zm/tougao
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

测试修改

Browse Source
This commit is contained in:
chengxl
2025-11-06 17:00:20 +08:00
parent 6d867d6088
commit 3c459af7fc
1 changed files with 191 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

201
application/common/ArticleParserService.php
View File

@@ -33,9 +33,162 @@ class ArticleParserService
$this->phpWord = $reader->load($filePath);
$this->sections = $this->phpWord->getSections();
} catch (\Exception $e) {
// 预处理：移除 DOCX 中的 EMF 图片
$processedFilePath = $this->removeEmfFromDocx($filePath);
// 加载处理后的文档
$reader = IOFactory::createReader();
$reader->setReadDataOnly(false);
Settings::setCompatibility(false);
Settings::setOutputEscapingEnabled(true);
$this->phpWord = $reader->load($processedFilePath);
$this->sections = $this->phpWord->getSections();
// 可选：删除临时处理文件（避免冗余）
var_dump($processedFilePath);
unlink($processedFilePath);
return json_encode(['status' => 5, 'msg' => $e->getMessage()]);
}
}
/**
* 移除 DOCX 压缩包内的所有 EMF 图片
* @param string $docxPath 原 DOCX 文件路径
* @return string 处理后的临时 DOCX 路径
*/
private function removeEmfFromDocx($docxPath){
$zip = new ZipArchive();
if ($zip->open($docxPath) !== true) {
throw new \Exception("无法打开 DOCX 文件：{$docxPath}");
}
// 1. 创建临时目录用于解压
$tempDir = rtrim(ROOT_PATH,'/').'/runtime/'.uniqid('docx_temp_');
mkdir($tempDir, 0700, true);
// 2. 解压 DOCX 到临时目录
$zip->extractTo($tempDir);
$zip->close();
// 3. 递归删除所有 EMF 文件
$dirIterator = new RecursiveDirectoryIterator($tempDir);
$iterator = new RecursiveIteratorIterator($dirIterator);
foreach ($iterator as $file) {
if ($file->isFile() && strtolower(pathinfo($file, PATHINFO_EXTENSION)) === 'emf') {
unlink($file->getPathname());
}
}
// 4. 重新打包为 DOCX
$processedPath = $tempDir . '_processed.docx';
$newZip = new ZipArchive();
if ($newZip->open($processedPath, ZipArchive::CREATE | ZipArchive::OVERWRITE) !== true) {
throw new \Exception("无法创建处理后的 DOCX 文件");
}
// 遍历临时目录，添加所有文件到新压缩包
$this->addFilesToZip($tempDir, $newZip);
$newZip->close();
// 5. 删除临时解压目录
$this->deleteDir($tempDir);
return $processedPath;
}
/**
* 递归添加目录文件到 ZipArchive
* @param string $dir 目录路径
* @param ZipArchive $zip ZipArchive 实例
*/
private function addFilesToZip($dir, $zip)
{
$files = scandir($dir);
foreach ($files as $file) {
if ($file === '.' || $file === '..') continue;
$filePath = $dir . '/' . $file;
if (is_dir($filePath)) {
$this->addFilesToZip($filePath, $zip);
} else {
// 计算压缩包内的相对路径（避免冗余目录层级）
$relativePath = str_replace(dirname($dir) . '/', '', $filePath);
$zip->addFile($filePath, $relativePath);
}
}
}
/**
* 递归删除目录
* @param string $dir 目录路径
*/
private function deleteDir($dir){
// 1. 基础校验：非空字符串且为有效目录
if (trim($dir) === '' || !is_dir($dir)) {
return false;
}
// 2. 统一路径格式（去除尾部分隔符，避免跨系统差异）
$dir = rtrim($dir, DIRECTORY_SEPARATOR);
$dirName = basename($dir);
// 3. 前缀强校验：仅处理docx_temp_开头的目录
if (strpos($dirName, 'docx_temp_') !== 0) {
return false;
}
// 4. 路径归属校验（缓存realpath结果，减少I/O）
$runtimeDir = rtrim(ROOT_PATH, '/') . '/runtime';
$realDir = realpath($dir);
$realRuntimeDir = realpath($runtimeDir);
if ($realDir === false || $realRuntimeDir === false || strpos($realDir, $realRuntimeDir) !== 0) {
return false;
}
// 5. 扫描目录（带错误抑制，处理权限问题）
$files = @scandir($dir);
if ($files === false) {
return false;
}
$isFullyDeleted = true; // 标记是否完全删除
// 6. 递归处理子项
foreach ($files as $file) {
if ($file === '.' || $file === '..') {
continue;
}
$filePath = $dir . DIRECTORY_SEPARATOR . $file;
$realFilePath = realpath($filePath);
// 子路径校验：必须是当前目录的子项（防符号链接跳转）
if ($realFilePath === false || strpos($realFilePath, $realDir) !== 0) {
$isFullyDeleted = false;
continue;
}
if (is_dir($realFilePath)) {
// 递归删除子目录，继承校验逻辑
if (!$this->deleteDir($realFilePath)) {
$isFullyDeleted = false;
}
} else {
// 尝试删除文件（失败则标记未完全删除）
if (!@unlink($realFilePath)) {
$isFullyDeleted = false;
}
}
}
// 7. 最终删除目录（确保空目录才删除）
$remainingFiles = @scandir($dir);
if ($remainingFiles !== false && count($remainingFiles) <= 2) {
@rmdir($dir);
return $isFullyDeleted; // 若子项完全删除，则返回true
}
return false;
}
// 上传并解析文档的入口方法
public static function uploadAndParse($sFileUrl){
@@ -79,9 +232,11 @@ class ArticleParserService
// 提取文章标题
private function getTitle(){
if(empty($this->sections)){
return '';
}
$title = '';
$maxLength = 0;
foreach ($this->sections as $section) {
foreach ($section->getElements() as $element) {
$text = $this->getTextFromElement($element);
@@ -93,6 +248,9 @@ class ArticleParserService
}
}
}
if(!empty($title) && !mb_check_encoding($title, 'UTF-8')){
$title = mb_convert_encoding($title, 'UTF-8', 'GBK');
}
return $title;
}
// 提取作者
@@ -462,6 +620,9 @@ class ArticleParserService
if(empty($value['name']) && empty($value['superscript'])){
continue;
}
if(!mb_check_encoding($value['name'], 'UTF-8')){
$value['name'] = mb_convert_encoding($value['name'], 'UTF-8', 'GBK');
}
if(!empty($value['name']) && !empty($value['is_report']) && $value['is_report'] == 1){
$aReport[] = $value['name'];
}
@@ -701,12 +862,12 @@ class ArticleParserService
$institution = trim($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
}
if(!empty($institution) && !mb_check_encoding($institution, 'UTF-8')){
$institution = mb_convert_encoding($institution, 'UTF-8', 'GBK');
}
$aCompany[$number] = $institution;
}
return $aCompany;
@@ -831,7 +992,7 @@ class ArticleParserService
private function getContentAfterText($targetText,$return_type = 2){
$found = false;
$content = [];
$stopKeywords = ['关键词', 'Key words', '摘要', 'Abstract'];
$stopKeywords = ['关键词', 'Key words', '摘要', 'Abstract','ABSTRACT'];
$maxLines = 200;
$lineNumber = 0;
foreach ($this->sections as $section) {
@@ -873,7 +1034,7 @@ class ArticleParserService
return $content;
}
$content = implode("\n", $content);
if(!mb_check_encoding($content, 'UTF-8')){
if(!empty($content) && !mb_check_encoding($content, 'UTF-8')){
$content = mb_convert_encoding($content, 'UTF-8', 'GBK');
}
return $content;
@@ -952,7 +1113,9 @@ class ArticleParserService
$text = preg_replace('/[\x00-\x1F\x7F-\x9F]/', ' ', $text); // 移除控制字符
$text = str_replace(["\t", "\r", "\n"], ' ', $text); // 统一空白字符
$text = preg_replace('/\s+/', ' ', $text); // 合并多个空格
if(!empty($text) && !mb_check_encoding($text, 'UTF-8')){
$text = mb_convert_encoding($text, 'UTF-8', 'GBK');
}
return $text;
}
@@ -993,7 +1156,8 @@ class ArticleParserService
$sContent .= "\n";
}
}
if(!mb_check_encoding($sContent, 'UTF-8')){
if(!empty($sContent) && !mb_check_encoding($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）
@@ -1002,7 +1166,7 @@ class ArticleParserService
// 3. 提取摘要
$abstract = '';
$abstractPattern = '/Abstract\s*([\s\S]*?)(?=Keywords:|$)/i';
$abstractPattern = '/Abstract\s*([\s\S]*?)(?=Keywords|$)/i';
if (preg_match($abstractPattern, $textContent, $abstractMatches)) {
$abstract = trim($abstractMatches[1]);
$abstract = preg_replace('/\n+/', ' ', $abstract);
@@ -1010,7 +1174,8 @@ class ArticleParserService
// 4. 提取关键词（核心：仅保留两种强制匹配逻辑）
$keywords = [];
// $keywordPattern = '/Keywords:\s*([\s\S]*?)(?=\s*\d+\.|[;,]\s*[\r\n]+\s*[\r\n]+|(?i)\bintroduction|abbreviations\b|$)/i';
$keywordPattern = '/Keywords:\s*(.*?)\s*Keywords-End-Flag/s';
$keywordPattern = '/Keywords\s*(.*?)\s*Keywords-End-Flag/s';
if (preg_match($keywordPattern, $textContent, $keywordMatches)) {
$keywordStr = trim($keywordMatches[1]);
@@ -1025,6 +1190,22 @@ class ArticleParserService
return !empty($item) && !ctype_space($item);
});
}
if(empty($keywords)){
$keywordPattern = '/Keywords\s*([\s\S]*?)(?=Introduction|$)/i';
if (preg_match($keywordPattern, $textContent, $keywordMatches)) {
$keywordStr = trim($keywordMatches[1]);
// 清理关键词列表格式（去除换行、末尾多余符号）
$keywordStr = preg_replace('/\n+/', ' ', $keywordStr);
$keywordStr = rtrim($keywordStr, ';,. '); // 去除末尾分号、逗号等
$keywordStr = trim($keywordStr);
// 分割并过滤有效关键词
$keywords = preg_split('/[,;]\s*/', $keywordStr);
$keywords = array_filter(array_map('trim', $keywords), function($item) {
return !empty($item) && !ctype_space($item);
});
}
}
return [
'status' => 1,
'msg' => '提取成功',