Module:PageSummary

此模块的文档可以在Module:PageSummary/doc创建

local p = {}

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 "页面不存在" end
    local content = title:getContent()

    -- 1. 提取第一张图（提取动作在清理前完成）
    local firstImage = string.match(content, "%[%[([Ff]ile:.-)[|%]]") or 
                       string.match(content, "%[%[([Ii]mage:.-)[|%]]") or
                       string.match(content, "%[%[(文件:.-)[|%]]") or
                       string.match(content, "%[%[(图像:.-)[|%]]")

    -- 2. 处理内容：预加载 1000 字符
    local rawText = mw.ustring.sub(content, 1, 1000)
    
    -- A. 高级清理图片：处理带有长描述和嵌套链接的图片
    -- 循环匹配 [[File: ... ]] 结构，尽量处理嵌套情况
    local function stripImages(text)
        -- 匹配 [[(File|文件|Image|图像): ... ]] 
        -- 此正则通过排除法尽可能多地包含字符，直到匹配到对应的闭合括号
        local patterns = {"%[%[[Ff]ile:.-%]%]", "%[%[[Ii]mage:.-%]%]", "%[%[文件:.-%]%]", "%[%[图像:.-%]%]" }
        for _, pat in ipairs(patterns) do
            while string.find(text, pat) do
                text = string.gsub(text, pat, "")
            end
        end
        return text
    end
    rawText = stripImages(rawText)
    
    -- B. 移除 Wiki 列表符号 (*, #, :, ;)
    rawText = string.gsub(rawText, "\n%s*[*#:]+", " ")
    
    -- C. 移除标题符号 == 标题 ==
    rawText = string.gsub(rawText, "\n==+.-==+", " ")
    
    -- D. 移除模板 {{...}}
    local limit = 15
    while string.find(rawText, "{{.-}}") and limit > 0 do
        rawText = string.gsub(rawText, "{{[^{}]+}}", "")
        limit = limit - 1
    end
    
    -- E. 移除引用 <ref> 和 注释
    rawText = string.gsub(rawText, "<ref.-</ref>", "")
    rawText = string.gsub(rawText, "<ref.->", "")
    rawText = string.gsub(rawText, "<!%-%-.-%-%->", "")
    
    -- F. 清理多余的空白符、换行符和特殊占位符（如 &nbsp;）
    rawText = string.gsub(rawText, "&nbsp;", " ")
    rawText = string.gsub(rawText, "\n", " ")
    rawText = string.gsub(rawText, "%s+", " ")
    
    local cleanText = mw.text.trim(rawText)

    -- 3. 截断逻辑：150 字
    local targetLen = 150
    local summary = ""
    if mw.ustring.len(cleanText) <= targetLen then
        summary = cleanText
    else
        summary = mw.ustring.sub(cleanText, 1, targetLen)
        local restText = mw.ustring.sub(cleanText, targetLen + 1)
        
        -- 链接延展逻辑
        local lastOpen = 0
        local lastClose = 0
        local tempPos = 0
        while true do
            local found = string.find(summary, "%[%[", tempPos + 1)
            if not found then break end
            lastOpen = found
            tempPos = found
        end
        tempPos = 0
        while true do
            local found = string.find(summary, "%]%]", tempPos + 1)
            if not found then break end
            lastClose = found
            tempPos = found
        end

        if lastOpen > lastClose then
            local endOfLink = string.find(restText, "%]%]")
            if endOfLink then
                summary = summary .. string.sub(restText, 1, endOfLink + 1)
            end
        end
        summary = summary .. "..."
    end

    -- 4. 渲染
    local container = mw.html.create('div')
        :css({
            ['margin'] = '15px 0',
            ['background'] = 'transparent',
            ['display'] = 'flow-root'
        })

    -- 原生格式图片
    if firstImage then
        container:wikitext('[[' .. firstImage .. '|150px|right|link=' .. pageName .. ']]')
    end

    -- 标题
    container:tag('div')
        :css({
            ['font-size'] = '1.4em',
            ['font-weight'] = 'bold',
            ['margin-bottom'] = '8px'
        })
        :wikitext('[[' .. pageName .. ']]')

    -- 摘要文本
    container:tag('div')
        :css({ 
            ['line-height'] = '1.6',
            ['color'] = '#202122'
        })
        :wikitext(summary)

    return tostring(container)
end

return p