这是您想要的简单解决方案，它是一个动态解决方案，因为它会处理找到的任何标签并仅比较文本内容。 findDiff() 会发现差异并以输出和不同单词的数组作为参数调用回调函数。
JSFiddle：https://jsfiddle.net/9svuc7om/18/
/**
* Parse and construct an Array of PDF text tokens
* @params {string} text The PDF text to be parsed
* @return {object} The parsed Array of tokens
*/
function parsePDFText(text) {
var token = text.split(' ');
for (var i=0,l=token.length; i<l; i++) {
// remove token of first space and consecutive space
if (token[i] == '') {
token.splice(i, 1);
}
}
return token;
}
/**
* Return the minimum indexOf among all the arguments
* @params {...number} index The indexOf
* @return {number} The minimum indexOf, -1 if all arguments are -1
*/
function findMinIndex() {
var min;
for (var i = 0, l = arguments.length; i < l; i++) {
// indexOf() returns -1 if not found
if (arguments[i] === -1) {
continue;
}
if (typeof min === 'undefined' || arguments[i] < min) {
min = arguments[i];
}
}
return min || -1;
}
/**
* Parse and construct an Array of HTML tokens
* @params {string} text The HTML text to be parsed
* @return {object} The parsed Array of tokens
*/
function parseHTMLText(text) {
var currentIndex = 0,
tl = text.length,
tokens = [],
token, firstChar, endPos;
while (currentIndex < tl) {
// determine the next token type
firstChar = text.charAt(currentIndex);
if (firstChar == '<') {
// a tag
// find the position of closing tag, assume all tags are well formed
endPos = text.indexOf('>', currentIndex + 1) + 1;
token = {
type: 'tag',
content: text.slice(currentIndex, endPos),
valid: true
}
currentIndex = endPos;
} else if (firstChar == ' ') {
// a space
token = {
type: 'space',
content: ' ',
valid: true
}
currentIndex++;
} else {
// a character, possibliy part of a word
// find the end of the word
// assume a word is delimitered either by tags or space
endPos = findMinIndex(text.indexOf('<', currentIndex), text.indexOf(' ', currentIndex));
// endPos is `-1` if there are not delimiter anymore, end of string reached
if (endPos === -1) {
endPos = tl;
}
token = {
type: 'text',
content: text.slice(currentIndex, endPos),
valid: true
}
currentIndex = endPos;
}
tokens.push(token);
}
return tokens;
}
/**
* Find the difference between pdf text and html text and pass the output and differenc to a callback function
* @params {string} pdfText The pdf text
* @params {string} htmlText The html text
* @params {function} callback The callback function
*/
function findDiff(pdfText, htmlText, callback) {
var output = '', // the final output
diff = [], // the array of different words
pdfTokens = parsePDFText(pdfText),
htmlTokens = parseHTMLText(htmlText),
j=0, hl=htmlTokens.length;
// the pdf text is the reference point, i.e. all the words in pdf text should always be present in html text as well
for (var i=0,pl=pdfTokens.length; i<pl; i++) {
// find the first occurrence of the pdf text
for(; j<hl; j++) {
if (htmlTokens[j].type != 'text') {
// exclude comparison to non-text
continue;
}
// check if the two text matches
if (htmlTokens[j].content == pdfTokens[i]) {
// a match is found
j++;
break;
} else {
// push the different html token into `diff` array
diff.push(htmlTokens[j].content);
// set the `valid` field of token to false
htmlTokens[j].valid = false;
}
}
}
// invalidate the rest of the html text
for(; j<hl; j++) {
if (htmlTokens[j].type == 'text') {
htmlTokens[j].valid = false;
}
}
// concat the final string to output
for (j=0; j<hl; j++) {
if (htmlTokens[j].valid) {
output += htmlTokens[j].content;
}
}
callback(output, diff);
}
你可以通过使用来调用函数
findDiff(text1, text2, function(output, diff) {
console.log(output);
console.log(diff);
});
但是，此解决方案存在一些限制
- 假定 pdf 中的所有内容都存在于 HTML 文本中
- 它只处理
<> 和空格，如果有其他可能的分隔符，例如选项卡，需要额外的代码
- 它假设所有标签都是格式正确的，并且文本内容之间不会有结束标签（如果你需要你应该使用
&gt;&lt;）
- 该功能是一种简化的解决方案，尚未经过全面测试。您不能期望它有任何保证，并且需要进行一些调整。我建议只提供
body 内的内容，甚至是更窄的范围，而不是整个 HTML 文件（如果你的情况可能的话），因为 HTML 文件的内容会有太多变化。