '{'. * Can correctly highlight `array-initializer` when '{' is placed at new line or with a comment in between. * Introduced a new scope `array-initializer`, in correspondence with `inner-class`. * Patterns are more neat and uniform. ### Possible Drawbacks * Undetected possible regression issues. ### Applicable Issues Fix #172 Fix #199 Fix redhat-developer/vscode-java#728
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Description of the Change
Dot should also end 'anonymous-classes-and-new', in order to propery highlight method call of temporary variable created by 'new'.
Question mark should also end 'anonymous-classes-and-new', in order to proper break out circumstances like
new Random() ? true : false;Right square brace should also end 'anonymous-classes-and-new', for circumstances like
[new Random()].Xor should also end
anonymous-classes-and-new, for circumstances likenew Integer(1) ^ new Integer(2).Alternate Designs
Non was considered.Having to enumerate these much punctuations made me think of proposing a general and simple regex to end new expression. But soon I found out that I couldn't find a way to tell when
)or]is met, whether there will be a{}block next to it.Benefits
Possible Drawbacks
No obvious drawback.
Applicable Issues
Fix #213 -> dot
Fix #179 -> question mark
Fix #180 -> right square
Fix #192 -> xor