manticore-smash 3.1.0 → 3.3.0

data/lib/manticore.rb

@@ -22,3 +22,8 @@ require_relative 'xmlutils/xpath'
 require_relative 'xmlutils/formatters'
 require_relative 'xmlutils/xml_doc'
 require_relative 'mdutils/rediscount'
+require_relative 'xlsxkit/zip_reader'
+require_relative 'xlsxkit/zip_writer'
+require_relative 'xlsxkit/sax_parser'
+require_relative 'xlsxkit/workbook'
+require_relative 'xlsxkit/writer'

data/lib/xlsxkit/sax_parser.rb

@@ -0,0 +1,294 @@
+# frozen_string_literal: false
+
+# Copyright (C) 2024 Manticore Authors
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published
+# by the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+require 'stringio'
+
+module XlsxKit
+  ##
+  # 轻量 SAX 风格 XML 解析器，专为 XLSX 内部 XML 流式处理设计。
+  #
+  # 与 XmlUtils::TreeParser（DOM 全量建树）不同，本解析器不在内存中构建节点树，
+  # 而是逐标签触发回调（start_element / end_element / characters），内存占用恒定。
+  #
+  # 核心算法：增量缓冲 — 每次从 IO 读取一块数据追加到 @buf，
+  # 尝试提取已完整的标签/文本段并回调，不完整的尾部保留到下次。
+  # 这样即使数百 MB 的 sheet XML 也只占用几十 KB 内存。
+  #
+  # Handler 协议（鸭子类型，无需继承）：
+  #   start_element(name, attrs)   — 遇到开始标签
+  #   end_element(name)            — 遇到结束标签
+  #   characters(text)             — 遇到文本内容（可能分多次回调）
+
+  class SAXParser
+    BUFFER_SIZE = 65536  # 64KB 读取块
+
+    def self.parse(source, handler)
+      new(source, handler).parse
+    end
+
+    def initialize(source, handler)
+      @source  = source
+      @handler = handler
+      @buf     = ''.b
+      @eof     = false
+      @i       = 0  # 已消费位置
+    end
+
+    def parse
+      until @eof && @i >= @buf.length
+        fill_buffer if !@eof && (@buf.length - @i) < BUFFER_SIZE
+        step || (@eof = true)
+      end
+      # 尾部残余文本
+      if @i < @buf.length
+        text = @buf[@i..]
+        emit_text(text) unless text.strip.empty?
+      end
+    end
+
+    private
+
+    def fill_buffer
+      chunk = @source.read(BUFFER_SIZE)
+      if chunk && !chunk.empty?
+        @buf << chunk
+      else
+        @eof = true
+      end
+    end
+
+    ##
+    # 尝试从 @buf 的 @i 位置提取一个完整元素（文本段或标签），
+    # 成功返回 true（已消费），失败返回 nil（需要更多数据）。
+    def step
+      return nil if @i >= @buf.length
+
+      lt = @buf.index('<', @i)
+
+      # --- 情况一：当前位置是 '<'，解析标签 ---
+      if @i == lt
+        parse_tag
+      # --- 情况二：有 '<' 在前方，先处理文本 ---
+      elsif lt
+        text = @buf[@i...lt]
+        @i = lt
+        emit_text(text) unless text.strip.empty?
+        true  # 文本已消费，下次循环处理标签
+      # --- 情况三：无 '<'，整个缓冲区都是文本（或已到 EOF）---
+      else
+        if @eof
+          text = @buf[@i..]
+          @i = @buf.length
+          emit_text(text) unless text.strip.empty?
+          true
+        else
+          # 留最后 1 字节（可能是 '<' 的一半... 虽然不会发生），等下次
+          nil
+        end
+      end
+    end
+
+    #---------------------------------------------------------------------------
+    # 标签解析
+    #---------------------------------------------------------------------------
+
+    def parse_tag
+      # 确保 '>' 存在
+      return nil unless ensure_char('>')
+
+      ch = @buf[@i + 1]
+
+      case ch
+      when '!'
+        parse_special
+      when '?'
+        parse_pi
+      when '/'
+        parse_end_tag
+      else
+        parse_start_tag
+      end
+    end
+
+    def parse_start_tag
+      # 注意：属性值中可能包含 '>'，因此需逐字符扫描，跳过引号区域
+      gt_pos = find_tag_end(@i)
+      return nil unless gt_pos
+
+      raw = @buf[(@i + 1)...gt_pos]
+      @i = gt_pos + 1
+
+      self_closing = raw.end_with?('/')
+      raw = raw[0..-2] if self_closing
+
+      name, attrs = extract_name_and_attrs(raw)
+      @handler.start_element(name, attrs) if @handler.respond_to?(:start_element)
+      @handler.end_element(name) if self_closing && @handler.respond_to?(:end_element)
+
+      true
+    end
+
+    def parse_end_tag
+      gt_pos = @buf.index('>', @i)
+      return nil unless gt_pos
+
+      name = @buf[(@i + 2)...gt_pos].strip
+      @i = gt_pos + 1
+      @handler.end_element(name) if @handler.respond_to?(:end_element)
+      true
+    end
+
+    def parse_special
+      # <!-- comment --> 或 <![CDATA[ ... ]]>
+      if @buf[(@i + 2), 2] == '--'
+        end_idx = @buf.index('-->', @i)
+        return nil unless end_idx
+        @i = end_idx + 3
+      elsif @buf[(@i + 2), 7] == '[CDATA['
+        end_idx = @buf.index(']]>', @i)
+        return nil unless end_idx
+
+        data = @buf[(@i + 9)...end_idx]
+        emit_text(data)
+        @i = end_idx + 3
+      else
+        gt_pos = @buf.index('>', @i)
+        return nil unless gt_pos
+        @i = gt_pos + 1
+      end
+      true
+    end
+
+    def parse_pi
+      close = @buf.index('?>', @i)
+      if close
+        @i = close + 2
+      else
+        gt_pos = @buf.index('>', @i)
+        return nil unless gt_pos
+        @i = gt_pos + 1
+      end
+      true
+    end
+
+    #---------------------------------------------------------------------------
+    # 辅助：标签内扫描（跳过引号区域内的 '>'）
+    #---------------------------------------------------------------------------
+
+    def find_tag_end(start)
+      i = start + 1
+      in_quote = nil
+
+      while i < @buf.length
+        ch = @buf[i]
+
+        if in_quote
+          i += 1
+          in_quote = nil if ch == in_quote
+        elsif ch == '"' || ch == "'"
+          in_quote = ch
+          i += 1
+        elsif ch == '>'
+          return i
+        else
+          i += 1
+        end
+      end
+
+      # 未找到 '>'
+      nil
+    end
+
+    def ensure_char(target)
+      @buf.index(target, @i) ? true : false
+    end
+
+    def extract_name_and_attrs(raw)
+      # raw 形如: 'c r="A1" t="s"' 或 'c r="A1"/'
+      i = 0
+      i += 1 while raw[i] && raw[i] =~ /[A-Za-z0-9_:.\-]/
+      name = raw[0...i]
+
+      attrs = {}
+      while i < raw.length
+        i += 1 while raw[i] =~ /\s/
+        break if i >= raw.length
+
+        # 属性名
+        nstart = i
+        i += 1 while raw[i] && raw[i] !~ /[\s=\/]/
+        aname = raw[nstart...i]
+
+        # 跳过空格
+        i += 1 while raw[i] =~ /\s/
+
+        if raw[i] == '='
+          i += 1
+          i += 1 while raw[i] =~ /\s/
+          quote = raw[i]
+          if quote == '"' || quote == "'"
+            i += 1
+            vend = raw.index(quote, i)
+            attrs[aname] = unescape(raw[i...vend])
+            i = vend + 1
+          else
+            vstart = i
+            i += 1 while raw[i] && raw[i] !~ /[\s\/]/
+            attrs[aname] = raw[vstart...i]
+          end
+        else
+          attrs[aname] = nil
+        end
+      end
+
+      [name, attrs]
+    end
+
+    #---------------------------------------------------------------------------
+    # 实体反转义
+    #---------------------------------------------------------------------------
+
+    ENTITY_MAP = {
+      'amp'  => '&',
+      'lt'   => '<',
+      'gt'   => '>',
+      'quot' => '"',
+      'apos' => "'"
+    }.freeze
+
+    def emit_text(text)
+      decoded = unescape(text)
+      @handler.characters(decoded) if @handler.respond_to?(:characters)
+    end
+
+    def unescape(str)
+      return '' if str.nil? || str.empty?
+      str = str.dup.force_encoding('UTF-8')
+      str.gsub!(/&(amp|lt|gt|quot|apos|#\d+|#x[0-9A-Fa-f]+);/) do
+        m = $1
+        if m.start_with?('#x')
+          [m[2..].to_i(16)].pack('U')
+        elsif m.start_with?('#')
+          [m[1..].to_i].pack('U')
+        else
+          ENTITY_MAP[m] || "&#{m};"
+        end
+      end
+      str
+    end
+  end
+end