Module:PageSummary:修订间差异
武外梗百科 爱国好学自强图新的百科全书
更多操作
无编辑摘要 标签:手工回退 |
无编辑摘要 |
||
| 第3行: | 第3行: | ||
local function safeStrip(text) | local function safeStrip(text) | ||
if not text then return "" end | if not text then return "" end | ||
-- 1. 移除引用、注释和 HTML 标签 | |||
text = string.gsub(text, "<ref.-</ref>", "") | text = string.gsub(text, "<ref.-</ref>", "") | ||
text = string.gsub(text, "<ref.->", "") | text = string.gsub(text, "<ref.->", "") | ||
text = string.gsub(text, "<!%-%-.-%-%->", "") | text = string.gsub(text, "<!%-%-.-%-%->", "") | ||
-- 2. 核心修复:粉碎带链接的图片描述 | |||
-- 原理:先匹配 [[文件:...[[...]]...]] 这种结构,防止正则在第一个 ]] 处截断 | |||
local imagePrefixes = {"[Ff]ile:", "[Ii]mage:", "文件:", "图像:"} | local imagePrefixes = {"[Ff]ile:", "[Ii]mage:", "文件:", "图像:"} | ||
for _, prefix in ipairs(imagePrefixes) do | for _, prefix in ipairs(imagePrefixes) do | ||
-- 先处理最复杂的:带有一层嵌套链接的图片 | |||
text = string.gsub(text, "%[%[" .. prefix .. "[^%[%]]+ %[%[[^%[%]]+%]%] [^%[%]]+ %]%]", "") | |||
-- 再处理普通的图片 | |||
text = string.gsub(text, "%[%[" .. prefix .. ".-%]%]", "") | text = string.gsub(text, "%[%[" .. prefix .. ".-%]%]", "") | ||
end | end | ||
-- 3. 移除模板 (限制5层) | |||
for i = 1, 5 do | for i = 1, 5 do | ||
local newText = string.gsub(text, "{{[^{}]+}}", "") | local newText = string.gsub(text, "{{[^{}]+}}", "") | ||
| 第18行: | 第26行: | ||
end | end | ||
-- 4. 移除 Wiki 格式符号 (包括 ==标题== 等) | |||
text = string.gsub(text, "\n==+.-==+", " ") | text = string.gsub(text, "\n==+.-==+", " ") | ||
text = string.gsub(text, "\n%s*[*#:]+", " ") | text = string.gsub(text, "\n%s*[*#:]+", " ") | ||
| 第23行: | 第32行: | ||
text = string.gsub(text, "\n", " ") | text = string.gsub(text, "\n", " ") | ||
text = string.gsub(text, "%s+", " ") | text = string.gsub(text, "%s+", " ") | ||
return mw.text.trim(text) | return mw.text.trim(text) | ||
end | end | ||
| 第29行: | 第39行: | ||
local pageName = frame.args[1] or "" | local pageName = frame.args[1] or "" | ||
local title = mw.title.new(pageName) | local title = mw.title.new(pageName) | ||
if not title or not title.exists then return " | if not title or not title.exists then return "" end | ||
local content = title:getContent() | local content = title:getContent() | ||
-- 预取的内容长度,500足够 | |||
local rawExcerpt = mw.ustring.sub(content, 1, 500) | local rawExcerpt = mw.ustring.sub(content, 1, 500) | ||
-- 提取第一张图(用于卡片展示) | |||
local firstImage = string.match(rawExcerpt, "%[%[([Ff]ile:.-)[|%]]") or | local firstImage = string.match(rawExcerpt, "%[%[([Ff]ile:.-)[|%]]") or | ||
string.match(rawExcerpt, "%[%[([Ii]mage:.-)[|%]]") or | string.match(rawExcerpt, "%[%[([Ii]mage:.-)[|%]]") or | ||
string.match(rawExcerpt, "%[%[(文件:.-)[|%]]") | string.match(rawExcerpt, "%[%[(文件:.-)[|%]]") | ||
-- 使用改进后的 safeStrip | |||
local cleanText = safeStrip(rawExcerpt) | local cleanText = safeStrip(rawExcerpt) | ||
-- 截断 150 字 | |||
local targetLen = 150 | local targetLen = 150 | ||
local summary = mw.ustring.sub(cleanText, 1, targetLen) | local summary = mw.ustring.sub(cleanText, 1, targetLen) | ||
while mw.ustring.sub(summary, -1) == "[" do | -- 清除末尾可能存在的残缺符号 | ||
while mw.ustring.sub(summary, -1) == "[" or mw.ustring.sub(summary, -1) == "|" do | |||
summary = mw.ustring.sub(summary, 1, -2) | summary = mw.ustring.sub(summary, 1, -2) | ||
end | end | ||
-- 链接完整性检查与补全 | |||
local _, opens = string.gsub(summary, "%[%[", "") | local _, opens = string.gsub(summary, "%[%[", "") | ||
local _, closes = string.gsub(summary, "%]%]", "") | local _, closes = string.gsub(summary, "%]%]", "") | ||
| 第55行: | 第71行: | ||
summary = summary .. string.sub(rest, 1, endLink + 1) | summary = summary .. string.sub(rest, 1, endLink + 1) | ||
else | else | ||
-- 补全失败则反向截断,确保不产生红字或代码残留 | |||
local lastOpen = string.find(summary:reverse(), "%[%[") | local lastOpen = string.find(summary:reverse(), "%[%[") | ||
if lastOpen then summary = string.sub(summary, 1, #summary - lastOpen - 1) end | if lastOpen then | ||
summary = string.sub(summary, 1, #summary - lastOpen - 1) | |||
end | |||
end | end | ||
end | end | ||
summary = mw.text.trim(summary) | summary = mw.text.trim(summary) | ||
if mw.ustring.len(cleanText) > targetLen then summary = summary .. "..." end | if mw.ustring.len(cleanText) > targetLen then | ||
summary = summary .. "..." | |||
end | |||
-- 渲染:原生标题(无下划线),原生图片(无框) | |||
local container = mw.html.create('div'):css({['margin']='15px 0', ['display']='flow-root'}) | local container = mw.html.create('div'):css({['margin']='15px 0', ['display']='flow-root'}) | ||
if firstImage then | if firstImage then | ||
| 第68行: | 第90行: | ||
end | end | ||
container:tag('div') | container:tag('div') | ||
:css({['font-size']='1. | :css({['font-size']='1.3em', ['font-weight']='bold', ['margin-bottom']='6px'}) | ||
:wikitext('[[' .. pageName .. ']]') | :wikitext('[[' .. pageName .. ']]') | ||
container:tag('div') | container:tag('div') | ||
2025年12月21日 (日) 18:19的版本
此模块的文档可以在Module:PageSummary/doc创建
local p = {}
local function safeStrip(text)
if not text then return "" end
-- 1. 移除引用、注释和 HTML 标签
text = string.gsub(text, "<ref.-</ref>", "")
text = string.gsub(text, "<ref.->", "")
text = string.gsub(text, "<!%-%-.-%-%->", "")
-- 2. 核心修复:粉碎带链接的图片描述
-- 原理:先匹配 [[文件:...[[...]]...]] 这种结构,防止正则在第一个 ]] 处截断
local imagePrefixes = {"[Ff]ile:", "[Ii]mage:", "文件:", "图像:"}
for _, prefix in ipairs(imagePrefixes) do
-- 先处理最复杂的:带有一层嵌套链接的图片
text = string.gsub(text, "%[%[" .. prefix .. "[^%[%]]+ %[%[[^%[%]]+%]%] [^%[%]]+ %]%]", "")
-- 再处理普通的图片
text = string.gsub(text, "%[%[" .. prefix .. ".-%]%]", "")
end
-- 3. 移除模板 (限制5层)
for i = 1, 5 do
local newText = string.gsub(text, "{{[^{}]+}}", "")
if newText == text then break end
text = newText
end
-- 4. 移除 Wiki 格式符号 (包括 ==标题== 等)
text = string.gsub(text, "\n==+.-==+", " ")
text = string.gsub(text, "\n%s*[*#:]+", " ")
text = string.gsub(text, "&[Nn][Bb][Ss][Pp];", " ")
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()
-- 预取的内容长度,500足够
local rawExcerpt = mw.ustring.sub(content, 1, 500)
-- 提取第一张图(用于卡片展示)
local firstImage = string.match(rawExcerpt, "%[%[([Ff]ile:.-)[|%]]") or
string.match(rawExcerpt, "%[%[([Ii]mage:.-)[|%]]") or
string.match(rawExcerpt, "%[%[(文件:.-)[|%]]")
-- 使用改进后的 safeStrip
local cleanText = safeStrip(rawExcerpt)
-- 截断 150 字
local targetLen = 150
local summary = mw.ustring.sub(cleanText, 1, targetLen)
-- 清除末尾可能存在的残缺符号
while mw.ustring.sub(summary, -1) == "[" or mw.ustring.sub(summary, -1) == "|" do
summary = mw.ustring.sub(summary, 1, -2)
end
-- 链接完整性检查与补全
local _, opens = string.gsub(summary, "%[%[", "")
local _, closes = string.gsub(summary, "%]%]", "")
if opens > closes then
local rest = mw.ustring.sub(cleanText, targetLen + 1, targetLen + 100)
local endLink = string.find(rest, "]]", 1, true)
if endLink then
summary = summary .. string.sub(rest, 1, endLink + 1)
else
-- 补全失败则反向截断,确保不产生红字或代码残留
local lastOpen = string.find(summary:reverse(), "%[%[")
if lastOpen then
summary = string.sub(summary, 1, #summary - lastOpen - 1)
end
end
end
summary = mw.text.trim(summary)
if mw.ustring.len(cleanText) > targetLen then
summary = summary .. "..."
end
-- 渲染:原生标题(无下划线),原生图片(无框)
local container = mw.html.create('div'):css({['margin']='15px 0', ['display']='flow-root'})
if firstImage then
container:wikitext('[[' .. firstImage .. '|150px|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']='1.6', ['color']='#202122'})
:wikitext(summary)
return tostring(container)
end
return p