init

2026-04-18 23:04:03 +08:00
parent 51a380e480
commit 1a4f3bff7d
12 changed files with 1023 additions and 0 deletions
@@ -0,0 +1,141 @@
+import re
+from typing import List, Dict, Tuple, Optional
+
+class ManifestParser:
+    def __init__(self):
+        self.tasks = []  # 存储解析后的任务指令
+
+    def parse(self, content: str):
+        """解析 manifest.txt 或邮件正文，生成内部指令结构"""
+        lines = [line.strip() for line in content.splitlines() if line.strip() and not line.strip().startswith('#')]
+        self.tasks = []
+        current_scope = None  # 当前压缩包作用域，None表示根目录
+
+        i = 0
+        while i < len(lines):
+            line = lines[i]
+            # 处理压缩包作用域开始
+            if line.startswith('in ') and line.endswith(':'):
+                archive_name = line[3:-1].strip()
+                current_scope = archive_name
+                i += 1
+                continue
+            elif line == 'in .':
+                current_scope = None
+                i += 1
+                continue
+
+            # 检查行内是否包含 "in xxx" 临时作用域
+            inline_archive = None
+            if ' in ' in line:
+                parts = line.split(' in ')
+                cmd_part = parts[0].strip()
+                archive_part = parts[1].strip()
+                # 如果archive_part还有更多内容（如"to: xxx"），需重新组合？假设格式严格："指令 in archive.zip"
+                if ' ' in archive_part:
+                    # 复杂情况，暂不支持，要求用户使用单独行
+                    pass
+                else:
+                    inline_archive = archive_part
+                    line = cmd_part
+
+            # 解析指令
+            if line.startswith('to:'):
+                # 全局默认
+                targets = self._parse_targets(line[3:])
+                self.tasks.append({
+                    'type': 'global',
+                    'targets': targets,
+                    'scope': current_scope,
+                    'inline_archive': inline_archive
+                })
+            elif line.startswith('from:'):
+                # 按格式批量转换
+                match = re.match(r'from:\s*(\S+)\s+to:\s*(.+)', line)
+                if match:
+                    src_format = match.group(1).lower()
+                    targets = self._parse_targets(match.group(2))
+                    self.tasks.append({
+                        'type': 'by_format',
+                        'src_format': src_format,
+                        'targets': targets,
+                        'scope': current_scope,
+                        'inline_archive': inline_archive
+                    })
+            elif line.startswith('from.name:'):
+                # 精准文件名匹配
+                rest = line[10:].strip()
+                if ' to.name:' in rest:
+                    # 重命名
+                    parts = rest.split(' to.name:')
+                    src_name = parts[0].strip()
+                    dst_spec = parts[1].strip()
+                    # 解析 dst_spec: "newname.ext(size)"
+                    dst_name, size = self._parse_dst_with_size(dst_spec)
+                    self.tasks.append({
+                        'type': 'rename',
+                        'src_name': src_name,
+                        'dst_name': dst_name,
+                        'size': size,
+                        'scope': current_scope,
+                        'inline_archive': inline_archive
+                    })
+                else:
+                    # to: 格式
+                    parts = rest.split(' to:')
+                    src_name = parts[0].strip()
+                    targets = self._parse_targets(parts[1])
+                    self.tasks.append({
+                        'type': 'by_name',
+                        'src_name': src_name,
+                        'targets': targets,
+                        'scope': current_scope,
+                        'inline_archive': inline_archive
+                    })
+            elif line.startswith('to.name:'):
+                # 多版本导出中的 to.name 行，需要结合前面的 from.name 处理
+                # 我们在主流程中采用累积方式，这里简单处理，实际在main中会连续读取
+                # 为简化，不在此处实现多行组合，而是在主程序中按行处理上下文
+                # 让主程序负责收集连续的 to.name
+                pass
+            else:
+                logger.warning(f"无法识别的指令: {line}")
+            i += 1
+        return self.tasks
+
+    def _parse_targets(self, target_str: str) -> List[Tuple[str, Optional[Tuple]]]:
+        """解析 "webp, png(16:9), jpg(800x600)" -> [('webp',None), ('png',('16:9')), ('jpg',('800x600'))]"""
+        items = [t.strip() for t in target_str.split(',')]
+        result = []
+        for item in items:
+            size = None
+            if '(' in item and item.endswith(')'):
+                fmt, size_part = item.split('(', 1)
+                size_str = size_part.rstrip(')')
+                if ':' in size_str:
+                    # 比例
+                    w_ratio, h_ratio = map(int, size_str.split(':'))
+                    size = ('ratio', w_ratio, h_ratio)
+                elif 'x' in size_str:
+                    w, h = map(int, size_str.split('x'))
+                    size = ('pixel', w, h)
+                result.append((fmt.lower(), size))
+            else:
+                result.append((item.lower(), None))
+        return result
+
+    def _parse_dst_with_size(self, dst_spec: str) -> Tuple[str, Optional[Tuple]]:
+        """解析 "banner.webp(800x600)" -> ('banner.webp', ('pixel',800,600))"""
+        size = None
+        if '(' in dst_spec and dst_spec.endswith(')'):
+            name, size_part = dst_spec.split('(', 1)
+            size_str = size_part.rstrip(')')
+            if ':' in size_str:
+                w, h = map(int, size_str.split(':'))
+                size = ('ratio', w, h)
+            elif 'x' in size_str:
+                w, h = map(int, size_str.split('x'))
+                size = ('pixel', w, h)
+            return name.strip(), size
+        else:
+            return dst_spec.strip(), None