From 17467185533158b987c02988361a7ec054832b6e Mon Sep 17 00:00:00 2001 From: Amrouche Hamza Date: Mon, 8 Jul 2019 18:46:19 +0200 Subject: [PATCH] [DomCrawler] add a value() method, normalize whitespaces --- src/Symfony/Component/DomCrawler/CHANGELOG.md | 1 + src/Symfony/Component/DomCrawler/Crawler.php | 14 +++++++++++--- .../DomCrawler/Tests/AbstractCrawlerTest.php | 12 ++++++++++++ 3 files changed, 24 insertions(+), 3 deletions(-) diff --git a/src/Symfony/Component/DomCrawler/CHANGELOG.md b/src/Symfony/Component/DomCrawler/CHANGELOG.md index ad8c47982ea62..56863b7c73735 100644 --- a/src/Symfony/Component/DomCrawler/CHANGELOG.md +++ b/src/Symfony/Component/DomCrawler/CHANGELOG.md @@ -8,6 +8,7 @@ CHANGELOG * Added `Crawler::matches()` method. * Added `Crawler::closest()` method. * Added `Crawler::outerHtml()` method. +* Added an argument to the `Crawler::text()` method to opt-in normalizing whitespaces. 4.3.0 ----- diff --git a/src/Symfony/Component/DomCrawler/Crawler.php b/src/Symfony/Component/DomCrawler/Crawler.php index 169b056e4a572..a1975644eec29 100644 --- a/src/Symfony/Component/DomCrawler/Crawler.php +++ b/src/Symfony/Component/DomCrawler/Crawler.php @@ -591,7 +591,9 @@ public function nodeName() } /** - * Returns the node value of the first node of the list. + * Returns the text of the first node of the list. + * + * Pass true as the 2nd argument to normalize whitespaces. * * @param mixed $default When provided and the current node is empty, this value is returned and no exception is thrown * @@ -599,7 +601,7 @@ public function nodeName() * * @throws \InvalidArgumentException When current node is empty */ - public function text(/* $default = null */) + public function text(/* $default = null, $normalizeWhitespace = true */) { if (!$this->nodes) { if (0 < \func_num_args()) { @@ -609,7 +611,13 @@ public function text(/* $default = null */) throw new \InvalidArgumentException('The current node list is empty.'); } - return $this->getNode(0)->nodeValue; + $text = $this->getNode(0)->nodeValue; + + if (\func_num_args() > 1 && func_get_arg(1)) { + return trim(preg_replace('/(?:\s{2,}+|[^\S ])/', ' ', $text)); + } + + return $text; } /** diff --git a/src/Symfony/Component/DomCrawler/Tests/AbstractCrawlerTest.php b/src/Symfony/Component/DomCrawler/Tests/AbstractCrawlerTest.php index 8de52465a8565..4a60ca83b0b4b 100644 --- a/src/Symfony/Component/DomCrawler/Tests/AbstractCrawlerTest.php +++ b/src/Symfony/Component/DomCrawler/Tests/AbstractCrawlerTest.php @@ -253,6 +253,14 @@ public function testEq() $this->assertCount(0, $crawler->eq(100), '->eq() returns an empty crawler if the nth node does not exist'); } + public function testNormalizeWhiteSpace() + { + $crawler = $this->createTestCrawler()->filterXPath('//p'); + $this->assertSame('Elsa <3', $crawler->text(null, true), '->text(null, true) returns the text with normalized whitespace'); + $this->assertNotSame('Elsa <3', $crawler->text(null, false)); + $this->assertNotSame('Elsa <3', $crawler->text()); + } + public function testEach() { $data = $this->createTestCrawler()->filterXPath('//ul[1]/li')->each(function ($node, $i) { @@ -1291,6 +1299,10 @@ public function createTestCrawler($uri = null)

Two Bis

Three Bis

+ Elsa + <3 +

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.

Alternative Proxies: