问题描述
|
如果我有某个字符串输入，那么我想与另一个字符串进行比较，并使用最大可能的匹配值将输入字符串的匹配内容包装在另一个字符串中。如何最好地将火柴包裹在标签中？这是一个不平凡的问题。
基本上，我想将输入的字符串与另一个字符串匹配，使用跨度标记显示在输入的字符串中找到的目标的匹配部分。
从输入字符串开头开始匹配（最大匹配）
搜索词的部分匹配项应突出显示（例如，请参见\“ barge \”，\“ barges \”）
特殊字符符应与输入的\“ fred / dred \”相匹配，这将是两个字。
输入字符串将根据用户输入内容而有所不同。
从头开始匹配输入字符串
匹配出现的每个单词
如果用户输入包含多个单词的字符串，则我想从第二个字符串中出现的位置开始，以渐进方式包装它们的匹配项。它们在输入的字符串的开头/结尾可能有空格，也可能没有空格。我希望将最大的部分包裹起来。
输入字符串示例：
\"brown cats cannot be white cats\"
\"blue pigs \"
\"large,charged/marged barge pigs\"
\"<span class=\'wrapper\'>brown cats cannot be white cats</span>\"
\"Hi bill,brown cats cannot be white cats and cows are not blue pigs,blue melons are large but not batteries charged barges with white cats carry coal\"
\"Hi bill,<span class=\'wrapper\'>brown cats cannot be white cats</span> and cows are not blue pigs,blue melons are large but not batteries charged barges with <span class=\'wrapper\'>white cats</span> carry coal\"
\"Hi bill,brown cats cannot be white cats and cows are not <span class=\'wrapper\'>blue pigs</span>,blue melons are large but not batteries charged barges with white cats carry coal\"
\"Hi bill,brown cats cannot be white cats and cows are not blue <span class=\'wrapper\'>pigs</span>,blue melons are large but not batteries <span class=\'wrapper\'>charged</span> <span class=\'wrapper\'>barge</span>s with white cats carry coal\"
\"brown cats cannot be white cats\"
\"brown cats cannot be white\"
\"brown cats cannot be\"
\"brown cats cannot\"
\"brown cats\"
\"brown\"
\"brown\" \"cats\" \"cannot\" \"be\" \"white\" \"cats\"
function replaceWords(wordsy,text) {
var re = \'(\' + wordsy + \')(?![^<]*(?:<\\/script|>))\',regExp = new RegExp(re,\'ig\'),sTag = \"<span class=\'wrapper\'>\",eTag = \"</span>\";
return text.replace(regExp,sTag + \'$&\' + eTag);
};
var matchstring = \"brown cats cannot be white cats\";
var wrapstring = \"Hi bill,blue melons are large but not batteries charged barges with white cats carry coal\";
var words = myValue.split(\" \");
var i = words.length; while (i--) {
wrapstring = replaceWords(words[i],wrapstring );
};
解决方法
我有一个有趣的解决方案，应该可以作为您的原始规格。它尚未经过压力测试，我不确定是否要处理大量文本，并且它会进行很多正则表达式匹配。不一定是最干净或最简单的解决方案，但它可以正常工作。
特点和局限性：
它处理匹配字符串中最奇怪的情况，例如重复的单词，非常相似或重复的短语等。
目前，您无法可靠地在源字符串中包含
[
]
N
2*N + 5
word[2]
word[3][2][1]
one[1] two[2] three[3]
one[1] []two three[3]
any number of word characters before [index]
any number of word characters after []
var matchstring = \'Brown cats cannot be white cats\';
var wrapstring = \'Hi bill,brown cats cannot be white cats and cows are not blue pigs,blue melons are large but not batteries charged barges with white cats carry coal,and the word \"cannot\" should match \';
// Pre-process matchstring to make it a flat list of words
// separated by single spaces.
matchstring = matchstring.replace(/\\W+/g,\' \');
var wrapStart = \'<span class=\"wrapped\">\';
var wrapEnd = \'</span>\';
var matcharray = matchstring.split(\' \');
var i,reg;
// Mark all matched words with indices
// one -> one[1]
for (i = 0; i < matcharray.length; i++) {
reg = new RegExp(\'\\\\b\' + matcharray[i] + \'\\\\b\',\'ig\');
wrapstring = wrapstring.replace(reg,\'$&[\' + i + \']\');
}
// Mark all inner words
// one[1] two[2] three[3] -> one[1] []two[2] three[3]
for (i = 1; i < matcharray.length; i++) {
reg = new RegExp(\'\\\\b(\\\\w+)([\\\\]\\\\d\\\\[]*\\\\[\' + (i - 1) + \'\\\\][\\\\]\\\\d\\\\[]*)(\\\\W+)(\\\\w+)([\\\\]\\\\d\\\\[]*\\\\[\' + i + \'\\\\][\\\\]\\\\d\\\\[]*)(?=\\\\W+\\\\w+[\\\\[\\\\d\\\\]]*\\\\[\' + (i + 1) + \'\\\\])\',\'$1$2$3[]$4$5\');
}
// Remove indices from inner words
// one[1] []two[2] three[3] -> one[1] []two three[3]
wrapstring = wrapstring.replace(/\\[\\](\\w+)[\\[\\d\\]]*/g,\'[]$1\');
// Start tags
// one[1] []two three[3] -> {one []two three[3]
wrapstring = wrapstring.replace(/(\\w+)\\[[\\[\\d\\]]+\\](\\W+)\\[\\]/g,wrapStart + \'$1$2[]\');
// End tags
// {one []two three[3] -> {one []two three}
wrapstring = wrapstring.replace(/\\[\\](\\w+\\W+\\w+)\\[[\\[\\d\\]]+\\]/g,\'$1\' + wrapEnd);
// Wrap double words
// one[1] two[2] -> {one two}
wrapstring = wrapstring.replace(/(\\w+)\\[[\\[\\d\\]]+\\](\\W+\\w+)\\[[\\[\\d\\]]*\\]/g,wrapStart + \'$1$2\' + wrapEnd);
// Orphan words
// unmatched matched[1] unmatched -> unmatched {matched} unmatched
wrapstring = wrapstring.replace(/(\\w+)\\[[\\[\\d\\]]+\\]/g,wrapStart + \'$1\' + wrapEnd);
// Remove left-over tags
// []word -> word
wrapstring = wrapstring.replace(/\\[\\]/g,\'\');
alert(wrapstring);
reg = new RegExp(\'\\\\b\' + matcharray[i] + \'\\\\b\',\'ig\');
\\bword\\b
word
reg = new RegExp(\'\\\\b\' + matcharray[i] + \'\\\\w*\\\\b\',\'ig\');
word
\\w*
\\\\
\\
\\bword\\w*\\b
\\b\\w*word\\b
\\b\\w*word\\w*\\b
\\b(\\w*word|word\\w*)\\b
danger
endanger
cat
cats
cannot
foot
\\b(foot|feet)\\b
\\bword(s|es|)\\b
word
words
wordes
fly
\\bfl(y|ies)\\b
index
\\bind(ex|exes|ices)\\b
?
?red
\\w
?
\\b\\wred\\b
dred
\\w+
\\w*
\\bf\\w+d\\b
fed
feed
\\w*
/* Note: str1 and str2 are the two overlapping portions of the strings */
var largestMatch = 0;
var currMatch = 0;
for (var i = 0; i < str1.length; i++) {
if (str1[i] == str2[i]) currMatch++;
else currMatch = 0;
largestMatch = Math.max(largestMatch,currMatch);
}
// largestMatch is the size of the largest section of matched characters
function findStringLimit(searchChar,searchCharIndex,searchedString) {
return searchedString.substring(0,searchedString.lastIndexOf(searchChar,searchCharIndex));
};
function replaceWords(wordsy,text) {
var re = \'(\' + wordsy + \')(?![^<]*(?:<\\/script|>))\',regExp = new RegExp(re,\'ig\'),sTag = \"<span class=\'wrappedWord\'>\",eTag = \"</span>\";
return text.replace(regExp,sTag + \'$&\' + eTag);
};
var longstring = $(\'#mystring\');
var htmlString =longstring .html(); // instance html
myValue = \"Brown cats cannot be white cats\";
myValue = myValue.replace(/^\\s+|\\s+$/g,\"\");//trim whitespace at each end
var words = myValue.split(\" \");
var allPhrases = [];
allPhrases.push(myValue);
var i = words.length;
while (i--) {
allPhrases.push(findStringLimit(\" \",allPhrases[(words.length - i) - 1].length,allPhrases[(words.length - i) - 1]));
};
var i = allPhrases.length;
while (i--) {
if (allPhrases[i] != \"\") words = words.concat(allPhrases[i]);
};
var i = words.length;
while (i--) {
htmlString = replaceWords(words[i],htmlString);
};
longstring.html(htmlString);