武外梗百科 - 用户贡献 [zh]

Module:PageSummary

2026-02-17T16:02:06Z

240D:C010:18:4:0:0:0:95：

local p = {}

-- 剥离 [[File:]] [[文件:]] [[Category:]] 等，避免污染摘要文本
local function stripImages(text)
if not text or text == "" then return "" end

text = text:gsub("%[%[(文件|Image|图像):", "[[File:")
:gsub("%[%[(分类|Category):", "[[Category:")

local out = {}
local i = 1
local len = #text

while i <= len do
if text:byte(i) == 91 and text:byte(i+1) == 91 then
local chunk = text:sub(i, i+15)
if chunk:match("^%[%[File:") or chunk:match("^%[%[Category:") then
local depth = 1
local j = i + 2
while j <= len and depth > 0 do
if text:byte(j) == 91 and text:byte(j+1) == 91 then
depth = depth + 1
j = j + 2
elseif text:byte(j) == 93 and text:byte(j+1) == 93 then
depth = depth - 1
j = j + 2
else
j = j + 1
end
end
i = j
else
table.insert(out, text:sub(i, i+1)) -- 保留普通 [[ ]]
i = i + 2
end
else
table.insert(out, text:sub(i, i))
i = i + 1
end

if i > 1200 then break end
end

return table.concat(out)
end

-- 清理各种干扰元素
local function cleanFinal(text)
if not text or text == "" then return "" end

text = text
:gsub("{{[^{}]*}}", "")
:gsub("<ref.-</?ref[^>]*>", "")
:gsub("<!%-%-.-%-%->", "")
:gsub("\n==+[^=]+==+", " ")
:gsub("\n%s*[*#:;]+", " ")
:gsub("%s*\n%s*", " ")
:gsub("%s+", " ")

return mw.text.trim(text)
end

function p.getSummaryAndImage(frame)
local pageName = mw.text.trim(frame.args[1] or "")
if pageName == "" then return "" end

local title = mw.title.new(pageName)
if not title or not title.exists then return "" end

local content = title:getContent()
if not content or content == "" then return "" end

-- 取前400字符，提高抓到图片的概率（中文页面首图常在前部）
local rawExcerpt = mw.ustring.sub(content, 1, 400) or ""

-- 提取第一张图片（多写法分开匹配 + 清理）
local firstImage =
mw.ustring.match(rawExcerpt, "%[%[File:([^|%]]+)") or
mw.ustring.match(rawExcerpt, "%[%[文件:([^|%]]+)") or
mw.ustring.match(rawExcerpt, "%[%[Image:([^|%]]+)") or
mw.ustring.match(rawExcerpt, "%[%[图像:([^|%]]+)")

-- 备用：抓完整 [[...]] 并验证命名空间
if not firstImage then
local candidate = mw.ustring.match(rawExcerpt, "%[%[([^%]]+)%]%]")
if candidate then
local ns, fname = mw.ustring.match(candidate, "^([^:]+):(.+)$")
if ns and (ns == "File" or ns == "文件" or ns == "Image" or ns == "图像") then
firstImage = ns .. ":" .. fname
end
end
end

-- 清理文件名（去除隐藏字符、空格等）
if firstImage then
firstImage = mw.text.trim(firstImage)
firstImage = mw.ustring.gsub(firstImage, "^[%s%z]+", "")
firstImage = mw.ustring.gsub(firstImage, "[%s%z]+$", "")
end

local step1 = stripImages(rawExcerpt) or ""
local cleanText = cleanFinal(step1) or ""

local targetLen = 120
local summary = cleanText:sub(1, targetLen * 2) or ""

local ulen = mw.ustring.len(summary) or 0
if ulen > targetLen then
summary = mw.ustring.sub(summary, 1, targetLen) or ""

local rev = summary:reverse()
local lastBreak = rev:find("[ %s%p，。！？；：,…]") or 1
if lastBreak > 1 and lastBreak < 20 then
summary = mw.ustring.sub(summary, 1, #summary - lastBreak + 1) or ""
end

summary = mw.text.trim(summary) .. "..."
end

summary = mw.text.trim(summary)

-- HTML 构建
local container = mw.html.create('div')
:css('margin-bottom', '25px')
:css('display', 'flow-root')

container:tag('div')
:css('font-size', '1.3em')
:css('font-weight', 'bold')
:css('margin-bottom', '6px')
:css('border-bottom', '1px solid #eee')
:wikitext('[[' .. pageName .. ']]')

local textDiv = container:tag('div')
:css('line-height', '1.6')
:css('color', '#222')
:css('font-size', '14px')

if firstImage and firstImage ~= "" then
local imgWikitext = '[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]'
textDiv:wikitext(imgWikitext)
end

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T16:01:28Z

240D:C010:18:4:0:0:0:95：

local p = {}

local function stripImages(text)
if not text or text == "" then return "" end

text = text:gsub("%[%[(文件|Image|图像):", "[[File:")
:gsub("%[%[(分类|Category):", "[[Category:")

local out = {}
local i = 1
local len = #text

while i <= len do
if text:byte(i) == 91 and text:byte(i+1) == 91 then
local chunk = text:sub(i, i+15)
if chunk:match("^%[%[File:") or chunk:match("^%[%[Category:") then
local depth = 1
local j = i + 2
while j <= len and depth > 0 do
if text:byte(j) == 91 and text:byte(j+1) == 91 then
depth = depth + 1
j = j + 2
elseif text:byte(j) == 93 and text:byte(j+1) == 93 then
depth = depth - 1
j = j + 2
else
j = j + 1
end
end
i = j
else
table.insert(out, text:sub(i, i+1))
i = i + 2
end
else
table.insert(out, text:sub(i, i))
i = i + 1
end

if i > 1200 then break end
end

return table.concat(out)
end

local function cleanFinal(text)
if not text or text == "" then return "" end

text = text
:gsub("{{[^{}]*}}", "")
:gsub("<ref.-</?ref[^>]*>", "")
:gsub("<!%-%-.-%-%->", "")
:gsub("\n==+[^=]+==+", " ")
:gsub("\n%s*[*#:;]+", " ")
:gsub("%s*\n%s*", " ")
:gsub("%s+", " ")

return mw.text.trim(text)
end

function p.getSummaryAndImage(frame)
local pageName = mw.text.trim(frame.args[1] or "")
if pageName == "" then return "" end

local title = mw.title.new(pageName)
if not title or not title.exists then return "" end

local content = title:getContent()
if not content or content == "" then return "" end

local rawExcerpt = mw.ustring.sub(content, 1, 400) or "" -- 加大到400字符，提高抓图概率

-- 改进：优先从 title 对象取页面关联图片（如果有代表图）
local firstImage = title:getImage() -- MediaWiki 1.35+ 支持，优先用这个

if not firstImage or firstImage == "" then
-- 后备：从文本匹配（分开匹配，避免捕获组中文问题）
firstImage = mw.ustring.match(rawExcerpt, "%[%[File:([^|%]]+)")
or mw.ustring.match(rawExcerpt, "%[%[文件:([^|%]]+)")
or mw.ustring.match(rawExcerpt, "%[%[Image:([^|%]]+)")
or mw.ustring.match(rawExcerpt, "%[%[图像:([^|%]]+)")

if not firstImage then
local full = mw.ustring.match(rawExcerpt, "%[%[([^%]]+)%]%]")
if full then
local ns, name = mw.ustring.match(full, "^(%a+):(.+)$")
if ns and (ns:lower() == "file" or ns == "文件" or ns:lower() == "image" or ns == "图像") then
firstImage = ns .. ":" .. name
end
end
end
end

if firstImage then
firstImage = mw.text.trim(firstImage)
firstImage = mw.ustring.gsub(firstImage, "[%z%s]+", "") -- 清理隐藏字符
end

local step1 = stripImages(rawExcerpt) or ""
local cleanText = cleanFinal(step1) or ""

local targetLen = 120
local summary = cleanText:sub(1, targetLen * 2) or ""

local ulen = mw.ustring.len(summary) or 0
if ulen > targetLen then
summary = mw.ustring.sub(summary, 1, targetLen) or ""

local rev = summary:reverse()
local lastBreak = rev:find("[ %s%p，。！？；：,…]") or 1
if lastBreak > 1 and lastBreak < 20 then
summary = mw.ustring.sub(summary, 1, #summary - lastBreak + 1) or ""
end

summary = mw.text.trim(summary) .. "..."
end

summary = mw.text.trim(summary)

-- 输出容器
local container = mw.html.create('div')
:css('margin-bottom', '25px')
:css('display', 'flow-root')

container:tag('div')
:css('font-size', '1.3em')
:css('font-weight', 'bold')
:css('margin-bottom', '6px')
:css('border-bottom', '1px solid #eee')
:wikitext('[[' .. pageName .. ']]')

local textDiv = container:tag('div')
:css('line-height', '1.6')
:css('color', '#222')
:css('font-size', '14px')

-- 关键：如果有图，输出完整 wikitext；否则不输出图片部分
if firstImage and firstImage ~= "" then
-- 限制大小 + right 浮动 + link 到页面
local imgWikitext = '[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]'
textDiv:wikitext(imgWikitext)
end

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T15:58:17Z

240D:C010:18:4:0:0:0:95：

local p = {}

-- 剥离 [[File:]] [[文件:]] [[Category:]] 等，防止出现在摘要文本中
local function stripImages(text)
if not text or text == "" then return "" end

-- 统一常见前缀写法，方便后续匹配
text = text:gsub("%[%[(文件|Image|图像):", "[[File:")
:gsub("%[%[(分类|Category):", "[[Category:")

local out = {}
local i = 1
local len = #text

while i <= len do
if text:byte(i) == 91 and text:byte(i+1) == 91 then -- [[
local chunk = text:sub(i, i+15)
if chunk:match("^%[%[File:") or chunk:match("^%[%[Category:") then
local depth = 1
local j = i + 2
while j <= len and depth > 0 do
if text:byte(j) == 91 and text:byte(j+1) == 91 then
depth = depth + 1
j = j + 2
elseif text:byte(j) == 93 and text:byte(j+1) == 93 then
depth = depth - 1
j = j + 2
else
j = j + 1
end
end
i = j
else
table.insert(out, "[[")
i = i + 2
end
else
table.insert(out, text:sub(i, i))
i = i + 1
end

if i > 1200 then break end
end

return table.concat(out)
end

-- 清理模板、引用、注释、标题、列表等
local function cleanFinal(text)
if not text or text == "" then return "" end

text = text
:gsub("{{[^{}]*}}", "")
:gsub("<ref.-</?ref[^>]*>", "")
:gsub("<!%-%-.-%-%->", "")
:gsub("\n==+[^=]+==+", " ")
:gsub("\n%s*[*#:;]+", " ")
:gsub("%s*\n%s*", " ")
:gsub("%s+", " ")

return mw.text.trim(text)
end

function p.getSummaryAndImage(frame)
local pageName = mw.text.trim(frame.args[1] or "")
if pageName == "" then return "" end

local title = mw.title.new(pageName)
if not title or not title.exists then return "" end

local content = title:getContent()
if not content or content == "" then return "" end

local rawExcerpt = mw.ustring.sub(content, 1, 300) or ""

-- 提取第一张图片 - 分开匹配各种常见写法，减少编码问题
local firstImage =
mw.ustring.match(rawExcerpt, "%[%[File:([^|%]]+)") or
mw.ustring.match(rawExcerpt, "%[%[文件:([^|%]]+)") or
mw.ustring.match(rawExcerpt, "%[%[Image:([^|%]]+)") or
mw.ustring.match(rawExcerpt, "%[%[图像:([^|%]]+)")

-- 备用方案：抓完整链接再校验命名空间
if not firstImage then
local full = mw.ustring.match(rawExcerpt, "%[%[([^%]]+)%]%]")
if full then
local ns, name = mw.ustring.match(full, "^([^:]+):(.+)$")
if ns and (ns == "File" or ns == "文件" or ns == "Image" or ns == "图像") then
firstImage = ns .. ":" .. name
end
end
end

-- 额外清理文件名（去除隐藏字符、两端空格等）
if firstImage then
firstImage = mw.text.trim(firstImage or "")
firstImage = mw.ustring.gsub(firstImage, "^[%z%s]+", "")
firstImage = mw.ustring.gsub(firstImage, "[%z%s]+$", "")
end

local step1 = stripImages(rawExcerpt) or ""
local cleanText = cleanFinal(step1) or ""

local targetLen = 120
local summary = cleanText:sub(1, targetLen * 2) or ""

local ulen = mw.ustring.len(summary) or 0
if ulen > targetLen then
summary = mw.ustring.sub(summary, 1, targetLen) or ""

local rev = summary:reverse()
local lastBreak = rev:find("[ %s%p，。！？；：,…]") or 1
if lastBreak > 1 and lastBreak < 20 then
summary = mw.ustring.sub(summary, 1, #summary - lastBreak + 1) or ""
end

summary = mw.text.trim(summary) .. "..."
end

summary = mw.text.trim(summary)

-- HTML 输出
local container = mw.html.create('div')
:css('margin-bottom', '25px')
:css('display', 'flow-root')

container:tag('div')
:css('font-size', '1.3em')
:css('font-weight', 'bold')
:css('margin-bottom', '6px')
:css('border-bottom', '1px solid #eee')
:wikitext('[[' .. pageName .. ']]')

local textDiv = container:tag('div')
:css('line-height', '1.6')
:css('color', '#222')
:css('font-size', '14px')

if firstImage and firstImage ~= "" then
local imgWikitext = '[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]'
textDiv:wikitext(imgWikitext)
end

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T15:54:56Z

240D:C010:18:4:0:0:0:95：

local p = {}

-- ────────────────────────────────────────────────
-- 剥离图片、分类等 [[File:…]] [[Category:…]] 结构，避免出现在摘要中
-- ────────────────────────────────────────────────
local function stripImages(text)
if not text or text == "" then return "" end

-- 统一常见命名空间写法，减少匹配分支
text = text:gsub("%[%[(文件|File|Image|图像):", "[[File:")
:gsub("%[%[(分类|Category):", "[[Category:")

local out = {}
local i = 1
local len = #text

while i <= len do
if text:byte(i) == 91 and text:byte(i+1) == 91 then -- [[
local chunk = text:sub(i, i+15)
if chunk:match("^%[%[File:") or chunk:match("^%[%[Category:") then
-- 跳过整段嵌套链接 [[ ... ]]
local depth = 1
local j = i + 2
while j <= len and depth > 0 do
if text:byte(j) == 91 and text:byte(j+1) == 91 then
depth = depth + 1
j = j + 2
elseif text:byte(j) == 93 and text:byte(j+1) == 93 then
depth = depth - 1
j = j + 2
else
j = j + 1
end
end
i = j -- 跳到 ]] 之后或文本末尾
else
-- 普通内部链接，保留
table.insert(out, "[[")
i = i + 2
end
else
table.insert(out, text:sub(i,i))
i = i + 1
end

if i > 1200 then break end -- 极端安全阀
end

return table.concat(out)
end

-- ────────────────────────────────────────────────
-- 清理模板、参考文献、标题、列表等，得到纯文本摘要
-- ────────────────────────────────────────────────
local function cleanFinal(text)
if not text or text == "" then return "" end

text = text
:gsub("{{[^{}]*}}", "") -- 模板（简单版，够用）
:gsub("<ref.-</?ref[^>]*>", "") -- 参考文献
:gsub("<!%-%-.-%-%->", "") -- HTML 注释
:gsub("\n==+[^=]+==+", " ") -- 各级标题
:gsub("\n%s*[*#:;]+", " ") -- 列表、定义列表
:gsub("%s*\n%s*", " ") -- 所有换行 → 单个空格
:gsub("%s+", " ") -- 压缩连续空格

return mw.text.trim(text)
end

-- ────────────────────────────────────────────────
-- 主函数：生成带首图的简短摘要 + 标题
-- ────────────────────────────────────────────────
function p.getSummaryAndImage(frame)
local pageName = mw.text.trim(frame.args[1] or "")
if pageName == "" then return "" end

local title = mw.title.new(pageName)
if not title or not title.exists then return "" end

local content = title:getContent()
if not content or content == "" then return "" end

-- 安全取前 300 个 unicode 字符（避免字节截断中文）
local rawExcerpt = mw.ustring.sub(content, 1, 300)

-- 提取第一张图片（优先标准写法）
local firstImage = mw.ustring.match(rawExcerpt, "%[%[(File|文件|Image|图像):([^|%]]+)")

-- 备选：更宽松匹配，但只接受文件类
if not firstImage then
local candidate = mw.ustring.match(rawExcerpt, "%[%[([^|%]]+)%]")
if candidate then
local ns = candidate:match("^([^:]+):")
if ns and (ns:lower() == "file" or ns == "文件" or ns:lower() == "image" or ns == "图像") then
firstImage = candidate
end
end
end

local step1 = stripImages(rawExcerpt) or ""
local cleanText = cleanFinal(step1) or ""

local targetLen = 120
local summary = cleanText:sub(1, targetLen * 2) or ""

local ulen = mw.ustring.len(summary) or 0
if ulen > targetLen then
summary = mw.ustring.sub(summary, 1, targetLen) or ""

-- 尽量避免截在中文单词/句子中间
local rev = summary:reverse()
local lastBreak = rev:find("[ %s%p，。！？；：,…]") or 1
if lastBreak > 1 and lastBreak < 20 then
summary = mw.ustring.sub(summary, 1, #summary - lastBreak + 1) or ""
end

summary = mw.text.trim(summary) .. "..."
end

summary = mw.text.trim(summary)

-- ── HTML 输出 ───────────────────────────────────
local container = mw.html.create('div')
:css('margin-bottom', '25px')
:css('display', 'flow-root')

-- 标题
container:tag('div')
:css('font-size', '1.3em')
:css('font-weight', 'bold')
:css('margin-bottom', '6px')
:css('border-bottom', '1px solid #eee')
:wikitext('[[' .. pageName .. ']]')

-- 内容区（右浮图 + 文字）
local textDiv = container:tag('div')
:css('line-height', '1.6')
:css('color', '#222')
:css('font-size', '14px')

if firstImage and firstImage ~= "" then
firstImage = mw.text.trim(firstImage)
local img = '[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]'
textDiv:wikitext(img)
end

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T15:53:07Z

240D:C010:18:4:0:0:0:95：

local p = {}

-- stripImages 函数保持不变（已使用字节级，但对图片提取影响小）

-- cleanFinal 函数保持不变

function p.getSummaryAndImage(frame)
local pageName = mw.text.trim(frame.args[1] or "")
if pageName == "" then return "" end

local title = mw.title.new(pageName)
if not title or not title.exists then return "" end

local content = title:getContent()
if not content or content == "" then return "" end

-- 用 mw.ustring 安全截取（避免字节切割中文）
local rawExcerpt = mw.ustring.sub(content, 1, 300) -- 字符级 300 个字符，够用

-- 改用 mw.ustring 的模式匹配，更安全处理中文文件名
local firstImage
-- 尝试匹配 File: 或文件: 开头的
firstImage = mw.ustring.match(rawExcerpt, "%[%[(File|文件|Image|图像):([^|%]]+)")

-- 如果没匹配到，再试更宽松的（但优先上面）
if not firstImage then
firstImage = mw.ustring.match(rawExcerpt, "%[%[([^|%]]+)%]")
if firstImage and not (firstImage:find("File:") or firstImage:find("文件:")
or firstImage:find("Image:") or firstImage:find("图像:")) then
firstImage = nil -- 不是文件就丢掉
end
end

local step1 = stripImages(rawExcerpt) or ""
local cleanText = cleanFinal(step1) or ""

local targetLen = 120
local summary = cleanText:sub(1, targetLen * 2) or ""

local ulen = mw.ustring.len(summary) or 0
if ulen > targetLen then
summary = mw.ustring.sub(summary, 1, targetLen) or ""

local rev = summary:reverse()
local lastBreak = rev:find("[ %s%p，。！？；：]") or 1
if lastBreak > 1 and lastBreak < 20 then
summary = mw.ustring.sub(summary, 1, #summary - lastBreak + 1) or ""
end

summary = mw.text.trim(summary) .. "..."
end

summary = mw.text.trim(summary)

-- HTML 构建
local container = mw.html.create('div')
:css('margin-bottom', '25px')
:css('display', 'flow-root')

container:tag('div')
:css('font-size', '1.3em')
:css('font-weight', 'bold')
:css('margin-bottom', '6px')
:css('border-bottom', '1px solid #eee')
:wikitext('[[' .. pageName .. ']]')

local textDiv = container:tag('div')
:css('line-height', '1.6')
:css('color', '#222')
:css('font-size', '14px')

-- 关键修复：如果有图片名，确保用 mw.ustring 处理
if firstImage and firstImage ~= "" then
-- 额外清理：去掉可能的前后空格（虽然 match 已捕获干净）
firstImage = mw.text.trim(firstImage)

-- 用 [[File:xxx|...]] 格式，确保 MediaWiki 正确解析
local imgWikitext = '[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]'
textDiv:wikitext(imgWikitext)
end

textDiv:wikitext(summary)

return tostring(container)
end

return p