武外梗百科 - 用户贡献 [zh]

Module:PageSummary

2026-02-18T01:32:54Z

240D:C010:132:2:0:0:0:AA：

local p = {}

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

-- 模式全部重新写，避免复制隐藏字符
text = string.gsub(text, "%[%[(文件|Image|图像):", "[[File:")
text = string.gsub(text, "%[%[(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:find("^%[%[File:") or chunk:find("^%[%[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

-- 所有模式都重新手写，使用 string.gsub（模式是纯 ASCII）
text = string.gsub(text, "{{[^{}]*}}", "")
text = string.gsub(text, "<ref.-</?ref[^>]*>", "")
text = string.gsub(text, "<!%-%-.-%-%->", "")
text = string.gsub(text, "\n==+[^=]+==+", " ")
text = string.gsub(text, "\n%s*[*#:;]+", " ")
text = string.gsub(text, "%s*\n%s*", " ")
text = string.gsub(text, "%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

-- 简单清洗 content，避免传入 gsub 的文本有问题
content = string.gsub(content, "[\0-\31\127]", "") -- 移除控制字符
local rawExcerpt = mw.ustring.sub(content, 1, 800) or ""

-- 提取第一张图片（使用 find + 手动截取）
local firstImage = nil
local prefixes = {"File:", "文件:"}
for _, prefix in ipairs(prefixes) do
local pattern = "%[%[" .. prefix
local start = mw.ustring.find(rawExcerpt, pattern)
if start then
local pipePos = mw.ustring.find(rawExcerpt, "|", start)
local endPos = mw.ustring.find(rawExcerpt, "%]%]", start)
local boundary = pipePos or endPos
if boundary then
local candidate = mw.ustring.sub(rawExcerpt, start + #pattern + 1, boundary - 1)
candidate = mw.text.trim(candidate)
if candidate ~= "" and mw.ustring.find(candidate, "%.") then
firstImage = prefix .. candidate
break
end
end
end
end

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

local targetLen = 120
local summary = cleanText

local ulen = mw.ustring.len(summary) or 0
if ulen > targetLen then
local temp = mw.ustring.sub(summary, 1, targetLen + 20)

local breakPos = nil
for i = mw.ustring.len(temp), 1, -1 do
local char = mw.ustring.sub(temp, i, i)
if char:match("[ %s，。！？；：,…、]") then
breakPos = i
break
end
end

if breakPos and breakPos > targetLen - 30 then
summary = mw.ustring.sub(temp, 1, breakPos)
else
summary = mw.ustring.sub(temp, 1, targetLen)
end

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

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
textDiv:wikitext('[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]')
end

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T16:16:56Z

240D:C010:132:2:0:0:0:AA：

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, 800) or ""

-- 强力提取第一张图片（支持带参数的写法）
local firstImage = mw.ustring.match(rawExcerpt, "%[%[File:([^|]+)")
or mw.ustring.match(rawExcerpt, "%[%[文件:([^|]+)")

-- 验证图片是否有效
if firstImage then
firstImage = mw.text.trim(firstImage)
firstImage = mw.ustring.gsub(firstImage, "[%z%s<>\"]+", "")
if mw.ustring.len(firstImage) < 5 or not mw.ustring.find(firstImage, "[:%.]") then
firstImage = nil
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)

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')

-- 只有图片有效时才输出图片代码，彻底杜绝泄漏
if firstImage and firstImage ~= "" then
textDiv:wikitext('[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]')
end

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T15:51:29Z

240D:C010:132:2:0:0:0:AA：

local p = {}

-- 更高效的图片/分类/文件链接剥离（字节级处理为主）
local function stripImages(text)
if not text or text == "" then return "" end

-- 统一常见前缀，减少分支
text = text:gsub("%[%[(文件|File|Image|图像):", "[[File:")

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+12)
if chunk:match("^%[%[File:")
or chunk:match("^%[%[Category:")
or chunk:match("^%[%[分类:")
then
-- 跳过整段嵌套 [[...]]
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
else
table.insert(out, "[[")
i = i + 2
end
else
table.insert(out, text:sub(i, i))
i = i + 1
end

if i > 800 then break end -- 安全阀
end

return table.concat(out)
end

-- 合并清理步骤，减少重复扫描
local function cleanFinal(text)
if not text or text == "" then return "" end

text = text
:gsub("{{[^{}]*}}", "") -- 模板
:gsub("<ref.-</?ref[^>]*>", "") -- 参考文献
:gsub("<!%-%-.-%-%->", "") -- 注释
:gsub("\n==+.+==+", " ") -- 标题
:gsub("\n%s*[*#:;]+", " ") -- 列表
:gsub("%s*\n%s*", " ") -- 换行转空格
:gsub("%s+", " ") -- 压缩空格

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

-- 只取前450字节（足够首图+摘要）
local rawExcerpt = content:sub(1, 450)

-- 提取第一张图片（兼容多种写法）
local firstImage = rawExcerpt:match("%[%[(File|文件|Image|图像):([^|%]]+)")

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

local targetLen = 120
-- 提前多取一点，便于后续精确截断
local summary = cleanText:sub(1, targetLen * 2) or ""

-- 关键修复：判断前确保不是 nil
local ulen = mw.ustring.len(summary) or 0

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

-- 尝试避免截断在单词/中文中间（简单向后找空格或标点）
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 输出部分
local container = mw.html.create('div')
:css('margin-bottom', '25px')
:css('display', 'flow-root')

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

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

textDiv:wikitext(summary)

return tostring(container)
end

return p

Module:PageSummary

2026-02-17T15:50:14Z

240D:C010:132:2:0:0:0:AA：

local p = {}

-- 更高效的图片/链接/分类剥离（避免逐字符 + 频繁 sub）
local function stripImages(text)
if not text or text == "" then return "" end

-- 预先把常见命名空间前缀统一处理，减少后续匹配分支
text = text:gsub("%[%[(文件|File|Image|图像):", "[[File:")

local out = {}
local i = 1
local len = #text -- 优先用字节长度，性能更好（中文环境影响可接受）

while i <= len do
local byte = text:byte(i)
if byte == 91 and text:byte(i+1) == 91 then -- [[
local chunk = text:sub(i, i+12)
if chunk:match("^%[%[File:")
or chunk:match("^%[%[Category:")
or chunk:match("^%[%[分类:")
then
-- 跳过整段 [[...]]
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
-- 普通 [[ ，保留
table.insert(out, "[")
i = i + 1
end
else
table.insert(out, text:sub(i, i))
i = i + 1
end

-- 安全阀（防止极端情况卡死）
if i > 800 then break end
end

return table.concat(out)
end

-- 合并多次 gsub，提升正则效率
local function cleanFinal(text)
if not text or text == "" then return "" end

-- 一次性处理多种需要删除的模式
text = text
:gsub("{{[^{}]*}}", "") -- 模板（包括嵌套一层的情况已足够）
:gsub("<ref.-</?ref[^>]*>", "") -- 参考文献（更宽松匹配）
:gsub("<!%-%-.-%-%->", "") -- 注释
:gsub("\n==+.+==+", " ") -- 标题
:gsub("\n%s*[*#:;]+", " ") -- 列表
:gsub("%s*\n%s*", " ") -- 换行 → 空格（合并多行）
:gsub("%s+", " ") -- 压缩连续空格

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 then return "" end

-- 只取前450字节（中文约200-300字），足够提取首图 + 摘要
local rawExcerpt = content:sub(1, 450)

-- 提取第一张图片（更宽松匹配，兼容更多写法）
local firstImage = rawExcerpt:match("%[%[(File|文件|Image|图像):([^|%]]+)")

-- 核心剥离 + 清理
local step1 = stripImages(rawExcerpt)
local cleanText = cleanFinal(step1)

local targetLen = 120
local summary = cleanText:sub(1, targetLen * 2) -- 多取一点，后面精确截断

-- 更安全的截断（避免截断在 [[ 中间）
local len = mw.ustring.len(summary)
if len > targetLen then
summary = mw.ustring.sub(summary, 1, targetLen)

-- 简单向后找最近的空格或标点，避免断字
local lastSpace = summary:reverse():find("[ %s%p]") or 1
if lastSpace > 1 and lastSpace < 15 then
summary = mw.ustring.sub(summary, 1, #summary - lastSpace + 1)
end
summary = mw.text.trim(summary) .. "..."
end

-- HTML 构建部分基本保持原样（性能占比很低）
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
local textDiv = container:tag('div')
:css('line-height', '1.6')
:css('color', '#222')
:css('font-size', '14px')

-- 图片右浮 + 链接到页面
if firstImage then
textDiv:wikitext(
'[[' .. firstImage .. '|120px|right|link=' .. pageName .. ']]'
)
end

textDiv:wikitext(summary)

return tostring(container)
end

return p