Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[SuperEditorSpellCheck] Add API to ignore text during spellcheck (Resolves #2564) #2567

Merged
merged 2 commits into from
Feb 11, 2025
Merged

[SuperEditorSpellCheck] Add API to ignore text during spellcheck (Resolves #2564) #2567

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
bde27b7
[SuperEditorSpellCheck] Add API to ignore text during spellcheck (Res…
angelosilvestre Feb 7, 2025
96476e0
Remove rule from method signature
angelosilvestre Feb 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
68 changes: 68 additions & 0 deletions super_editor_spellcheck/example/lib/main_super_editor_spellcheck.dart
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,6 +82,11 @@ class _SuperEditorSpellcheckScreenState extends State<_SuperEditorSpellcheckScre
_spellingAndGrammarPlugin = SpellingAndGrammarPlugin(
iosControlsController: _iosControlsController,
androidControlsController: _androidControlsController,
ignoreRules: [
SpellingIgnoreRules.byAttribution(boldAttribution),
SpellingIgnoreRules.byAttributionFilter((attr) => attr is LinkAttribution),
SpellingIgnoreRules.byPattern(RegExp(r'#\w+')),
],
);

_editor = createDefaultDocumentEditor(
Expand Down Expand Up @@ -119,6 +124,58 @@ class _SuperEditorSpellcheckScreenState extends State<_SuperEditorSpellcheckScre
attributions: {},
),
]);
_editor.execute([
InsertNodeAfterNodeRequest(
existingNodeId: _editor.context.document.last.id,
newNode: ParagraphNode(id: Editor.createNodeId(), text: AttributedText('')),
)
]);
_editor.execute([
InsertAttributedTextRequest(
DocumentPosition(
nodeId: _editor.context.document.last.id,
nodePosition: _editor.context.document.last.endPosition,
),
AttributedText(
'The spellchecking can be configured to ignore spelling errors for some situation, like links: https://www.populr.com, '
'tags: #framwork, or text with specific attributions, like bold attbution.',
AttributedSpans(
attributions: [
const SpanMarker(
attribution: LinkAttribution('https://www.populr.com'),
offset: 94,
markerType: SpanMarkerType.start,
),
const SpanMarker(
attribution: LinkAttribution('https://www.populr.com'),
offset: 115,
markerType: SpanMarkerType.end,
),
const SpanMarker(
attribution: PatternTagAttribution(),
offset: 124,
markerType: SpanMarkerType.start,
),
const SpanMarker(
attribution: PatternTagAttribution(),
offset: 132,
markerType: SpanMarkerType.end,
),
const SpanMarker(
attribution: boldAttribution,
offset: 176,
markerType: SpanMarkerType.start,
),
const SpanMarker(
attribution: boldAttribution,
offset: 189,
markerType: SpanMarkerType.end,
),
],
),
),
)
]);
});
}

Expand All @@ -133,6 +190,17 @@ class _SuperEditorSpellcheckScreenState extends State<_SuperEditorSpellcheckScre
autofocus: true,
editor: _editor,
stylesheet: defaultStylesheet.copyWith(
inlineTextStyler: (attributions, existingStyle) {
TextStyle style = defaultInlineTextStyler(attributions, existingStyle);

if (attributions.whereType<PatternTagAttribution>().isNotEmpty) {
style = style.copyWith(
color: Colors.orange,
);
}

return style;
},
addRulesAfter: [
if (Theme.of(context).brightness == Brightness.dark) ..._darkModeStyles,
],
Expand Down
160 changes: 144 additions & 16 deletions super_editor_spellcheck/lib/src/super_editor/spelling_and_grammar_plugin.dart
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
import 'dart:async';
import 'dart:ui';

import 'package:collection/collection.dart';
import 'package:flutter/foundation.dart';
import 'package:flutter/material.dart';
import 'package:flutter/services.dart';
Expand Down Expand Up @@ -33,6 +34,9 @@ class SpellingAndGrammarPlugin extends SuperEditorPlugin {
/// is required when running on Android.
/// - [iosControlsController]: the controls controller to use when running on iOS. This is
/// required when running on iOS.
/// - [ignoreRules]: a list of rules that determine ranges that should be ignored from spellchecking.
/// It can be used, for example, to ignore links or text with specific attributions. See [SpellingIgnoreRules]
/// for a list of built-in rules.
SpellingAndGrammarPlugin({
bool isSpellingCheckEnabled = true,
UnderlineStyle spellingErrorUnderlineStyle = defaultSpellingErrorUnderlineStyle,
Expand All @@ -42,6 +46,7 @@ class SpellingAndGrammarPlugin extends SuperEditorPlugin {
Color? selectedWordHighlightColor,
SuperEditorAndroidControlsController? androidControlsController,
SuperEditorIosControlsController? iosControlsController,
List<SpellingIgnoreRule> ignoreRules = const [],
}) : _isSpellCheckEnabled = isSpellingCheckEnabled,
_isGrammarCheckEnabled = isGrammarCheckEnabled {
assert(defaultTargetPlatform != TargetPlatform.android || androidControlsController != null,
Expand All @@ -66,6 +71,8 @@ class SpellingAndGrammarPlugin extends SuperEditorPlugin {
: null),
);

_ignoreRules = ignoreRules;

_contentTapHandler = switch (defaultTargetPlatform) {
TargetPlatform.android => SuperEditorAndroidSpellCheckerTapHandler(
popoverController: _popoverController,
Expand All @@ -89,6 +96,8 @@ class SpellingAndGrammarPlugin extends SuperEditorPlugin {
/// misspelled word.
final _selectedWordLink = LeaderLink();

late final List<SpellingIgnoreRule> _ignoreRules;

late final SpellingAndGrammarReaction _reaction;

/// Whether this reaction checks spelling in the document.
Expand Down Expand Up @@ -138,7 +147,7 @@ class SpellingAndGrammarPlugin extends SuperEditorPlugin {
editor.context.put(spellingErrorSuggestionsKey, _spellingErrorSuggestions);
_contentTapHandler?.editor = editor;

_reaction = SpellingAndGrammarReaction(_spellingErrorSuggestions, _styler);
_reaction = SpellingAndGrammarReaction(_spellingErrorSuggestions, _styler, _ignoreRules);
editor.reactionPipeline.add(_reaction);

// Do initial spelling and grammar analysis, in case the document already
Expand Down Expand Up @@ -245,12 +254,14 @@ extension SpellingAndGrammarEditorExtensions on Editor {
/// An [EditReaction] that runs spelling and grammar checks on all [TextNode]s
/// in a given [Document].
class SpellingAndGrammarReaction implements EditReaction {
SpellingAndGrammarReaction(this._suggestions, this._styler);
SpellingAndGrammarReaction(this._suggestions, this._styler, this._ignoreRules);

final SpellingErrorSuggestions _suggestions;

final SpellingAndGrammarStyler _styler;

final List<SpellingIgnoreRule> _ignoreRules;

bool isSpellCheckEnabled = true;

set spellingErrorUnderlineStyle(UnderlineStyle style) => _styler.spellingErrorUnderlineStyle = style;
Expand All @@ -274,12 +285,13 @@ class SpellingAndGrammarReaction implements EditReaction {
/// Checks every [TextNode] in the given document for spelling and grammar
/// errors and stores them for visual styling.
void analyzeWholeDocument(EditContext editorContext) {
final document = editorContext.document;
for (final node in editorContext.document) {
if (node is! TextNode) {
continue;
}

_findSpellingAndGrammarErrors(node);
_findSpellingAndGrammarErrors(node, document);
}
}

Expand Down Expand Up @@ -351,19 +363,19 @@ class SpellingAndGrammarReaction implements EditReaction {
continue;
}

_findSpellingAndGrammarErrors(textNode);
_findSpellingAndGrammarErrors(textNode, document);
}
}

Future<void> _findSpellingAndGrammarErrors(TextNode textNode) async {
Future<void> _findSpellingAndGrammarErrors(TextNode textNode, Document document) async {
if (defaultTargetPlatform == TargetPlatform.macOS) {
await _findSpellingAndGrammarErrorsOnMac(textNode);
await _findSpellingAndGrammarErrorsOnMac(textNode, document);
} else if (defaultTargetPlatform == TargetPlatform.android || defaultTargetPlatform == TargetPlatform.iOS) {
await _findSpellingAndGrammarErrorsOnMobile(textNode);
await _findSpellingAndGrammarErrorsOnMobile(textNode, document);
}
}

Future<void> _findSpellingAndGrammarErrorsOnMac(TextNode textNode) async {
Future<void> _findSpellingAndGrammarErrorsOnMac(TextNode textNode, Document document) async {
// TODO: Investigate whether we can parallelize spelling and grammar checks
// for a given node (and whether it's worth the complexity).
final textErrors = <TextError>{};
Expand All @@ -374,6 +386,8 @@ class SpellingAndGrammarReaction implements EditReaction {
final requestId = _asyncRequestIds[textNode.id]! + 1;
_asyncRequestIds[textNode.id] = requestId;

final plainText = _filterIgnoredRanges(textNode, document);

int startingOffset = 0;
TextRange prevError = TextRange.empty;
final locale = PlatformDispatcher.instance.locale;
Expand All @@ -382,16 +396,16 @@ class SpellingAndGrammarReaction implements EditReaction {
if (isSpellCheckEnabled) {
do {
prevError = await _macSpellChecker.checkSpelling(
stringToCheck: textNode.text.text,
stringToCheck: plainText,
startingOffset: startingOffset,
language: language,
);

if (prevError.isValid) {
final word = textNode.text.text.substring(prevError.start, prevError.end);
final word = plainText.substring(prevError.start, prevError.end);

// Ask platform for spelling correction guesses.
final guesses = await _macSpellChecker.guesses(range: prevError, text: textNode.text.text);
final guesses = await _macSpellChecker.guesses(range: prevError, text: plainText);

textErrors.add(
TextError.spelling(
Expand Down Expand Up @@ -419,7 +433,7 @@ class SpellingAndGrammarReaction implements EditReaction {
prevError = TextRange.empty;
do {
final result = await _macSpellChecker.checkGrammar(
stringToCheck: textNode.text.text,
stringToCheck: plainText,
startingOffset: startingOffset,
language: language,
);
Expand All @@ -428,7 +442,7 @@ class SpellingAndGrammarReaction implements EditReaction {
if (prevError.isValid) {
for (final grammarError in result.details) {
final errorRange = grammarError.range;
final text = textNode.text.text.substring(errorRange.start, errorRange.end);
final text = plainText.substring(errorRange.start, errorRange.end);
textErrors.add(
TextError.grammar(
nodeId: textNode.id,
Expand Down Expand Up @@ -461,7 +475,7 @@ class SpellingAndGrammarReaction implements EditReaction {
_suggestions.putSuggestions(textNode.id, spellingSuggestions);
}

Future<void> _findSpellingAndGrammarErrorsOnMobile(TextNode textNode) async {
Future<void> _findSpellingAndGrammarErrorsOnMobile(TextNode textNode, Document document) async {
final textErrors = <TextError>{};
final spellingSuggestions = <TextRange, SpellingError>{};

Expand All @@ -471,16 +485,18 @@ class SpellingAndGrammarReaction implements EditReaction {
final requestId = _asyncRequestIds[textNode.id]! + 1;
_asyncRequestIds[textNode.id] = requestId;

final plainText = _filterIgnoredRanges(textNode, document);

final suggestions = await _mobileSpellChecker.fetchSpellCheckSuggestions(
PlatformDispatcher.instance.locale,
textNode.text.toPlainText(),
plainText,
);
if (suggestions == null) {
return;
}

for (final suggestion in suggestions) {
final misspelledWord = textNode.text.substring(suggestion.range.start, suggestion.range.end);
final misspelledWord = plainText.substring(suggestion.range.start, suggestion.range.end);
spellingSuggestions[suggestion.range] = SpellingError(
word: misspelledWord,
nodeId: textNode.id,
Expand Down Expand Up @@ -514,6 +530,83 @@ class SpellingAndGrammarReaction implements EditReaction {
// see suggestions and select them.
_suggestions.putSuggestions(textNode.id, spellingSuggestions);
}

/// Filters out ranges that should be ignored from spellchecking.
///
/// This method replaces the ignored ranges with whitespaces so that the spellchecker
/// doesn't see them.
String _filterIgnoredRanges(TextNode node, Document document) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we remove the document from this? This seems to spread the document around to a bunch of callsites, and in theory the broader document shouldn't be necessary to evaluate rules for patterns, attributions, or explicit text spans.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Removed.

final ranges = _ignoreRules //
.map((rule) => rule(node, document))
.expand((listOfRanges) => listOfRanges)
.toList();

final text = node.text.toPlainText();

if (ranges.isEmpty) {
// We don't have any ranges to remove, short circuit.
return text;
}

final buffer = StringBuffer();

final mergedRanges = _mergeOverlappingRanges(ranges);
int currentOffset = 0;
for (final range in mergedRanges) {
if (range.start > currentOffset) {
// We have text before the ignored range. Add it.
buffer.write(text.substring(currentOffset, range.start));
}

// Fill the ignored range with whitespaces.
buffer.write(' ' * (range.end - range.start));

currentOffset = range.end;
}

// Add the remaining text, after the last ignored range, if any.
if (currentOffset < text.length) {
buffer.write(text.substring(currentOffset));
}

return buffer.toString();
}

/// Merges overlapping ranges in the given list of [ranges].
///
/// Returns a new sorted list of ranges where overlapping ranges are merged.
List<TextRange> _mergeOverlappingRanges(List<TextRange> ranges) {
final sortedRanges = ranges.sorted((a, b) {
if (a.start < b.start) {
return -1;
} else if (a.start > b.start) {
return 1;
}

return a.end - b.end;
});

TextRange currentRange = sortedRanges.first;

final mergedRanges = <TextRange>[];
for (int i = 1; i < sortedRanges.length; i++) {
final nextRange = sortedRanges[i];
if (currentRange.end >= nextRange.start) {
// The ranges overlap, merge them.
currentRange = TextRange(
start: currentRange.start,
end: nextRange.end,
);
} else {
// The ranges don't overlap.
mergedRanges.add(currentRange);
currentRange = nextRange;
}
}
mergedRanges.add(currentRange);

return mergedRanges;
}
}

/// A [ContentTapDelegate] that shows the suggestions popover when the user taps on
Expand Down Expand Up @@ -785,3 +878,38 @@ class _SpellCheckerContentTapDelegate extends ContentTapDelegate {

Editor? editor;
}

/// A function that determines ranges to be ignored from spellchecking.
typedef SpellingIgnoreRule = List<TextRange> Function(TextNode node, Document document);

/// A collection of built-in rules for ignoring spans of text from spellchecking.
class SpellingIgnoreRules {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should add a rule for explicit text spans, just in case an app already knows the range of text to ignore.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Like we talked before, the knowledge about ignored ranges lives only in the SpellingAndGrammarReaction. So, a rule that takes a TextRange would apply that range for all nodes in the document. A rule to ignore a portion of the document would need to take a DocumentRange and a Document, to generate different TextRanges for each node, as the rules return the ignored range for a single TextNode.

Regarding adding option directly on our native spellchecker plugin, I'm not sure it's a good idea, because the plugin basically just exposes the macOS API's, so I think we shouldn't be adding arbitrary parameters.

/// Creates a rule that ignores text spans that match the given [pattern].
static SpellingIgnoreRule byPattern(Pattern pattern) {
return (TextNode node, Document document) {
return pattern
.allMatches(node.text.toPlainText())
.map((match) => TextRange(start: match.start, end: match.end))
.toList();
};
}

/// Creates a rule that ignores text spans that have the given [attribution].
static SpellingIgnoreRule byAttribution(Attribution attribution) {
return byAttributionFilter((candidate) => candidate == attribution);
}

/// Creates a rule that ignore text spans that have at least one atribution that matches the given [filter].
static SpellingIgnoreRule byAttributionFilter(AttributionFilter filter) {
return (TextNode node, Document document) {
return node.text.spans
.getAttributionSpansInRange(
attributionFilter: filter,
start: 0,
end: node.text.toPlainText().length - 1, // -1 to make end of range inclusive.
)
.map((span) => TextRange(start: span.start, end: span.end + 1)) // +1 to make the end exclusive.
.toList();
};
}
}
Loading