Module:PageSummary

此模块的文档可以在Module:PageSummary/doc创建

local p = {}

-- 核心剥离逻辑：状态机识别，绝对不回溯，绝对不死循环
local function stripImages(text)
    if not text then return "" end
    
    local out = {}
    local i = 1
    local len = mw.ustring.len(text)
    
    while i <= len do
        local foundImage = false
        -- 检查是否是图片起始
        local chunk = mw.ustring.sub(text, i, i + 10)
        if chunk:match("^%[%[[Ff]ile:") or chunk:match("^%[%[[Ii]mage:") or 
           chunk:match("^%[%[文件:") or chunk:match("^%[%[图像:") then
            
            -- 进入图片跳过模式
            local depth = 0
            local j = i
            while j <= len do
                local char2 = mw.ustring.sub(text, j, j + 1)
                if char2 == "[[" then
                    depth = depth + 1
                    j = j + 2
                elseif char2 == "]]" then
                    depth = depth - 1
                    j = j + 2
                    if depth <= 0 then
                        i = j -- 跳过整段图片
                        foundImage = true
                        break
                    end
                else
                    j = j + 1
                end
            end
        end
        
        if not foundImage then
            table.insert(out, mw.ustring.sub(text, i, i))
            i = i + 1
        end
        
        -- 安全熔断：如果处理太长，强制中断
        if i > 600 then break end
    end
    
    return table.concat(out)
end

local function cleanFinal(text)
    -- 移除模板
    text = string.gsub(text, "{{[^{}]+}}", "")
    text = string.gsub(text, "{{[^{}]+}}", "")
    -- 移除引用和注释
    text = string.gsub(text, "<ref.-</ref>", "")
    text = string.gsub(text, "<ref.->", "")
    text = string.gsub(text, "<!%-%-.-%-%->", "")
    -- 移除 Wiki 格式
    text = string.gsub(text, "\n==+.-==+", " ")
    text = string.gsub(text, "\n%s*[*#:]+", " ")
    text = string.gsub(text, "\n", " ")
    text = string.gsub(text, "%s+", " ")
    return mw.text.trim(text)
end

function p.getSummaryAndImage(frame)
    local pageName = frame.args[1] or ""
    local title = mw.title.new(pageName)
    if not title or not title.exists then return "" end
    
    local content = title:getContent()
    local rawExcerpt = mw.ustring.sub(content, 1, 400)

    -- 提取第一张图（用于显示）
    local firstImage = string.match(rawExcerpt, "%[%[([Ff]ile:[^|%]%s]+)") or 
                       string.match(rawExcerpt, "%[%[(文件:[^|%]%s]+)")

    -- 执行两步清理
    local step1 = stripImages(rawExcerpt)
    local cleanText = cleanFinal(step1)

    -- 截断
    local targetLen = 140
    local summary = mw.ustring.sub(cleanText, 1, targetLen)
    
    -- 清理末尾断头链接
    local _, opens = string.gsub(summary, "%[%[", "")
    local _, closes = string.gsub(summary, "%]%]", "")
    if opens > closes then
        local lastOpen = string.find(summary:reverse(), "%[%[")
        if lastOpen then
            summary = string.sub(summary, 1, #summary - lastOpen - 1)
        end
    end

    summary = mw.text.trim(summary)
    if mw.ustring.len(cleanText) > targetLen then summary = summary .. "..." end

    -- 渲染
    local container = mw.html.create('div'):css({['margin-bottom']='25px', ['display']='flow-root'})
    if firstImage then
        container:wikitext('[[' .. firstImage .. '|130px|right|link=' .. pageName .. ']]')
    end
    container:tag('div')
        :css({['font-size']='1.3em', ['font-weight']='bold', ['margin-bottom']='6px'})
        :wikitext('[[' .. pageName .. ']]')
    container:tag('div')
        :css({['line-height']='1.5', ['color']='#222'})
        :wikitext(summary)

    return tostring(container)
end

return p