跳转到内容

Module:Furigana：修订间差异

以互动方式浏览历史
下一编辑→
创建页面，内容为“local p = {} local getArgs = require('module:arguments').getArgs function main(args) local rb = args[1] local rt = args[2] local rt2 = args[3] local newrb = '' local newrt = '' if rt then local lrb = mw.ustring.len(rb) local lrt = mw.ustring.len(rt) local diff = lrt * 0.75 - lrb local space = '' local important = rt2 and '!important' or '' if diff > 0 then space = tostring(math.floor(diff / (lrb + 1) * 1000) / 1000)..'em' newrb = '<span…”
 
无编辑摘要
第1行： 第1行：
local p = {}
local p = {}
local getArgs = require('module:arguments').getArgs


function main(args)
-- 转化为upvalue，提升速度
local rb = args[1]
local type = type
local rt = args[2]
local ipairs = ipairs
local rt2 = args[3]
local concat = table.concat
local newrb = ''
local u_gsub = mw.ustring.gsub
local newrt = ''
local u_sub = mw.ustring.sub
if rt then
 
local lrb = mw.ustring.len(rb)
local KANJI_LIKE_UNICODE_RANGES = {
local lrt = mw.ustring.len(rt)
{ 0x2E80,  0x2EFF },  -- CJK部首补充
local diff = lrt * 0.75 - lrb
{ 0x3005,  0x3007 },  -- "々、〆、〇"
local space = ''
{ 0x31C0,  0x31EF },  -- CJK笔画
local important = rt2 and '!important' or ''
{ 0x3400,  0x4DBF },  -- CJK统一表意文字扩展A
if diff > 0 then
{ 0x4E00,  0x9FFF },  -- CJK统一表意文字
space = tostring(math.floor(diff / (lrb + 1) * 1000) / 1000)..'em'
{ 0xF900,  0xFAFF },  -- CJK兼容表意文字
newrb = '<span style="margin-left:'..space..important..';letter-spacing:'..space..important..'">'..rb..'</span>'
{ 0x20000, 0x2A6DF }, -- CJK统一表意文字扩展B
newrt = rt
{ 0x2A700, 0x2EE5F }, -- CJK统一表意文字扩展C-I
{ 0x2F800, 0x2FA1F }, -- CJK兼容表意文字补充
{ 0x30000, 0x323AF }, -- CJK统一表意文字扩展G-H
}
local KANJI_LIKE_PATTERN = (function ()
local char = mw.ustring.char
local parts = {}
for i, range in ipairs(KANJI_LIKE_UNICODE_RANGES) do
parts[i] = type(range) == 'table' and char(range[1])..'-'..char(range[2]) or char(range)
end
return '['..concat(parts)..']'
end)()
 
 
--[[
把<tag>、</tag>与其他文本分开。
```lua
divide_tags_and_non_tags('A<span title="（title）">B</span>C')
-->
{
{ type = 'unknown', content = 'A' },
{ type = 'tag', content = '<span title="（title）">' },
{ type = 'unknown', content = 'B' },
{ type = 'tag', content = '</span>' },
{ type = 'unknown', content = 'C' },
```
]]
---@param text string
---@return {type: 'tag' | 'unkown', content: string}[]
local function extract_tags(text)
local result = {}
local rest = text
:gsub(
'(.-)(</?[:%a_][:%w_-.]*[^<>]*>)',
function (non_tag, tag)
if non_tag ~= '' then
result[#result+1] = { type = 'unknown', content = non_tag }
end
result[#result+1] = { type = 'tag', content = tag }
return ''
end
)
if rest ~= '' then
result[#result+1] = { type = 'unknown', content = rest }
end
return result
end
 
local EXTRACT_RB_RT_PATTERN = '(.-)('..KANJI_LIKE_PATTERN..'+)(%b（）)'
---@param text string
---@param into_tokens ({type: 'ruby-part' | 'tag' | 'text', content: string} | {type: 'rb-rt', rb: string, rt: string})[]
local function extract_rb_rt(text, into_tokens)
for line_with_lf in text:gmatch('[^\n]*\n*') do
local rest = u_gsub(
line_with_lf,
EXTRACT_RB_RT_PATTERN,
function (before, rb, rt)
if before ~= '' then
into_tokens[#into_tokens+1] = { type = 'text', content = before }
end
into_tokens[#into_tokens+1] = {
type = 'rb-rt',
rb = rb,
rt = u_sub(rt, 2, -2),  -- 去除括号
}
return ''
end
)
if rest ~= '' then
into_tokens[#into_tokens+1] = { type = 'text', content = rest }
end
end
end
 
---细节见测试 [[Module:Furigana/tests]]。
---@param tokens {type: 'tag' | 'unkown', content: string}[]
---@param into_tokens ({type: 'ruby-part' | 'tag' | 'text', content: string} | {type: 'rb-rt', rb: string, rt: string})[]
local function extract_others(tokens, into_tokens)
local ruby_start_index = nil
 
for i, token in ipairs(tokens) do
if ruby_start_index then  -- 如果处于<ruby>标签内，则所有token都是ruby-part
into_tokens[#into_tokens+1] = { type = 'ruby-part', content = token.content }
if token.type == 'tag' and token.content:match('^</[Rr][Uu][Bb][Yy]') then
ruby_start_index = nil
end
elseif token.type == 'tag' then
if token.content:match('^<[Rr][Uu][Bb][Yy]') then
into_tokens[#into_tokens+1] = { type = 'ruby-part', content = token.content }
ruby_start_index = i
else
---@diagnostic disable-next-line: assign-type-mismatch
into_tokens[#into_tokens+1] = token
end
else
else
space = tostring(math.floor(-diff / (lrt + 1) / 0.75 * 1000) / 1000)..'em'
-- assert(token.type == 'unknown')
newrb = rb
extract_rb_rt(token.content, into_tokens)
newrt = '<span style="margin-left:'..space..';letter-spacing:'..space..'">'..rt..'</span>'
end
end
else
newrb = rb
newrt = '<span class="photrans-s">'..rb..'</span>'
end
end
return newrb, newrt, rt2
end
end


function p.one(frame)
---将`texts`解析为tokens，细节见测试[[Module:Furigana/tests]]。
local args = getArgs(frame)
---@param texts string[]
local rb = args[1] or ''
---@return ({type: 'ruby-part' | 'tag' | 'text', content: string} | {type: 'rb-rt', rb: string, rt: string})[]
local rt = args[2] or '&nbsp;'
local function parse(texts)
return '<ruby><rb>'..rb..'</rb><rt'..(args[2] and '>' or ' class="photrans-s">')..
local tokens = {}
(args.on and rt or '<span>'..rt..'</span>')..'</rt></ruby>'
for _, text in ipairs(texts) do
extract_others(extract_tags(text), tokens)
end
return tokens
end
end
p.parse = parse


function p.multi(frame)
---用`render_ruby`渲染`tokens`中的`rb-rt` token，其余token不变，最终连接为一个字符串。
local args = getArgs(frame)
---@param tokens ({type: 'ruby-part' | 'tag' | 'text', content: string} | {type: 'rb-rt', rb: string, rt: string})[]
local fullrb = ''
---@param render_ruby fun(rb: string, rt: string): string
local fullrt = ''
---@return string
local flag = false
local function render(tokens, render_ruby)
local fullrt2 = '<span class="photrans-off">'
local result = {}
for k, v in ipairs(args) do
for i, token in ipairs(tokens) do
local eachArg = mw.text.split(v, '_')
result[i] = token.type == 'rb-rt' and render_ruby(token.rb, token.rt) or token.content
if args.on then
end
eachArg[3] = true
return concat(result)
end
p.render = render
 
---将`texts`中用“汉字（注音）”表示的振假名替换为`render_ruby`的返回值。
---@param texts string | string[]
---@param render_ruby fun(rb: string, rt: string): string
---@return string
function p.process(texts, render_ruby)
if type(texts) == 'string' then
texts = { texts }
end
local tokens = parse(texts)
return render(tokens, render_ruby)
end
 
function p.main(frame, args)
if not args then
local parent = frame:getParent()
local templates = { ['Template:Furigana'] = true, ['Template:Sandbox'] = true }
if parent and templates[parent:getTitle()] then
frame = parent
end
args = frame.args
end
local template = args.template or 'Ruby'
 
if mw.isSubsting() then
local prefix = '{{'..template..'|'
return p.process(args, function (rb, rt)
return prefix..rb..'|'..rt..'}}'
end)
end
 
local is_compatible_mode = args.compatible and args.compatible ~= ''
if not is_compatible_mode then
-- 快速模式，通过字符串替换，避免多次调用frame:expandTemplate
local placeholder_pattern = '!@$r[bt]$@!'
local rb_placeholder = '!@$rb$@!'
local rt_placeholder = '!@$rt$@!'
local template_string = frame:expandTemplate {
title = template,
args = { rb_placeholder, rt_placeholder },
}
if template_string:find(placeholder_pattern) then
return p.process(args, function (rb, rt)
return template_string:gsub(placeholder_pattern, { [rb_placeholder] = rb, [rt_placeholder] = rt })
end)
end
end
local newrb, newrt, rt2 = main(eachArg)
fullrb = fullrb..newrb
fullrt = fullrt..newrt
flag = flag or rt2
fullrt2 = fullrt2..(rt2 and newrt or '<span class="photrans-s">'..eachArg[1]..'</span>')
end
end
fullrt2 = fullrt2..'</span>'
 
return '<ruby><rb>'..fullrb..'</rb><rt><span>'..fullrt..'</span>'..(flag and fullrt2 or '')..'</rt></ruby>'
return p.process(args, function (rb, rt)
return frame:expandTemplate {
title = template,
args = { rb, rt },
}
end)
end
end


return p
return p
检索自“https://wikimas.org/wiki/Module:Furigana”