xaay/dr_py
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

修复drpy支持磁力链接，style属性处理等问题

Browse Source
This commit is contained in:
晚风拂柳颜 2023-07-28 11:43:24 +08:00
parent dbc8e560a3
commit aff38eb7fe
8 changed files with 33 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
libs/drpy.js
View File

@ -55,7 +55,7 @@ function pre(){
} }
let rule = {}; let rule = {};
const VERSION = 'drpy1 3.9.47beta1 20230711'; const VERSION = 'drpy1 3.9.47beta15 20230728';
/** 已知问题记录 /** 已知问题记录
* 1.影魔的jinjia2引擎不支持 {{fl}}对象直接渲染 (有能力解决的话尽量解决下，支持对象直接渲染字符串转义,如果加了|safe就不转义)[影魔牛逼，最新的文件发现这问题已经解决了] * 1.影魔的jinjia2引擎不支持 {{fl}}对象直接渲染 (有能力解决的话尽量解决下，支持对象直接渲染字符串转义,如果加了|safe就不转义)[影魔牛逼，最新的文件发现这问题已经解决了]
* Array.prototype.append = Array.prototype.push; 这种js执行后有毛病,for in 循环列表会把属性给打印出来 (这个大毛病需要重点排除一下) * Array.prototype.append = Array.prototype.push; 这种js执行后有毛病,for in 循环列表会把属性给打印出来 (这个大毛病需要重点排除一下)
@ -111,6 +111,8 @@ var _pdfa;
var _pd; var _pd;
// const DOM_CHECK_ATTR = ['url', 'src', 'href', 'data-original', 'data-src']; // const DOM_CHECK_ATTR = ['url', 'src', 'href', 'data-original', 'data-src'];
const DOM_CHECK_ATTR = /(url|src|href|-original|-src|-play|-url|style)$/; const DOM_CHECK_ATTR = /(url|src|href|-original|-src|-play|-url|style)$/;
// 过滤特殊链接,不走urlJoin
const SPECIAL_URL = /^(ftp|magnet|thunder|ws):/;
const SELECT_REGEX = /:eq|:lt|:gt|#/g; const SELECT_REGEX = /:eq|:lt|:gt|#/g;
const SELECT_REGEX_A = /:eq|:lt|:gt/g; const SELECT_REGEX_A = /:eq|:lt|:gt/g;
@ -537,7 +539,7 @@ const defaultParser = {
if(typeof(uri)==='undefined'||!uri){ if(typeof(uri)==='undefined'||!uri){
uri = ''; uri = '';
} }
if(DOM_CHECK_ATTR.test(parse)){ if(DOM_CHECK_ATTR.test(parse) && !SPECIAL_URL.test(ret)){
if(/http/.test(ret)){ if(/http/.test(ret)){
ret = ret.substr(ret.indexOf('http')); ret = ret.substr(ret.indexOf('http'));
}else{ }else{
@ -569,6 +571,8 @@ function pdfh2(html,parse){
if(/style/.test(option.toLowerCase())&&/url\(/.test(result)){ if(/style/.test(option.toLowerCase())&&/url\(/.test(result)){
try { try {
result = result.match(/url\((.*?)\)/)[1]; result = result.match(/url\((.*?)\)/)[1];
// 2023/07/28新增 style取内部链接自动去除首尾单双引号
result = result.replace(/^['|"](.*)['|"]$/, "$1");
}catch (e) {} }catch (e) {}
} }
return result return result
@ -604,7 +608,7 @@ function pd2(html,parse,uri){
if(typeof(uri)==='undefined'||!uri){ if(typeof(uri)==='undefined'||!uri){
uri = ''; uri = '';
} }
if(DOM_CHECK_ATTR.test(parse)){ if(DOM_CHECK_ATTR.test(parse) && !SPECIAL_URL.test(ret)){
if(/http/.test(ret)){ if(/http/.test(ret)){
ret = ret.substr(ret.indexOf('http')); ret = ret.substr(ret.indexOf('http'));
}else{ }else{
@ -727,10 +731,12 @@ const parseTags = {
if(/style/.test(option.toLowerCase())&&/url\(/.test(result)){ if(/style/.test(option.toLowerCase())&&/url\(/.test(result)){
try { try {
result = result.match(/url\((.*?)\)/)[1]; result = result.match(/url\((.*?)\)/)[1];
// 2023/07/28新增 style取内部链接自动去除首尾单双引号
result = result.replace(/^['|"](.*)['|"]$/, "$1");
}catch (e) {} }catch (e) {}
} }
} }
if (result && base_url && DOM_CHECK_ATTR.test(option)) { if (result && base_url && DOM_CHECK_ATTR.test(option) && !SPECIAL_URL.test(result)) {
if (/http/.test(result)) { if (/http/.test(result)) {
result = result.substr(result.indexOf('http')); result = result.substr(result.indexOf('http'));
} else { } else {

2
libs/drpy.min.js vendored
View File

File diff suppressed because one or more lines are too long

10
libs/drpy2.js
View File

@ -41,7 +41,7 @@ function pre(){
let rule = {}; let rule = {};
let vercode = typeof(pdfl) ==='function'?'drpy2.1':'drpy2'; let vercode = typeof(pdfl) ==='function'?'drpy2.1':'drpy2';
const VERSION = vercode+' 3.9.47beta1 20230711'; const VERSION = vercode+' 3.9.47beta15 20230728';
/** 已知问题记录 /** 已知问题记录
* 1.影魔的jinjia2引擎不支持 {{fl}}对象直接渲染 (有能力解决的话尽量解决下，支持对象直接渲染字符串转义,如果加了|safe就不转义)[影魔牛逼，最新的文件发现这问题已经解决了] * 1.影魔的jinjia2引擎不支持 {{fl}}对象直接渲染 (有能力解决的话尽量解决下，支持对象直接渲染字符串转义,如果加了|safe就不转义)[影魔牛逼，最新的文件发现这问题已经解决了]
* Array.prototype.append = Array.prototype.push; 这种js执行后有毛病,for in 循环列表会把属性给打印出来 (这个大毛病需要重点排除一下) * Array.prototype.append = Array.prototype.push; 这种js执行后有毛病,for in 循环列表会把属性给打印出来 (这个大毛病需要重点排除一下)
@ -97,7 +97,9 @@ var _pdfh;
var _pdfa; var _pdfa;
var _pd; var _pd;
// const DOM_CHECK_ATTR = ['url', 'src', 'href', 'data-original', 'data-src']; // const DOM_CHECK_ATTR = ['url', 'src', 'href', 'data-original', 'data-src'];
const DOM_CHECK_ATTR = /(url|src|href|-original|-src|-play|-url)$/; const DOM_CHECK_ATTR = /(url|src|href|-original|-src|-play|-url|style)$/;
// 过滤特殊链接,不走urlJoin
const SPECIAL_URL = /^(ftp|magnet|thunder|ws):/;
const NOADD_INDEX = /:eq|:lt|:gt|:first|:last|^body$|^#/; // 不自动加eq下标索引 const NOADD_INDEX = /:eq|:lt|:gt|:first|:last|^body$|^#/; // 不自动加eq下标索引
const URLJOIN_ATTR = /(url|src|href|-original|-src|-play|-url|style)$/; // 需要自动urljoin的属性 const URLJOIN_ATTR = /(url|src|href|-original|-src|-play|-url|style)$/; // 需要自动urljoin的属性
const SELECT_REGEX = /:eq|:lt|:gt|#/g; const SELECT_REGEX = /:eq|:lt|:gt|#/g;
@ -544,6 +546,8 @@ function pdfh2(html,parse){
if(/style/.test(option.toLowerCase())&&/url\(/.test(result)){ if(/style/.test(option.toLowerCase())&&/url\(/.test(result)){
try { try {
result = result.match(/url\((.*?)\)/)[1]; result = result.match(/url\((.*?)\)/)[1];
// 2023/07/28新增 style取内部链接自动去除首尾单双引号
result = result.replace(/^['|"](.*)['|"]$/, "$1");
}catch (e) {} }catch (e) {}
} }
return result return result
@ -579,7 +583,7 @@ function pd2(html,parse,uri){
if(typeof(uri)==='undefined'||!uri){ if(typeof(uri)==='undefined'||!uri){
uri = ''; uri = '';
} }
if(DOM_CHECK_ATTR.test(parse)){ if(DOM_CHECK_ATTR.test(parse) && !SPECIAL_URL.test(ret)){
if(/http/.test(ret)){ if(/http/.test(ret)){
ret = ret.substr(ret.indexOf('http')); ret = ret.substr(ret.indexOf('http'));
}else{ }else{

2
libs/drpy2.min.js vendored
View File

File diff suppressed because one or more lines are too long

5
utils/HtmlParser.cs
View File

@ -176,6 +176,7 @@ namespace Peach.DataAccess
private static readonly Regex p = new ("url\\((.*?)\\)", RegexOptions.Multiline | RegexOptions.Singleline); private static readonly Regex p = new ("url\\((.*?)\\)", RegexOptions.Multiline | RegexOptions.Singleline);
private static readonly Regex NOAdd_INDEX = new (":eq|:lt|:gt|:first|:last|^body$|^#"); private static readonly Regex NOAdd_INDEX = new (":eq|:lt|:gt|:first|:last|^body$|^#");
private static readonly Regex URLJOIN_ATTR = new ("(url|src|href|-original|-src|-play|-url|style)$", RegexOptions.Multiline | RegexOptions.IgnoreCase); private static readonly Regex URLJOIN_ATTR = new ("(url|src|href|-original|-src|-play|-url|style)$", RegexOptions.Multiline | RegexOptions.IgnoreCase);
private static readonly Regex SPECIAL_URL = new ("^(ftp|magnet|thunder|ws):", RegexOptions.Multiline | RegexOptions.IgnoreCase);
private static String pdfh_html = ""; private static String pdfh_html = "";
private static String pdfa_html = ""; private static String pdfa_html = "";
private static Document? pdfh_doc = null; private static Document? pdfh_doc = null;
@ -317,13 +318,15 @@ namespace Peach.DataAccess
Match m = p.Match(result); Match m = p.Match(result);
if (m.Success) if (m.Success)
result = m.Groups[1]?.Value; result = m.Groups[1]?.Value;
result = Regex.Replace(result, "^['|\"](.*)['|\"]$", "$1");
} }
if (!string.IsNullOrWhiteSpace(result) && !string.IsNullOrWhiteSpace(Add_url))// (JSUtils.isNotEmpty(result) && JSUtils.isNotEmpty(Add_url)) if (!string.IsNullOrWhiteSpace(result) && !string.IsNullOrWhiteSpace(Add_url))// (JSUtils.isNotEmpty(result) && JSUtils.isNotEmpty(Add_url))
{ {
// 需要自动urljoin的属性 // 需要自动urljoin的属性
Match m = URLJOIN_ATTR.Match(option); Match m = URLJOIN_ATTR.Match(option);
Match n = SPECIAL_URL.Match(result);
//if (isUrl(option)) { //if (isUrl(option)) {
if (m.Success) if (m.Success && !n.Success)
{ {
if (result.Contains("http")) if (result.Contains("http"))
result = result[result.IndexOf("http")..]; result = result[result.IndexOf("http")..];

6
utils/HtmlParser.java
View File

@ -23,6 +23,7 @@ public class HtmlParser {
private static final Pattern p = Pattern.compile("url\\((.*?)\\)", Pattern.MULTILINE | Pattern.DOTALL); private static final Pattern p = Pattern.compile("url\\((.*?)\\)", Pattern.MULTILINE | Pattern.DOTALL);
private static final Pattern NOADD_INDEX = Pattern.compile(":eq|:lt|:gt|:first|:last|^body$|^#"); // 不自动加eq下标索引 private static final Pattern NOADD_INDEX = Pattern.compile(":eq|:lt|:gt|:first|:last|^body$|^#"); // 不自动加eq下标索引
private static final Pattern URLJOIN_ATTR = Pattern.compile("(url|src|href|-original|-src|-play|-url|style)$", Pattern.MULTILINE | Pattern.CASE_INSENSITIVE); // 需要自动urljoin的属性 private static final Pattern URLJOIN_ATTR = Pattern.compile("(url|src|href|-original|-src|-play|-url|style)$", Pattern.MULTILINE | Pattern.CASE_INSENSITIVE); // 需要自动urljoin的属性
private static final Pattern SPECIAL_URL = Pattern.compile("^(ftp|magnet|thunder|ws):", Pattern.MULTILINE | Pattern.CASE_INSENSITIVE); // 过滤特殊链接,不走urlJoin
private static Document pdfh_doc = null; private static Document pdfh_doc = null;
private static Document pdfa_doc = null; private static Document pdfa_doc = null;
@ -199,12 +200,15 @@ public class HtmlParser {
if (m.find()) { if (m.find()) {
result = m.group(1); result = m.group(1);
} }
// 2023/07/28新增 style取内部链接自动去除首尾单双引号
result = result.replaceAll("^['|\"](.*)['|\"]$", "$1");
} }
if (JSUtils.isNotEmpty(result) && JSUtils.isNotEmpty(add_url)) { if (JSUtils.isNotEmpty(result) && JSUtils.isNotEmpty(add_url)) {
// 需要自动urljoin的属性 // 需要自动urljoin的属性
Matcher m = URLJOIN_ATTR.matcher(option); Matcher m = URLJOIN_ATTR.matcher(option);
Matcher n = SPECIAL_URL.matcher(result);
//if (isUrl(option)) { //if (isUrl(option)) {
if (m.find()) { if (m.find() && !n.find() {
if (result.contains("http")) { if (result.contains("http")) {
result = result.substring(result.indexOf("http")); result = result.substring(result.indexOf("http"));
} else { } else {

5
utils/htmlParser.py
View File

@ -14,6 +14,7 @@ from jsonpath import jsonpath
PARSE_CACHE = True # 解析缓存 PARSE_CACHE = True # 解析缓存
NOADD_INDEX = ':eq|:lt|:gt|:first|:last|^body$|^#' # 不自动加eq下标索引 NOADD_INDEX = ':eq|:lt|:gt|:first|:last|^body$|^#' # 不自动加eq下标索引
URLJOIN_ATTR = '(url|src|href|-original|-src|-play|-url|style)$' # 需要自动urljoin的属性 URLJOIN_ATTR = '(url|src|href|-original|-src|-play|-url|style)$' # 需要自动urljoin的属性
SPECIAL_URL = '^(ftp|magnet|thunder|ws):' # 过滤特殊链接,不走urlJoin
class jsoup: class jsoup:
@ -193,11 +194,13 @@ class jsoup:
if self.contains(option.lower(), 'style') and self.contains(ret, 'url('): if self.contains(option.lower(), 'style') and self.contains(ret, 'url('):
try: try:
ret = re.search('url\((.*?)\)', ret, re.M | re.S).groups()[0] ret = re.search('url\((.*?)\)', ret, re.M | re.S).groups()[0]
# 2023/07/28新增 style取内部链接自动去除首尾单双引号
ret = re.sub(r"^['\"]|['\"]$", '', ret)
except: except:
pass pass
if ret and base_url: if ret and base_url:
# need_add = re.search(URLJOIN_ATTR, option, re.M | re.I) # need_add = re.search(URLJOIN_ATTR, option, re.M | re.I)
need_add = self.test(URLJOIN_ATTR, option) need_add = self.test(URLJOIN_ATTR, option) and not self.test(SPECIAL_URL, ret)
if need_add: if need_add:
if 'http' in ret: if 'http' in ret:
ret = ret[ret.find('http'):] ret = ret[ret.find('http'):]