Module:PageSummary

此模块的文档可以在Module:PageSummary/doc创建

local p = {}

-- 強化版清理：徹底根除內文圖片，保留加粗和連結
local function cleanContent(text)
    if not text then return "" end

    -- 1. 基礎清理（注釋、腳注、表格、模板）
    text = mw.ustring.gsub(text, "<!%-%-.-%-%->", "")
    text = mw.ustring.gsub(text, "<ref[^>]*>.-</ref>", "")
    text = mw.ustring.gsub(text, "<ref[^>]-/>", "")
    
    -- 移除表格 {| ... |}
    local prev
    repeat
        prev = text
        text = mw.ustring.gsub(text, "{|[^{}]*|}", "")
    until text == prev

    -- 移除模板 {{ ... }}
    local count = 0
    repeat
        prev = text
        text = mw.ustring.gsub(text, "{{[^{}]-}}", "")
        count = count + 1
    until text == prev or count > 10

    -- 2. 使用安全占位符保護連結，剔除圖片
    repeat
        prev = text
        text = mw.ustring.gsub(text, "%[%[%s*([^%[%]]-)%s*%]%]", function(inner)
            local low = mw.ustring.lower(inner)
            if low:match("^file:") or low:match("^image:") or 
               low:match("^文件:") or low:match("^圖像:") or
               low:match("^category:") or low:match("^分類:") then
                return "" 
            end
            return "LINKSTART" .. inner .. "LINKEND"
        end)
    until text == prev

    -- 3. 清理剩餘雜質
    text = mw.ustring.gsub(text, "\n==+.-==+", " ") 
    text = mw.ustring.gsub(text, "\n%s*[*#:]+", " ")
    text = mw.ustring.gsub(text, "\n+", " ")
    text = mw.ustring.gsub(text, "%s+", " ")

    -- 4. 初步恢復連結標籤
    text = mw.ustring.gsub(text, "LINKSTART", "[[")
    text = mw.ustring.gsub(text, "LINKEND", "]]")

    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

    -- 【調整】讀取前 1000 字以確保跨過開頭的表格/模板區
    local rawContent = title:getContent() or ""
    local limitedContent = mw.ustring.sub(rawContent, 1, 1000)

    -- 1. 提取第一張縮圖名
    local firstImage = mw.ustring.match(limitedContent, "%[%[%s*[Ff]ile%s*:([^|%]%s\n]+)") or 
                       mw.ustring.match(limitedContent, "%[%[%s*文件%s*:([^|%]%s\n]+)") or
                       mw.ustring.match(limitedContent, "%[%[%s*[Ii]mage%s*:([^|%]%s\n]+)")

    -- 2. 執行清理（此時會得到過濾掉干擾後的純文字）
    local cleanText = cleanContent(limitedContent)
    
    -- 3. 【調整】最終截取約 80 字
    local targetLen = 80
    local summary = ""
    
    if mw.ustring.len(cleanText) <= targetLen then
        summary = cleanText
    else
        -- 截取前 80 字並加上省略號
        summary = mw.ustring.sub(cleanText, 1, targetLen) .. "..."
        
        -- 閉合加粗標籤 '''
        local _, opens = mw.ustring.gsub(summary, "'''", "")
        if opens % 2 ~= 0 then summary = summary .. "'''" end
        
        -- 閉合或清理截斷的連結 [[...
        if mw.ustring.match(summary, "%[%[[^%]]*$") then
            summary = mw.ustring.gsub(summary, "%[%[[^%]]*$", "") .. "..."
        end
    end

    -- 4. 渲染 HTML
    local res = mw.html.create('div'):css({['display'] = 'flow-root', ['line-height'] = '1.5'})
    
    -- 標題
    res:tag('div')
       :css({['font-size'] = '1.15em', ['font-weight'] = 'bold', ['margin-bottom'] = '4px'})
       :wikitext('[[' .. pageName .. ']]')

    -- 右側縮圖
    if firstImage then
        res:wikitext('[[File:' .. firstImage .. '|100px|right|link=' .. pageName .. ']]')
    end

    -- 摘要正文
    res:wikitext(summary)

    return tostring(res)
end

return p