Mhurd has submitted this change and it was merged.
Change subject: Address nested parentheses in Share a Fact
......................................................................
Address nested parentheses in Share a Fact
Additionally, trim whitespace before semicolons.
Also, add tests for various utility instance methods.
Finally, remove the circle R from overlay per Moiz and Legal.
Bug T91846
Change-Id: I3c26877c7e4e220e84af9f192423384ea3b5c64e
---
M WikipediaUnitTests/NSString+WMFHTMLParsingTests.m
M wikipedia/Categories/NSString+WMFHTMLParsing.h
M wikipedia/Categories/NSString+WMFHTMLParsing.m
M wikipedia/View Controllers/ShareCard/ShareCard.xib
4 files changed, 177 insertions(+), 95 deletions(-)
Approvals:
Fjalapeno: Looks good to me, but someone else must approve
Mhurd: Looks good to me, approved
Bgerstle: Looks good to me, but someone else must approve
jenkins-bot: Verified
diff --git a/WikipediaUnitTests/NSString+WMFHTMLParsingTests.m
b/WikipediaUnitTests/NSString+WMFHTMLParsingTests.m
index 698180d..82de594 100644
--- a/WikipediaUnitTests/NSString+WMFHTMLParsingTests.m
+++ b/WikipediaUnitTests/NSString+WMFHTMLParsingTests.m
@@ -33,8 +33,52 @@
}
- (void)testAdequateSnippet {
- NSString *string = @"<p>Dog (woof) [horse] adequately long string</p>";
- XCTAssertEqualObjects([string wmf_getStringSnippetWithoutHTML], @"Dog
adequately long string");
+ NSString *string = @"<p>Dog (woof (w00t)) [horse] adequately long string
historically 40 characters.</p>";
+ XCTAssertEqualObjects([string wmf_getStringSnippetWithoutHTML],
+ @"Dog adequately long string historically 40
characters.");
+}
+
+- (void)testConsecutiveNewlinesCollapsing {
+ NSString *string = @"\n\nHola\n\n";
+ XCTAssertEqualObjects([string wmf_stringByCollapsingConsecutiveNewlines],
+ @"\nHola\n");
+}
+
+- (void)testNestedParenthesesRemoval {
+ NSString *string = @"He(a(b(c(d)e)f)g)llo";
+ XCTAssertEqualObjects([string
wmf_stringByRecursivelyRemovingParenthesizedContent],
+ @"Hello");
+}
+
+- (void)testBracketedContentRemoval {
+ NSString *string = @"J[aeio]ump";
+ XCTAssertEqualObjects([string wmf_stringByRemovingBracketedContent],
+ @"Jump");
+}
+
+- (void)testRemovalOfSpaceBeforeCommaAndSemicolon
+{
+ NSString *string = @"fish , squids ; eagles , crows";
+ XCTAssertEqualObjects([string
wmf_stringByRemovingWhiteSpaceBeforeCommasAndSemicolons],
+ @"fish, squids; eagles, crows");
+}
+
+- (void)testRemovalOfSpaceBeforePeriod {
+ NSString *string = @"Yes . No 。 Maybe ． So ｡";
+ XCTAssertEqualObjects([string wmf_stringByRemovingWhiteSpaceBeforePeriod],
+ @"Yes. No。 Maybe． So｡");
+}
+
+- (void)testConsecutiveSpacesCollapsing {
+ NSString *string = @" Metal ";
+ XCTAssertEqualObjects([string wmf_stringByCollapsingConsecutiveSpaces],
+ @" Metal ");
+}
+
+- (void)testRemovalOfLeadingOrTrailingSpacesNewlinesOrColons {
+ NSString *string = @"\n Syncopation:\n:";
+ XCTAssertEqualObjects([string
wmf_stringByRemovingLeadingOrTrailingSpacesNewlinesOrColons],
+ @"Syncopation");
}
@end
diff --git a/wikipedia/Categories/NSString+WMFHTMLParsing.h
b/wikipedia/Categories/NSString+WMFHTMLParsing.h
index 4eeb7e4..742406c 100644
--- a/wikipedia/Categories/NSString+WMFHTMLParsing.h
+++ b/wikipedia/Categories/NSString+WMFHTMLParsing.h
@@ -27,4 +27,12 @@
*/
- (NSString*)wmf_getStringSnippetWithoutHTML;
+- (NSString*)wmf_stringByCollapsingConsecutiveNewlines;
+- (NSString*)wmf_stringByRecursivelyRemovingParenthesizedContent;
+- (NSString*)wmf_stringByRemovingBracketedContent;
+- (NSString*)wmf_stringByRemovingWhiteSpaceBeforeCommasAndSemicolons;
+- (NSString*)wmf_stringByRemovingWhiteSpaceBeforePeriod;
+- (NSString*)wmf_stringByCollapsingConsecutiveSpaces;
+- (NSString*)wmf_stringByRemovingLeadingOrTrailingSpacesNewlinesOrColons;
+
@end
diff --git a/wikipedia/Categories/NSString+WMFHTMLParsing.m
b/wikipedia/Categories/NSString+WMFHTMLParsing.m
index 9c06619..75c0579 100644
--- a/wikipedia/Categories/NSString+WMFHTMLParsing.m
+++ b/wikipedia/Categories/NSString+WMFHTMLParsing.m
@@ -41,82 +41,124 @@
result : nil;
}
+
+#pragma mark - String simplification and cleanup
+ (NSString*)wmf_stringSnippetSimplifiedInString:(NSString*)string {
- NSString* result = [string
stringByReplacingOccurrencesOfString:@"&" withString:@"&"];
- NSError* err = nil;
- NSRegularExpression* newlinesRegex = [NSRegularExpression
-
regularExpressionWithPattern:@"\n{2,}"
- options:0
- error:&err];
- NSRange range = NSMakeRange(0, result.length);
- result = [newlinesRegex stringByReplacingMatchesInString:result
- options:0
- range:range
- withTemplate:@"\n"];
+ NSString* result = [string stringByReplacingOccurrencesOfString:@"&"
withString:@"&"];
+ result = [string stringByReplacingOccurrencesOfString:@">"
withString:@">"];
+ result = [string stringByReplacingOccurrencesOfString:@"<"
withString:@"<"];
+ result = [result wmf_stringByCollapsingConsecutiveNewlines];
+ result = [result wmf_stringByRecursivelyRemovingParenthesizedContent];
+ result = [result wmf_stringByRemovingBracketedContent];
+ result = [result wmf_stringByRemovingWhiteSpaceBeforeCommasAndSemicolons];
+ result = [result wmf_stringByRemovingWhiteSpaceBeforePeriod];
+ result = [result wmf_stringByCollapsingConsecutiveSpaces];
+ return [result
wmf_stringByRemovingLeadingOrTrailingSpacesNewlinesOrColons];
+}
-
- // We probably don't want to try to handle ideographic parens
- err = nil;
- NSRegularExpression* parensRegex = [NSRegularExpression
-
regularExpressionWithPattern:@"[(][^)]+[)]"
- options:0
- error:&err];
- range = NSMakeRange(0, result.length);
- result = [parensRegex stringByReplacingMatchesInString:result
+- (NSString*)wmf_stringByCollapsingConsecutiveNewlines {
+ static NSRegularExpression* newlinesRegex;
+ if (!newlinesRegex) {
+ newlinesRegex = [NSRegularExpression
+ regularExpressionWithPattern:@"\n{2,}"
+ options:0
+ error:nil];
+ }
+ return [newlinesRegex stringByReplacingMatchesInString:self
options:0
- range:range
- withTemplate:@""];
+ range:NSMakeRange(0,
self.length)
+ withTemplate:@"\n"];
+}
- // Nor do we want to try to handle ideographic brackets
- err = nil;
- NSRegularExpression* bracketsRegex = [NSRegularExpression
-
regularExpressionWithPattern:@"\\[[^]]+]"
- options:0
- error:&err];
- range = NSMakeRange(0, result.length);
- result = [bracketsRegex stringByReplacingMatchesInString:result
- options:0
- range:range
- withTemplate:@""];
+- (NSString*)wmf_stringByRecursivelyRemovingParenthesizedContent {
+ // We probably don't want to handle ideographic parens
+ static NSRegularExpression* parensRegex;
+ if (!parensRegex) {
+ parensRegex = [NSRegularExpression
+ regularExpressionWithPattern:@"[(][^()]+[)]"
+ options:0
+ error:nil];
+ }
+ NSString* string = [self copy];
+ NSString* oldResult;
+ NSRange range;
+ do {
+ oldResult = [string copy];
+ range = NSMakeRange(0, string.length);
+ string = [parensRegex stringByReplacingMatchesInString:string
+ options:0
+ range:range
+ withTemplate:@""];
+ } while (![oldResult isEqualToString:string]);
+ return string;
+}
+
+- (NSString*)wmf_stringByRemovingBracketedContent {
+ // We don't care about ideographic brackets
+ // Nested bracketing unseen thus far
+ static NSRegularExpression* bracketedRegex;
+ if (!bracketedRegex) {
+ bracketedRegex = [NSRegularExpression
+ regularExpressionWithPattern:@"\\[[^]]+]"
+ options:0
+ error:nil];
+ }
+ return [bracketedRegex stringByReplacingMatchesInString:self
+ options:0
+ range:NSMakeRange(0,
self.length)
+ withTemplate:@""];
+}
+
+- (NSString*)wmf_stringByRemovingWhiteSpaceBeforeCommasAndSemicolons {
// Unlike parens and brackets and unlike doubled up space in general,
- // we do not want whitespace preceding the comma or ideographic comma
- err = nil;
- NSRegularExpression* whitespaceCommaRegex = [NSRegularExpression
-
regularExpressionWithPattern:@"\\s+([,、])"
- options:0
-
error:&err];
- range = NSMakeRange(0, result.length);
- result = [whitespaceCommaRegex stringByReplacingMatchesInString:result
- options:0
- range:range
- withTemplate:@"$1"];
+ // we do not want whitespace preceding the comma, ideographic comma,
+ // or semicolon
+ static NSRegularExpression* spaceCommaColonRegex;
+ if (!spaceCommaColonRegex) {
+ spaceCommaColonRegex = [NSRegularExpression
+ regularExpressionWithPattern:@"\\s+([,、;])"
+ options:0
+ error:nil];
+ }
+ return [spaceCommaColonRegex stringByReplacingMatchesInString:self
+ options:0
+
range:NSMakeRange(0, self.length)
+ withTemplate:@"$1"];
+}
+- (NSString*)wmf_stringByRemovingWhiteSpaceBeforePeriod {
// Ideographic stops from TextExtracts, which were from OpenSearch
- err = nil;
- NSRegularExpression* whitespacePeriodRegex = [NSRegularExpression
-
regularExpressionWithPattern:@"\\s+([\\.|。|．|｡])"
-
options:0
-
error:&err];
- range = NSMakeRange(0, result.length);
- result = [whitespacePeriodRegex stringByReplacingMatchesInString:result
- options:0
- range:range
- withTemplate:@"$1"];
+ static NSRegularExpression* spacePeriodRegex;
+ if (!spacePeriodRegex) {
+ spacePeriodRegex = [NSRegularExpression
+
regularExpressionWithPattern:@"\\s+([\\.|。|．|｡])"
+ options:0
+ error:nil];
+ }
+ return [spacePeriodRegex stringByReplacingMatchesInString:self
+ options:0
+
range:NSMakeRange(0, self.length)
+ withTemplate:@"$1"];
+}
+- (NSString*)wmf_stringByCollapsingConsecutiveSpaces {
// In practice, we rarely care about doubled up whitespace in the
// string except for the actual space character
- err = nil;
- NSRegularExpression* spacesRegex = [NSRegularExpression
- regularExpressionWithPattern:@" {2,}"
- options:0
- error:&err];
- range = NSMakeRange(0, result.length);
- result = [spacesRegex stringByReplacingMatchesInString:result
- options:0
- range:range
- withTemplate:@" "];
+ static NSRegularExpression* spacesRegex;
+ if (!spacesRegex) {
+ spacesRegex = [NSRegularExpression
+ regularExpressionWithPattern:@" {2,}"
+ options:0
+ error:nil];
+ }
+ return [spacesRegex stringByReplacingMatchesInString:self
+ options:0
+ range:NSMakeRange(0,
self.length)
+ withTemplate:@" "];
+}
+- (NSString*)wmf_stringByRemovingLeadingOrTrailingSpacesNewlinesOrColons {
// Note about trailing colon characters: they usually look strange if kept,
// and removing them (plus spaces and newlines) doesn't often create merged
// words that look bad - these are usually at tag boundaries. For Latinized
@@ -124,18 +166,18 @@
// But as a matter of markup structure, something like a <p> tag
// shouldn't be </p> closed until something like <ul>...</ul> is closed.
// In fact, some sections have this layout, and some do not.
- err = nil;
- NSRegularExpression* leadingTrailingWhitespaceNewlineRegex =
[NSRegularExpression
-
regularExpressionWithPattern:@"^[\\s\n]+|[\\s\n:]+$"
-
options:0
-
error:&err];
- range = NSMakeRange(0, result.length);
- result = [leadingTrailingWhitespaceNewlineRegex
stringByReplacingMatchesInString:result
-
options:0
-
range:range
-
withTemplate:@""];
-
- return result;
+ static NSRegularExpression* leadTrailColonRegex;
+ if (!leadTrailColonRegex) {
+ leadTrailColonRegex = [NSRegularExpression
+
regularExpressionWithPattern:@"^[\\s\n]+|[\\s\n:]+$"
+ options:0
+ error:nil];
+ }
+ return [leadTrailColonRegex stringByReplacingMatchesInString:self
+ options:0
+
range:NSMakeRange(0, self.length)
+ withTemplate:@""];
}
+
@end
diff --git a/wikipedia/View Controllers/ShareCard/ShareCard.xib
b/wikipedia/View Controllers/ShareCard/ShareCard.xib
index b84bc4b..3bf301c 100644
--- a/wikipedia/View Controllers/ShareCard/ShareCard.xib
+++ b/wikipedia/View Controllers/ShareCard/ShareCard.xib
@@ -1,8 +1,8 @@
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<document type="com.apple.InterfaceBuilder3.CocoaTouch.XIB" version="3.0"
toolsVersion="6254" systemVersion="13F34" targetRuntime="iOS.CocoaTouch"
propertyAccessControl="none" useAutolayout="YES" useTraitCollections="YES">
+<document type="com.apple.InterfaceBuilder3.CocoaTouch.XIB" version="3.0"
toolsVersion="6751" systemVersion="13F1066" targetRuntime="iOS.CocoaTouch"
propertyAccessControl="none" useAutolayout="YES" useTraitCollections="YES">
<dependencies>
<deployment identifier="iOS"/>
- <plugIn identifier="com.apple.InterfaceBuilder.IBCocoaTouchPlugin"
version="6247"/>
+ <plugIn identifier="com.apple.InterfaceBuilder.IBCocoaTouchPlugin"
version="6736"/>
</dependencies>
<objects>
<placeholder placeholderIdentifier="IBFilesOwner" id="-1"
userLabel="File's Owner" customClass="WMFShareCardViewController">
@@ -45,7 +45,7 @@
</constraints>
</imageView>
<imageView userInteractionEnabled="NO"
contentMode="scaleAspectFit" horizontalHuggingPriority="251"
verticalHuggingPriority="251" image="Wikipedia_wordmark_gray.png"
translatesAutoresizingMaskIntoConstraints="NO" id="6HC-uN-YYr"
userLabel="Wikipedia Text Logo">
- <rect key="frame" x="437" y="300" width="165" height="42"/>
+ <rect key="frame" x="446" y="300" width="165" height="42"/>
<constraints>
<constraint firstAttribute="width" constant="165"
id="0M9-6P-ngj"/>
<constraint firstAttribute="height" constant="42"
id="lae-oj-fPO"/>
@@ -90,16 +90,6 @@
<color key="shadowColor" white="0.33333333333333331"
alpha="1" colorSpace="calibratedWhite"/>
<size key="shadowOffset" width="0.0" height="0.0"/>
</label>
- <label opaque="NO" userInteractionEnabled="NO"
contentMode="left" horizontalHuggingPriority="251"
verticalHuggingPriority="251" text="®" textAlignment="center"
lineBreakMode="tailTruncation" baselineAdjustment="alignBaselines"
adjustsFontSizeToFit="NO" preferredMaxLayoutWidth="14"
translatesAutoresizingMaskIntoConstraints="NO" id="5Yo-yI-rwI"
userLabel="Registered Character">
- <rect key="frame" x="595" y="306" width="14" height="14"/>
- <constraints>
- <constraint firstAttribute="height" constant="14"
id="pCZ-Xx-JCh"/>
- <constraint firstAttribute="width" constant="14"
id="zd9-Sv-7Sd"/>
- </constraints>
- <fontDescription key="fontDescription"
name="TimesNewRomanPSMT" family="Times New Roman" pointSize="12"/>
- <color key="textColor" red="0.87058823529411766"
green="0.87058823529411766" blue="0.87058823529411766" alpha="1"
colorSpace="calibratedRGB"/>
- <nil key="highlightedColor"/>
- </label>
</subviews>
<color key="backgroundColor" red="0.0" green="0.0" blue="0.0"
alpha="1" colorSpace="calibratedRGB"/>
<constraints>
@@ -110,17 +100,15 @@
<constraint firstAttribute="trailing" secondItem="ihv-br-w7Q"
secondAttribute="trailing" constant="594" id="Fa7-xI-sO6"/>
<constraint firstItem="QlQ-cG-u6Y" firstAttribute="top"
secondItem="iN0-l3-epB" secondAttribute="top" constant="1" id="I1H-MI-rGN"/>
<constraint firstItem="6HC-uN-YYr" firstAttribute="top"
secondItem="iN0-l3-epB" secondAttribute="top" constant="300" id="IRD-c1-xQ1"/>
- <constraint firstItem="6HC-uN-YYr" firstAttribute="leading"
secondItem="iN0-l3-epB" secondAttribute="leading" constant="437"
id="Jth-di-SuC"/>
<constraint firstAttribute="trailing" secondItem="dbe-nb-kfi"
secondAttribute="trailing" constant="250" id="PFY-gP-kCk"/>
<constraint firstAttribute="trailing" secondItem="rjC-wW-yPe"
secondAttribute="trailing" constant="250" id="SfX-Oa-xFJ"/>
- <constraint firstItem="5Yo-yI-rwI" firstAttribute="top"
secondItem="iN0-l3-epB" secondAttribute="top" constant="306" id="UqO-oR-9Ri"/>
<constraint firstItem="jvd-fN-IRx" firstAttribute="leading"
secondItem="iN0-l3-epB" secondAttribute="leading" constant="66"
id="WCt-t0-W80"/>
<constraint firstAttribute="trailing" secondItem="QlQ-cG-u6Y"
secondAttribute="trailing" constant="30" id="Y9l-1g-Oo7"/>
- <constraint firstItem="5Yo-yI-rwI" firstAttribute="left"
secondItem="6HC-uN-YYr" secondAttribute="right" constant="-7" id="fhg-D4-4AE"/>
- <constraint firstAttribute="trailing" secondItem="6HC-uN-YYr"
secondAttribute="trailing" constant="38" id="jVI-Mw-VWG"/>
+ <constraint firstAttribute="trailing" secondItem="6HC-uN-YYr"
secondAttribute="trailing" constant="29" id="dXe-Br-38A"/>
<constraint firstItem="ihv-br-w7Q" firstAttribute="top"
secondItem="dbe-nb-kfi" secondAttribute="bottom" constant="5" id="n83-t1-RUk"/>
<constraint firstItem="rjC-wW-yPe" firstAttribute="leading"
secondItem="iN0-l3-epB" secondAttribute="leading" constant="30"
id="nv3-ra-Qgn"/>
<constraint firstAttribute="trailing" secondItem="Sjf-Tp-DH9"
secondAttribute="trailing" constant="576" id="ooM-aW-c1A"/>
+ <constraint firstItem="6HC-uN-YYr" firstAttribute="leading"
secondItem="iN0-l3-epB" secondAttribute="leading" constant="446"
id="r6f-yl-9co"/>
<constraint firstItem="Sjf-Tp-DH9" firstAttribute="top"
secondItem="iN0-l3-epB" secondAttribute="top" constant="319" id="sf5-h3-ktL"/>
<constraint firstItem="dbe-nb-kfi" firstAttribute="leading"
secondItem="iN0-l3-epB" secondAttribute="leading" constant="30"
id="si1-eO-mqn"/>
<constraint firstItem="ihv-br-w7Q" firstAttribute="top"
secondItem="iN0-l3-epB" secondAttribute="top" constant="319" id="uzH-it-j9k"/>
--
To view, visit https://gerrit.wikimedia.org/r/196298
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I3c26877c7e4e220e84af9f192423384ea3b5c64e
Gerrit-PatchSet: 7
Gerrit-Project: apps/ios/wikipedia
Gerrit-Branch: master
Gerrit-Owner: Dr0ptp4kt <[email protected]>
Gerrit-Reviewer: Bgerstle <[email protected]>
Gerrit-Reviewer: Dr0ptp4kt <[email protected]>
Gerrit-Reviewer: Fjalapeno <[email protected]>
Gerrit-Reviewer: Mhurd <[email protected]>
Gerrit-Reviewer: jenkins-bot <>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
