升级
This commit is contained in:
@@ -18,7 +18,7 @@ class ArticleParserService
|
||||
public function __construct($filePath = '')
|
||||
{
|
||||
if (!file_exists($filePath)) {
|
||||
throw new Exception("文档不存在:{$filePath}");
|
||||
return json_encode(['status' => 5, 'msg' => '"文档不存在:{$filePath}"']);
|
||||
}
|
||||
try {
|
||||
// 关键配置:关闭“仅读数据”,保留完整节结构
|
||||
@@ -32,9 +32,8 @@ class ArticleParserService
|
||||
// $this->log("✅ 文档直接加载成功,节数量:{$sectionCount}");
|
||||
$this->phpWord = $reader->load($filePath);
|
||||
$this->sections = $this->phpWord->getSections();
|
||||
|
||||
} catch (\Exception $e) {
|
||||
return json(['status' => 'error', 'msg' => $e->getMessage()]);
|
||||
return json_encode(['status' => 5, 'msg' => $e->getMessage()]);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -260,6 +259,168 @@ class ArticleParserService
|
||||
// var_dump($aAuthorData);exit;
|
||||
// return ['author' => $aAuthorData,'report' => array_unique($aReport)];
|
||||
// }
|
||||
|
||||
// 提取作者
|
||||
private function parseAuthorsWithoutRegex($str = '') {
|
||||
if (empty($str)) {
|
||||
return [];
|
||||
}
|
||||
// 清理乱码和特殊字符(扩展全角数字处理)
|
||||
$str = mb_convert_encoding($str, 'UTF-8', 'auto');
|
||||
$str = str_replace(["\xC2\xA0", 'ï¼', '�', ',', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0'],
|
||||
[' ', ' ', ' ', ' ', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0'], $str);
|
||||
$str = trim(str_replace([' and ', ' AND ', ' And '], ', ', $str));
|
||||
|
||||
// 合并上标中数字与逗号间的空格(如"2, 3"→"2,3")
|
||||
$len = mb_strlen($str);
|
||||
$processed = '';
|
||||
for ($i = 0; $i < $len; $i++) {
|
||||
$char = mb_substr($str, $i, 1);
|
||||
if ($char === ',' && $i - 1 >= 0 && $i + 2 < $len) {
|
||||
$prevChar = mb_substr($str, $i - 1, 1);
|
||||
$next1 = mb_substr($str, $i + 1, 1);
|
||||
$next2 = mb_substr($str, $i + 2, 1);
|
||||
// 兼容全角数字转半角后的判断
|
||||
if ((ctype_digit($prevChar) || is_numeric($prevChar)) && $next1 === ' ' && (ctype_digit($next2) || is_numeric($next2))) {
|
||||
$processed .= $char;
|
||||
$i += 1;
|
||||
continue;
|
||||
}
|
||||
}
|
||||
$processed .= $char;
|
||||
}
|
||||
$str = $processed;
|
||||
|
||||
// 合并数字与符号间的空格(如"1 *"→"1*")
|
||||
$len = mb_strlen($str);
|
||||
$processed = '';
|
||||
for ($i = 0; $i < $len; $i++) {
|
||||
$char = mb_substr($str, $i, 1);
|
||||
if ((ctype_digit($char) || is_numeric($char)) && $i + 2 < $len) { // 支持数字判断
|
||||
$next1 = mb_substr($str, $i + 1, 1);
|
||||
$next2 = mb_substr($str, $i + 2, 1);
|
||||
if ($next1 === ' ' && in_array($next2, ['#', '*', '†', '‡', '§'])) { // 扩展符号支持
|
||||
$processed .= $char;
|
||||
$i += 2;
|
||||
$processed .= $next2;
|
||||
continue;
|
||||
}
|
||||
}
|
||||
$processed .= $char;
|
||||
}
|
||||
$str = $processed;
|
||||
|
||||
// 合并连续空格
|
||||
$len = mb_strlen($str);
|
||||
$processed = '';
|
||||
$prevSpace = false;
|
||||
for ($i = 0; $i < $len; $i++) {
|
||||
$char = mb_substr($str, $i, 1);
|
||||
if ($char === ' ') {
|
||||
if (!$prevSpace) {
|
||||
$processed .= $char;
|
||||
$prevSpace = true;
|
||||
}
|
||||
} else {
|
||||
$processed .= $char;
|
||||
$prevSpace = false;
|
||||
}
|
||||
}
|
||||
$str = trim($processed);
|
||||
|
||||
// 作者处理
|
||||
$authors = [];
|
||||
$currentName = '';
|
||||
$currentSuperscript = '';
|
||||
$inName = true;
|
||||
$len = mb_strlen($str);
|
||||
for ($i = 0; $i < $len; $i++) {
|
||||
$char = mb_substr($str, $i, 1);
|
||||
|
||||
// 处理作者分隔符:逗号+空格
|
||||
if ($char === ',' && $i + 1 < $len) {
|
||||
$nextChar = mb_substr($str, $i + 1, 1);
|
||||
if ($nextChar === ' ') {
|
||||
if (!empty($currentName)) {
|
||||
$currentSuperscript = rtrim($currentSuperscript, ',');
|
||||
$authors[] = [
|
||||
'name' => trim($currentName),
|
||||
'superscript' => trim($currentSuperscript)
|
||||
];
|
||||
}
|
||||
$currentName = '';
|
||||
$currentSuperscript = '';
|
||||
$inName = true;
|
||||
$i++;
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
||||
// 支持姓名中的点、连字符、特殊字母(如带重音的字母)
|
||||
if (ctype_alpha($char) || in_array($char, [' ', '.', '-', 'à', 'á', 'â', 'ã', 'ä', 'ç', 'è', 'é', 'ê', 'ë'])) {
|
||||
if ($inName) {
|
||||
$currentName .= $char;
|
||||
} else {
|
||||
$currentSuperscript = rtrim($currentSuperscript, ',');
|
||||
$authors[] = [
|
||||
'name' => trim($currentName),
|
||||
'superscript' => trim($currentSuperscript)
|
||||
];
|
||||
$currentName = $char;
|
||||
$currentSuperscript = '';
|
||||
$inName = true;
|
||||
}
|
||||
}
|
||||
// 解析上标(数字、逗号、#、*、†等)
|
||||
elseif ((ctype_digit($char) || is_numeric($char)) || in_array($char, ['#', '*', '†', ',', '‡', '§'])) {
|
||||
$inName = false;
|
||||
$currentSuperscript .= $char;
|
||||
}
|
||||
// 忽略其他字符
|
||||
else {
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
||||
// 处理最后一个作者
|
||||
if (!empty($currentName)) {
|
||||
$currentSuperscript = rtrim($currentSuperscript, ',');
|
||||
$authors[] = [
|
||||
'name' => trim($currentName),
|
||||
'superscript' => trim($currentSuperscript)
|
||||
];
|
||||
}
|
||||
|
||||
// 提取机构编号为数组、判断通讯作者和第一作者
|
||||
foreach ($authors as $index => &$author) {
|
||||
// 提取机构编号(兼容多字节数字)
|
||||
$institutionIds = [];
|
||||
$superscript = $author['superscript'];
|
||||
$numStr = '';
|
||||
for ($i = 0; $i < mb_strlen($superscript); $i++) {
|
||||
$c = mb_substr($superscript, $i, 1);
|
||||
if (ctype_digit($c) || is_numeric($c)) { // 支持数字判断
|
||||
$numStr .= $c;
|
||||
} else {
|
||||
if (!empty($numStr)) {
|
||||
$institutionIds[] = (int)$numStr;
|
||||
$numStr = '';
|
||||
}
|
||||
}
|
||||
}
|
||||
if (!empty($numStr)) {
|
||||
$institutionIds[] = (int)$numStr;
|
||||
}
|
||||
$institutionIds = array_values(array_unique($institutionIds));
|
||||
$author['company_id'] = $institutionIds;
|
||||
|
||||
// 判断第一作者(#标记)和通讯作者(*、†标记)
|
||||
$author['is_super'] = strpos($superscript, '#') !== false ? 1 : 0;
|
||||
$author['is_report'] = (strpos($superscript, '*') !== false || strpos($superscript, '†') !== false) ? 1 : 0;
|
||||
}
|
||||
unset($author); // 释放引用
|
||||
return $authors;
|
||||
}
|
||||
private function getAuthors($aParam = []) {
|
||||
$title = empty($aParam['title']) ? $this->getTitle() : $aParam['title'];
|
||||
$sAuthorContent = $this->getNextParagraphAfterText($title);
|
||||
@@ -291,95 +452,192 @@ class ArticleParserService
|
||||
$sAuthorContent = preg_replace('/\s+and\s+/i', ', ', $sAuthorContent); // and转逗号
|
||||
$sAuthorContent = preg_replace('/\s+/', ' ', $sAuthorContent); // 合并多余空格
|
||||
$sAuthorContent = trim($sAuthorContent);
|
||||
|
||||
// 处理作者
|
||||
$content = mb_convert_encoding($sAuthorContent, 'UTF-8', 'auto'); // 确保编码正确
|
||||
$content = str_replace("\xC2\xA0", ' ', $content); // 替换非-breaking空格为普通空格
|
||||
$content = preg_replace('/(\d+)\s*([*#†])/', '$1$2', $content); // 合并"1 *"为"1*"、"1 #"为"1#"
|
||||
$content = preg_replace('/,†/', ',†', $content); // 保留"1,†"格式(防止被拆分)
|
||||
//标记上标内的逗号+空格(多编号)
|
||||
$tempStr = preg_replace('/(\d+)\s*,\s*(\d+)/', '$1<SEP>$2', $content);
|
||||
// 原有步骤2:正则匹配(扩展上标符号支持,保持原有逻辑)
|
||||
$pattern = '/
|
||||
([A-Za-z\s\.\-]+?) # 姓名(支持缩写、空格)
|
||||
\s* # 姓名与上标间空格
|
||||
( # 上标组(扩展符号支持)
|
||||
\d+ # 起始数字
|
||||
(?:[†#*,]|<SEP>\d+)* # 允许:†#*符号、逗号、<SEP>+数字(兼容1,†、1,*等)
|
||||
)
|
||||
\s*,? # 作者间逗号(可选)
|
||||
(?=\s|$) # 确保后面是空格或结尾
|
||||
/ux';
|
||||
|
||||
preg_match_all($pattern, $tempStr, $matches);
|
||||
$authorList = [];
|
||||
if(!empty($matches[1])){
|
||||
foreach ($matches[1] as $i => $name) {
|
||||
$name = trim($name);
|
||||
$superscript = trim($matches[2][$i]);
|
||||
$superscript = str_replace('<SEP>', ',', $superscript); // 恢复多编号逗号
|
||||
$superscript = preg_replace('/,$/', '', $superscript); // 清理末尾逗号
|
||||
// 修复符号与数字间的空格(如原始"1 *"被误处理为"1*"的情况,保持原样)
|
||||
$superscript = preg_replace('/(\d)([*#†])/', '$1$2', $superscript);
|
||||
if (!empty($name)) {
|
||||
$authorList[] = [
|
||||
'name' => $name,
|
||||
'superscript' => $superscript
|
||||
];
|
||||
}
|
||||
}
|
||||
}else {
|
||||
// 按“两个或多个连续空格”拆分(姓名之间的分隔)
|
||||
$authorList = array_filter(
|
||||
array_map('trim',
|
||||
preg_split('/(,\p{Z}*|\p{Z}{2,})/u', $sAuthorContent)
|
||||
)
|
||||
);
|
||||
$aAuthor = $this->parseAuthorsWithoutRegex($sAuthorContent);
|
||||
if(empty($aAuthor)){
|
||||
return ['author' => [],'report' => []];
|
||||
}
|
||||
|
||||
$aReport = $aAuthorData = [];
|
||||
|
||||
// //处理作者
|
||||
$aAuthorData = [];
|
||||
$aReport = [];
|
||||
$namePattern = '/
|
||||
(?:[A-Za-z\s·\-\']+| # 英文姓名(支持空格、连字符)
|
||||
[\x{4e00}-\x{9fa5}]+| # 中文姓名
|
||||
[\x{1800}-\x{18AF}]+| # 蒙古文姓名
|
||||
[A-Z]\.) # 单字母缩写(如 J.)
|
||||
/ux';
|
||||
|
||||
foreach ($authorList as $authorStr){
|
||||
if (empty($authorStr)) continue;
|
||||
|
||||
//获取下标
|
||||
$superscript = empty($authorStr['superscript']) ? $authorStr : $authorStr['superscript'];
|
||||
$nameStr = empty($authorStr['name']) ? $authorStr : $authorStr['name'];
|
||||
|
||||
$companyId = [];
|
||||
$isSuper = 0;
|
||||
$isReport = 0;
|
||||
if (!empty($superscript)) {
|
||||
// 提取机构编号(忽略上标中的逗号,如1,† → 提取1)
|
||||
preg_match_all('/\d+/', $superscript, $numMatch);
|
||||
// 识别特殊符号(#为超级作者,*†为通讯作者)
|
||||
$isSuper = strpos($superscript, '#') !== false ? 1 : 0;
|
||||
$isReport = (strpos($superscript, '*') !== false || strpos($superscript, '†') !== false) ? 1 : 0;
|
||||
foreach ($aAuthor as $key => $value) {
|
||||
if(empty($value['name']) && empty($value['superscript'])){
|
||||
continue;
|
||||
}
|
||||
if (preg_match("/^([A-Za-z\s'\.-]+)/u", $nameStr, $match)) {
|
||||
$nameStr = trim($match[1]);
|
||||
}
|
||||
$aAuthorData[] = [
|
||||
'name' => $nameStr,
|
||||
'company_id' => empty($numMatch[0]) ? [] : $numMatch[0],
|
||||
'is_super' => $isSuper,
|
||||
'is_report' => $isReport
|
||||
];
|
||||
if ($isReport) {
|
||||
$aReport[] = $nameStr;
|
||||
if(!empty($value['name']) && !empty($value['is_report']) && $value['is_report'] == 1){
|
||||
$aReport[] = $value['name'];
|
||||
}
|
||||
$aAuthorData[] = $value;
|
||||
}
|
||||
return ['author' => $aAuthorData,'report' => array_unique($aReport)];
|
||||
}
|
||||
// private function getAuthors($aParam = []) {
|
||||
// $title = empty($aParam['title']) ? $this->getTitle() : $aParam['title'];
|
||||
// $sAuthorContent = $this->getNextParagraphAfterText($title);
|
||||
// if (empty($sAuthorContent)) {
|
||||
// return ['author' => [], 'report' => []];
|
||||
// }
|
||||
|
||||
// //编码修复
|
||||
// $possibleEncodings = [
|
||||
// 'Windows-1252', 'UTF-8', 'GBK', 'GB2312',
|
||||
// 'Latin-1', 'ISO-8859-1', 'CP1252'
|
||||
// ];
|
||||
// $encodedContent = @mb_convert_encoding($sAuthorContent, 'UTF-8', implode(',', $possibleEncodings));
|
||||
// $sAuthorContent = $encodedContent ?: $sAuthorContent;
|
||||
|
||||
// //清理不可见字符
|
||||
// $sAuthorContent = preg_replace('/[\x00-\x1F\x7F\x{200B}-\x{200F}]/u', '', $sAuthorContent);
|
||||
|
||||
// //修复特殊符号乱码
|
||||
// $symbolMap = [
|
||||
// '†' => '†', 'â ' => '†', 'â' => '†', '?†' => '†',
|
||||
// ':' => ':', ',' => ',', '—' => '-',
|
||||
// '啊' => '' // 针对性移除异常字符“啊”(若为固定乱码)
|
||||
// ];
|
||||
// $sAuthorContent = strtr($sAuthorContent, $symbolMap);
|
||||
|
||||
// //格式标准化
|
||||
// $sAuthorContent = str_replace([',', ';', ';', '、'], ',', $sAuthorContent); // 统一分隔符
|
||||
// $sAuthorContent = preg_replace('/\s+and\s+/i', ', ', $sAuthorContent); // and转逗号
|
||||
// $sAuthorContent = preg_replace('/\s+/', ' ', $sAuthorContent); // 合并多余空格
|
||||
// $sAuthorContent = trim($sAuthorContent);
|
||||
// var_dump($this->parseAuthorsWithoutRegex($sAuthorContent));exit;
|
||||
// // 关键预处理:兼容"and"分隔符、清理乱码、统一空格
|
||||
// $content = mb_convert_encoding($sAuthorContent, 'UTF-8', 'auto');
|
||||
// $content = str_replace(["\xC2\xA0", 'ï¼', '�', ','], ' ', $content); // 清理乱码和全角符号
|
||||
// $content = preg_replace('/\band\b/i', ',', $content); // 将 "and" 转为逗号(统一分隔符)
|
||||
// $content = preg_replace('/(\d+)\s*([*#†])/', '$1$2', $content); // 合并数字与符号间的空格(如"1 *"→"1*")
|
||||
// $content = trim(preg_replace('/\s+/', ' ', $content)); // 合并连续空格
|
||||
|
||||
// // 标记上标内的逗号(多编号处理)
|
||||
// $tempStr = preg_replace('/(\d+)\s*,\s*(\d+)/', '$1<SEP>$2', $content);
|
||||
|
||||
// // 核心正则(保持原有结构,扩展符号支持)
|
||||
// $pattern = '/
|
||||
// ([A-Za-z\s\.\-]+?) # 姓名(支持缩写、空格、连字符)
|
||||
// \s* # 姓名与上标间的空格(允许0或多个)
|
||||
// ( # 上标组(扩展兼容所有符号)
|
||||
// \d+ # 起始数字(至少1个数字)
|
||||
// (?:[†#*,]|<SEP>\d+)* # 允许:符号(†#*)、逗号、<SEP>+数字(多编号)
|
||||
// )
|
||||
// \s*,? # 作者间的逗号(可选,允许逗号前有空格)
|
||||
// (?=\s|$) # 确保后面是空格或字符串结尾(避免跨作者匹配)
|
||||
// /ux';
|
||||
|
||||
// preg_match_all($pattern, $tempStr, $matches);
|
||||
|
||||
// // 解析结果并格式化
|
||||
// $authorList = [];
|
||||
// if (!empty($matches[1])) {
|
||||
// foreach ($matches[1] as $i => $name) {
|
||||
// $name = trim($name);
|
||||
// $superscript = trim($matches[2][$i]);
|
||||
// $superscript = str_replace('<SEP>', ',', $superscript); // 恢复多编号逗号
|
||||
// $superscript = preg_replace('/,$/', '', $superscript); // 清理末尾多余逗号
|
||||
// if (!empty($name)) {
|
||||
// $authorList[] = [
|
||||
// 'name' => $name,
|
||||
// 'superscript' => $superscript
|
||||
// ];
|
||||
// }
|
||||
// }
|
||||
// }
|
||||
|
||||
// // 输出结果
|
||||
// echo "<pre>";
|
||||
// print_r($authorList);
|
||||
// echo "</pre>";
|
||||
// exit;
|
||||
|
||||
// // 处理作者
|
||||
// $content = mb_convert_encoding($sAuthorContent, 'UTF-8', 'auto'); // 确保编码正确
|
||||
// $content = str_replace("\xC2\xA0", ' ', $content); // 替换非-breaking空格为普通空格
|
||||
// $content = preg_replace('/(\d+)\s*([*#†])/', '$1$2', $content); // 合并"1 *"为"1*"、"1 #"为"1#"
|
||||
// $content = preg_replace('/,†/', ',†', $content); // 保留"1,†"格式(防止被拆分)
|
||||
|
||||
// //标记上标内的逗号+空格(多编号)
|
||||
// $tempStr = preg_replace('/(\d+)\s*,\s*(\d+)/', '$1<SEP>$2', $content);
|
||||
// // 原有步骤2:正则匹配(扩展上标符号支持,保持原有逻辑)
|
||||
// $pattern = '/
|
||||
// ([A-Za-z\s\.\-]+?) # 姓名(支持缩写、空格)
|
||||
// \s* # 姓名与上标间空格
|
||||
// ( # 上标组(扩展符号支持)
|
||||
// \d+ # 起始数字
|
||||
// (?:[†#*,]|<SEP>\d+)* # 允许:†#*符号、逗号、<SEP>+数字(兼容1,†、1,*等)
|
||||
// )
|
||||
// \s*,? # 作者间逗号(可选)
|
||||
// (?=\s|$) # 确保后面是空格或结尾
|
||||
// /ux';
|
||||
|
||||
// preg_match_all($pattern, $tempStr, $matches);
|
||||
// var_dump($matches);exit;
|
||||
// $authorList = [];
|
||||
// if(!empty($matches[1])){
|
||||
// foreach ($matches[1] as $i => $name) {
|
||||
// $name = trim($name);
|
||||
// $superscript = trim($matches[2][$i]);
|
||||
// $superscript = str_replace('<SEP>', ',', $superscript); // 恢复多编号逗号
|
||||
// $superscript = preg_replace('/,$/', '', $superscript); // 清理末尾逗号
|
||||
// // 修复符号与数字间的空格(如原始"1 *"被误处理为"1*"的情况,保持原样)
|
||||
// $superscript = preg_replace('/(\d)([*#†])/', '$1$2', $superscript);
|
||||
// if (!empty($name)) {
|
||||
// $authorList[] = [
|
||||
// 'name' => $name,
|
||||
// 'superscript' => $superscript
|
||||
// ];
|
||||
// }
|
||||
// }
|
||||
// }else {
|
||||
// // 按“两个或多个连续空格”拆分(姓名之间的分隔)
|
||||
// $authorList = array_filter(
|
||||
// array_map('trim',
|
||||
// preg_split('/(,\p{Z}*|\p{Z}{2,})/u', $sAuthorContent)
|
||||
// )
|
||||
// );
|
||||
// }
|
||||
|
||||
|
||||
// // //处理作者
|
||||
// $aAuthorData = [];
|
||||
// $aReport = [];
|
||||
// $namePattern = '/
|
||||
// (?:[A-Za-z\s·\-\']+| # 英文姓名(支持空格、连字符)
|
||||
// [\x{4e00}-\x{9fa5}]+| # 中文姓名
|
||||
// [\x{1800}-\x{18AF}]+| # 蒙古文姓名
|
||||
// [A-Z]\.) # 单字母缩写(如 J.)
|
||||
// /ux';
|
||||
|
||||
// foreach ($authorList as $authorStr){
|
||||
// if (empty($authorStr)) continue;
|
||||
|
||||
// //获取下标
|
||||
// $superscript = empty($authorStr['superscript']) ? $authorStr : $authorStr['superscript'];
|
||||
// $nameStr = empty($authorStr['name']) ? $authorStr : $authorStr['name'];
|
||||
|
||||
// $companyId = [];
|
||||
// $isSuper = 0;
|
||||
// $isReport = 0;
|
||||
// if (!empty($superscript)) {
|
||||
// // 提取机构编号(忽略上标中的逗号,如1,† → 提取1)
|
||||
// preg_match_all('/\d+/', $superscript, $numMatch);
|
||||
// // 识别特殊符号(#为超级作者,*†为通讯作者)
|
||||
// $isSuper = strpos($superscript, '#') !== false ? 1 : 0;
|
||||
// $isReport = (strpos($superscript, '*') !== false || strpos($superscript, '†') !== false) ? 1 : 0;
|
||||
// }
|
||||
// if (preg_match("/^([A-Za-z\s'\.-]+)/u", $nameStr, $match)) {
|
||||
// $nameStr = trim($match[1]);
|
||||
// }
|
||||
// $aAuthorData[] = [
|
||||
// 'name' => $nameStr,
|
||||
// 'company_id' => empty($numMatch[0]) ? [] : $numMatch[0],
|
||||
// 'is_super' => $isSuper,
|
||||
// 'is_report' => $isReport
|
||||
// ];
|
||||
// if ($isReport) {
|
||||
// $aReport[] = $nameStr;
|
||||
// }
|
||||
// }
|
||||
// return ['author' => $aAuthorData,'report' => array_unique($aReport)];
|
||||
// }
|
||||
|
||||
// 获取机构
|
||||
private function getCompany($aParam = []){
|
||||
@@ -388,32 +646,68 @@ class ArticleParserService
|
||||
//获取标题下的作者
|
||||
$sAuthorContent = empty($aParam['authors']) ? $this->getNextParagraphAfterText($title) : $aParam['authors'];
|
||||
//获取作者结构
|
||||
$sCompany = $this->getContentAfterText($sAuthorContent);
|
||||
if(empty($sCompany)){
|
||||
$allLines = $this->getContentAfterText($sAuthorContent,1);
|
||||
if(empty($allLines)){
|
||||
return [];
|
||||
}
|
||||
//编码修复
|
||||
// 2. 按序号分组,合并同一序号的多行内容
|
||||
$grouped = [];
|
||||
$currentNumber = null; // 当前序号
|
||||
foreach ($allLines as $line) {
|
||||
$line = trim($line);
|
||||
if (empty($line)) continue;
|
||||
|
||||
// 判断是否是新条目的开头:行首为数字(后续可接任意字符或直接接内容)
|
||||
$number = '';
|
||||
$i = 0;
|
||||
$lineLen = strlen($line);
|
||||
// 提取行首的连续数字(作为序号)
|
||||
while ($i < $lineLen && ctype_digit($line[$i])) {
|
||||
$number .= $line[$i];
|
||||
$i++;
|
||||
}
|
||||
|
||||
// 若行首有数字,则视为新条目
|
||||
if (!empty($number)) {
|
||||
$currentNumber = $number;
|
||||
// 提取序号后的内容(跳过数字后的符号/空格,保留核心内容)
|
||||
// 从数字后的位置开始,跳过可能的符号(./*)或空格
|
||||
while ($i < $lineLen && (in_array($line[$i], ['.', '*', ' ']))) {
|
||||
$i++;
|
||||
}
|
||||
$content = trim(substr($line, $i)); // 序号后的内容
|
||||
$grouped[$currentNumber] = $content;
|
||||
continue;
|
||||
}
|
||||
|
||||
// 非新条目,合并到当前序号的内容中
|
||||
if ($currentNumber !== null) {
|
||||
$grouped[$currentNumber] .= ' ' . $line;
|
||||
}
|
||||
}
|
||||
|
||||
//清理结果
|
||||
$possibleEncodings = [
|
||||
'Windows-1252', 'UTF-8', 'GBK', 'GB2312',
|
||||
'Latin-1', 'ISO-8859-1', 'CP1252'
|
||||
];
|
||||
$encodedContent = @mb_convert_encoding($sCompany, 'UTF-8', implode(',', $possibleEncodings));
|
||||
$sCompany = $encodedContent ?: $sCompany;
|
||||
//按行拆分,保留数字开头的行
|
||||
$sCompany = str_replace(["\r\n", "\r"], "\n", $sCompany);
|
||||
$aCompanyLines = explode("\n", $sCompany);
|
||||
$aCompanyLines = array_filter(array_map('trim', $aCompanyLines), function($line) {
|
||||
return preg_match('/^\d+/', $line); // 仅保留数字开头的行
|
||||
});
|
||||
|
||||
$aCompany = [];
|
||||
foreach ($aCompanyLines as $line) {
|
||||
if (preg_match('/^(\d+)\s*(.+)$/', $line, $match)) {
|
||||
if(empty($match[1]) || empty($match[2])){
|
||||
continue;
|
||||
}
|
||||
$aCompany[$match[1]] = ltrim(trim(ltrim($match[2]),'.'),' ');
|
||||
foreach ($grouped as $number => $institution) {
|
||||
$encodedContent = @mb_convert_encoding($institution, 'UTF-8', implode(',', $possibleEncodings));
|
||||
$sCompany = $encodedContent ?: $sCompany;
|
||||
$institution = preg_replace('/\s+/', ' ', $institution); // 合并多余空格
|
||||
$institution = rtrim($institution, '.');
|
||||
$institution = preg_replace('/^\d+\s+/', '', $institution);
|
||||
$institution = trim($institution); // 清理首尾空格
|
||||
preg_match('/(.*?, [A-Za-z]+ \d+, [A-Za-z]+)/', $institution, $institutionmatches);;
|
||||
$institution = trim($institutionmatches[1] ?? $institution);
|
||||
if(!mb_check_encoding($institution, 'UTF-8')){
|
||||
$institution = mb_convert_encoding($institution, 'UTF-8', 'GBK');
|
||||
}
|
||||
if (preg_match('/^(.*?)(?=\s*\*Email)/', $institution, $matches)) {
|
||||
$institution = trim($matches[1]); // trim() 去除内容前后多余空格
|
||||
}
|
||||
$aCompany[$number] = $institution;
|
||||
}
|
||||
return $aCompany;
|
||||
}
|
||||
@@ -451,11 +745,10 @@ class ArticleParserService
|
||||
$corrText = str_replace([':', '@'], [':', '@'], $corrText);
|
||||
$corrText = preg_replace('/\s+/', ' ', $corrText); // 统一空格
|
||||
$corrText = str_replace(' ', ' ', $corrText); // 去除多余空格
|
||||
|
||||
//按"*"分割通讯作者
|
||||
$corrBlocks = preg_split('/\s*\*\s*/', $corrText);
|
||||
$corrBlocks = array_filter(array_map('trim', $corrBlocks));
|
||||
|
||||
|
||||
$aCorresponding = [];
|
||||
foreach ($corrBlocks as $block) {
|
||||
//匹配通讯作者姓名
|
||||
@@ -466,7 +759,6 @@ class ArticleParserService
|
||||
preg_match('/(E[\s-]*mail|邮箱)[\s:]*([^\s]+@[^\s]+)/i', $block, $email);
|
||||
preg_match('/(Postal[\s-]*address|地址)[\s:]*([^,;]+)/i', $block, $address);
|
||||
preg_match('/(Tel|电话)[\s:]*([^\s]+)/i', $block, $tel);
|
||||
|
||||
$aCorresponding[] = [
|
||||
'name' => $sName,
|
||||
'email' => isset($email[2]) ? trim($email[2]) : '',
|
||||
@@ -474,6 +766,24 @@ class ArticleParserService
|
||||
'tel' => isset($tel[2]) ? trim($tel[2]) : ''
|
||||
];
|
||||
}
|
||||
if(empty($aCorresponding)){
|
||||
$pattern = '/Corresponding Authors: (.*?)(?=$|;)/s';
|
||||
preg_match($pattern, $corrText, $match);
|
||||
if (!empty($match[1])) {
|
||||
$corrContent = $match[1];
|
||||
// 提取每个作者的名称和邮箱(优化正则,支持更多字符)
|
||||
$authorPattern = '/([A-Za-z\s]+?),\s*E-mail:\s*([\w@\.\-]+)/';
|
||||
preg_match_all($authorPattern, $corrContent, $authors);
|
||||
if(!empty($authors[1])){
|
||||
for ($i = 0; $i < count($authors[1]); $i++) {
|
||||
$aCorresponding[] = [
|
||||
'name' => empty($authors[1][$i]) ? '' : trim($authors[1][$i]),
|
||||
'email' => empty($authors[2][$i]) ? '' : trim($authors[2][$i])
|
||||
];
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
return $aCorresponding;
|
||||
}
|
||||
|
||||
@@ -518,7 +828,7 @@ class ArticleParserService
|
||||
}
|
||||
|
||||
// 获取目标文本后的所有内容
|
||||
private function getContentAfterText($targetText){
|
||||
private function getContentAfterText($targetText,$return_type = 2){
|
||||
$found = false;
|
||||
$content = [];
|
||||
$stopKeywords = ['关键词', 'Key words', '摘要', 'Abstract'];
|
||||
@@ -559,7 +869,14 @@ class ArticleParserService
|
||||
}
|
||||
if (count($content) >= $maxLines || (isset($shouldStop) && $shouldStop)) break;
|
||||
}
|
||||
return implode("\n", $content);
|
||||
if($return_type == 1){
|
||||
return $content;
|
||||
}
|
||||
$content = implode("\n", $content);
|
||||
if(!mb_check_encoding($content, 'UTF-8')){
|
||||
$content = mb_convert_encoding($content, 'UTF-8', 'GBK');
|
||||
}
|
||||
return $content;
|
||||
}
|
||||
|
||||
// 统一提取元素文本
|
||||
@@ -676,6 +993,9 @@ class ArticleParserService
|
||||
$sContent .= "\n";
|
||||
}
|
||||
}
|
||||
if(!mb_check_encoding($sContent, 'UTF-8')){
|
||||
$sContent = mb_convert_encoding($sContent, 'UTF-8', 'GBK');
|
||||
}
|
||||
// 2. 基础文本清理(合并多余空格,保留有效换行)
|
||||
$textContent = preg_replace('/(\S)\s+/', '$1 ', $sContent);
|
||||
$textContent = trim($textContent);
|
||||
|
||||
Reference in New Issue
Block a user