function explode(delim, str, limit)
local result = {}
local m = {}
local offset = 1
local count = 1
while (limit == nil or count < limit) and match_with_offset(delim, str, m, offset) do
table.insert(result, string.sub(str, offset, m[1]-1))
offset = m[1] + string.len(delim)
count = count + 1
end
table.insert(result, string.sub(str, offset))
return result
end
function trim(str)
return string.match(str, '^ *(.-) *$')
end
function rtrim(str)
return string.match(str, '^(.-) *$')
end
function strspn(str, charlist)
return string.len(string.match(str, '^[' .. charlist .. ']+') or '')
end
function match_with_offset(pattern, str, match, offset)
offset = offset or 1
match[0] = string.match(str, pattern, offset)
match[1] = string.find(str, pattern, offset)
if match[0] == nil then
return false
else
return true
end
end
-- converted from Parser.php with a few adjustments
function parseTables(text)
local lines = explode('\n', text)
local out = ''
local td_history = {};
local last_tag_history = {};
local tr_history = {};
local tr_attributes = {};
local has_opened_tr = {};
local indent_level = 0;
for i, outLine in ipairs(lines) do
line = trim(outLine)
if line == '' then
out = out .. outLine .. '\n'
else--CONTINUE
local first_character = string.sub(line, 1, 1)
local first_two = string.sub(line, 1, 2)
local matches = {}
matches[1], matches[3], matches[2] = string.match(line, '^(:*)%s*({|)(.*)$')
if matches[3] ~= nil then
indent_level = string.len(matches[1] or '')
local attributes = matches[2] or ''
-- unstripBoth & fixTagAttributes
outLine = string.rep('<dl><dd>', indent_level) .. '<table ' .. attributes .. '>'
table.insert(td_history, false)
table.insert(last_tag_history, '')
table.insert(tr_history, false)
table.insert(tr_attributes, '')
table.insert(has_opened_tr, false)
elseif #td_history == 0 then
elseif first_two == '|}' then
line = '</table>' .. string.sub(line, 3)
local last_tag = table.remove(last_tag_history)
if not table.remove(has_opened_tr) then
line = '<tr><td></td></tr>' .. line
end
if table.remove(tr_history) then
line = '</tr>' .. line
end
if table.remove(td_history) then
line = '</' .. last_tag .. '>' .. line
end
table.remove(tr_attributes)
if indent_level > 0 then
outLine = rtrim(line) .. string.rep('</dd></dl>', indent_level)
else
outLine = line
end
elseif first_two == '|-' then
line = string.gsub(line, '^|%-+', '')
local attributes = line
-- unstripBoth & fixTagAttributes
table.remove(tr_attributes)
table.insert(tr_attributes, attributes)
line = ''
local last_tag = table.remove(last_tag_history)
table.remove(has_opened_tr)
table.insert(has_opened_tr, true)
if table.remove(tr_history) then
line = '</tr>'
end
if table.remove(td_history) then
line = '</' .. last_tag .. '>' .. line
end
outLine = line
table.insert(tr_history, false)
table.insert(td_history, false)
table.insert(last_tag_history, '')
elseif first_character == '|'
or first_character == '!'
or first_two == '|+' then
if first_two == '|+' then
first_character = '+'
line = string.sub(line, 3)
else
line = string.sub(line, 2)
end
if first_character == '!' then
-- replaceMarkup
line = string.gsub(line, '!!', '||')
end
local cells = explode('||', line)
outLine = ''
for i, cell in ipairs(cells) do
local previous = ''
if first_character ~= '+' then
local tr_after = table.remove(tr_attributes)
if not table.remove(tr_history) then
previous = '<tr ' .. tr_after .. '>'
end
table.insert(tr_history, true)
table.insert(tr_attributes, '')
table.remove(has_opened_tr)
table.insert(has_opened_tr, true)
end
local last_tag = table.remove(last_tag_history)
if table.remove(td_history) then
previous = '</' .. last_tag .. '>\n' .. previous
end
if first_character == '|' then
last_tag = 'td'
elseif first_character == '!' then
last_tag = 'th'
elseif first_character == '+' then
last_tag = 'caption'
else
last_tag = ''
end
table.insert(last_tag_history, last_tag)
local cell_data_iter = explode('|', cell, 2)
local cell_data = {}
for i, item in ipairs(cell_data_iter) do
table.insert(cell_data, item)
end
if string.match('[[', cell_data[1]) or string.match('%-{', cell_data[1]) then
cell = previous .. '<' .. last_tag .. '>' .. trim(cell)
elseif #cell_data == 1 then
cell = previous .. '<' .. last_tag .. '>' .. trim(cell_data[1])
else
local attributes = cell_data[1]
-- unstripBoth & fixTagAttributes
cell = previous .. '<' .. last_tag .. ' ' .. attributes .. '>' .. trim(cell_data[2])
end
outLine = outLine .. cell
table.insert(td_history, true)
end
end
out = out .. outLine .. '\n'
end--CONTINUE
end
while #td_history > 0 do
if table.remove(td_history) then
out = out .. '</td>'
end
if table.remove(tr_history) then
out = out .. '</tr>'
end
if not table.remove(has_opened_tr) then
out = out .. '<tr><td></td></tr>'
end
out = out .. '</table>'
end
if string.sub(out, -1) == '\n' then
out = string.sub(out, 1, -2)
end
if out == '<table><tr><td></td></tr></table>' then
out = ''
end
return out
end
-- converted from BlockLevelPass.php with a few adjustments
DTopen = false
lastParagraph = ''
COLON_STATE = {
['TEXT']=0,
['TAG']=1,
['TAGSTART']=2,
['CLOSETAG']=3,
['TAGSLASH']=4,
['COMMENT']=5,
['COMMENTDASH']=6,
['COMMENTDASHDASH']=7,
['LC']=8
}
function hasOpenParagraph()
return lastParagraph ~= ''
end
function closeParagraph(atTheEnd)
atTheEnd = atTheEnd or false
local result = ''
if hasOpenParagraph() then
result = '</' .. lastParagraph .. '>'
if not atTheEnd then
result = result .. '\n'
end
end
lastParagraph = ''
return result
end
function getCommon(st1, st2)
local shorter = math.min(string.len(st1), string.len(st2))
local count=0
for i=1, shorter do
if string.sub(st1, i, i) ~= string.sub(st2, i, i) then
break
end
count = count + 1
end
return count
end
function openList(char)
local result = closeParagraph()
if char == '*' then
result = result .. '<ul><li>'
elseif char == '#' then
result = result .. '<ol><li>'
elseif char == ':' then
result = result .. '<dl><dd>'
elseif char == ';' then
result = result .. '<dl><dt>'
DTopen = true
end
return result
end
function nextItem(char)
if char == '*' or char == '#' then
return '</li>\n<li>'
elseif char == ':' or char == ';' then
local close = '</dd>\n'
if DTopen then
close = '</dt>\n'
end
if char == ';' then
DTopen = true
return close .. '<dt>'
else
DTopen = false
return close .. '<dd>'
end
end
return ''
end
function closeList(char)
local text = ''
if char == '*' then
text = '</li></ul>'
elseif char == '#' then
text = '</li></ol>'
elseif char == ':' then
if DTopen then
DTopen = false
text = '</dt></dl>'
else
text = '</dd></dl>'
end
end
return text
end
function parseBlockLevel(text)
local textLines = explode('\n', text)
local lastPrefix = ''
local output = ''
local DTopen = false
local inBlockElem = false
local prefixLength = 0
local pendingPTag = false
local inBlockquote = false
local prefix2 = ''
for i, inputLine in ipairs(textLines) do
local lastPrefixLength = string.len(lastPrefix)
prefixLength = strspn(inputLine, '*#:;')
local prefix = string.sub(inputLine, 1, prefixLength)
prefix2 = string.gsub(prefix, ';', ':')
local t = string.sub(inputLine, prefixLength+1)
if prefixLength ~= 0 and lastPrefix == prefix2 then
output = output .. nextItem(string.sub(prefix, -1, -1))
pendingPTag = false
if string.sub(prefix, -1, -1) == ';' then
local term_t2 = {'', ''}
if findColonNoLinks(t, term_t2) ~= false then
t = term_t2[1]
output = output .. trim(term) .. nextItem(':')
end
end
elseif prefixLength ~= 0 or lastPrefixLength ~= 0 then
local commonPrefixLength = getCommon(prefix, lastPrefix)
pendingPTag = false
while commonPrefixLength < lastPrefixLength do
output = output .. closeList(string.sub(lastPrefix, lastPrefixLength, lastPrefixLength))
lastPrefixLength = lastPrefixLength - 1
end
if prefixLength <= commonPrefixLength and commonPrefixLength > 0 then
output = output .. nextItem(string.sub(prefix, commonPrefixLength, commonPrefixLength))
end
if DTopen and commonPrefixLength > 0 and string.sub(prefix, commonPrefixLength, commonPrefixLength) == ':' then
output = output .. nextItem(':')
end
if lastPrefix ~= '' and prefixLength > commonPrefixLength then
output = output .. '\n'
end
while prefixLength > commonPrefixLength do
local char = string.sub(prefix, commonPrefixLength + 1, commonPrefixLength + 1)
output = output .. openList(char)
if char == ';' then
local term_t2 = {'', ''}
if findColonNoLinks(t, term_t2) ~= false then
t = term_t2[1]
output = output .. trim(term) .. nextItem(':')
end
end
commonPrefixLength = commonPrefixLength + 1
end
if not prefixLength ~= 0 and lastPrefix ~= 0 then
output = output .. '\n'
end
lastPrefix = prefix2
end
if prefixLength == 0 then
local blockElems = {'table', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'pre', 'p', 'ul', 'ol', 'dl'}
local antiBlockElems = {'td', 'th'}
local openMatch = false
for i, elem in ipairs(blockElems) do
if string.match(t, '<' .. elem .. '%f[%W]') then
openMatch = true
end
end
if not openMatch then
for i, elem in ipairs(antiBlockElems) do
if string.match(t, '</' .. elem .. '%f[%W]') then
openMatch = true
end
end
end
if not openMatch then
for i, elem in ipairs({'tr', 'caption', 'dt', 'dd', 'li'}) do
if string.match(t, '</?' .. elem .. '%f[%W]') then
openMatch = true
end
end
end
local closeMatch = false
for i, elem in ipairs(blockElems) do
if string.match(t, '</' .. elem .. '%f[%W]') then
closeMatch = true
end
end
if not closeMatch then
for i, elem in ipairs(antiBlockElems) do
if string.match(t, '<' .. elem .. '%f[%W]') then
closeMatch = true
end
end
end
if not closeMatch then
for i, elem in ipairs({'center', 'blockquote', 'div', 'hr', 'mw:', 'aside', 'figure'}) do
if string.match(t, '</?' .. elem .. '%f[%W]') then
closeMatch = true
end
end
end
if openMatch or closeMatch then
pendingPTag = false
closeParagraph()
local bqOffset = 1
local bqMatch = {}
while match_with_offset('<(/?)blockquote[%s>]', t, bqMatch, bqOffset) do
inBlockquote = not bqMatch[0]
bqOffset = bqMatch[1] + string.len(bqMatch[0])
end
inBlockElem = not closeMatch
elseif not inBlockElem then
-- "and
-- trim(" triggers WAF
if trim(t) ~= ''
and string.sub(t, 1, 2) == ' '
and not inBlockquote then
t = string.sub(t, 2)
elseif string.match(t, '^<style%f[%W][^>]*>.-</style>$')
or string.match(t, '<link%f[%W][^>]*>%s*') then
if pendingPTag ~= '' and pendingPTag ~= false then
output = output .. closeParagraph()
pendingPTag = false
end
else
if trim(t) == '' then
if pendingPTag ~= '' and pendingPTag ~= false then
output = output .. pendingPTag .. '<br />'
pendingPTag = false
lastParagraph = 'p'
elseif lastParagraph ~= 'p' then
output = output .. closeParagraph()
pendintPTag = '<p>'
else
pendingPTag = '</p><p>'
end
elseif pendingPTag ~= '' and pendingPTag ~= false then
output = output .. pendingPTag
pendingPTag = false
lastParagraph = 'p'
elseif lastParagraph ~= 'p' then
output = output .. closeParagraph() .. '<p>'
lastParagraph = 'p'
end
end
end
end
if pendingPTag == false then
if prefixLength == 0 then
output = output .. t
if notLastLine or hasOpenParagraph() then
output = output .. '\n'
end
else
output = output .. trim(t)
end
end
end
while prefixLength > 0 do
output = output .. closeList(string.sub(prefix2, prefixLength, prefixLength))
prefixLength = prefixLength - 1
if prefixLength ~= 0 and hasOpenParagraph() then
output = output .. '\n'
end
end
output = output .. closeParagraph(true)
return output
end
function findColonNoLinks(str, before_after)
m = {}
if not (match_with_offset(':', str, m) or match_with_offset('<', str, m) or match_with_offset('-{', str, m)) then
return false
end
if m[0] == ':' then
local colonPos = m[1]
before_after[1] = string.sub(str, 1, colonPos+1)
before_after[2] = string.sub(str, colonPos+2)
return colonPos
end
local state = COLON_STATE.TEXT
local ltLevel = 0
local lcLevel = 0
local len = string.len(str)
for i = m[1], len-1 do
local c = string.sub(str, i, i)
if state == COLON_STATE.TEXT then
if c == '<' then
state = COLON_STATE.TAGSTART
elseif c == ':' then
if ltLevel == 0 then
before_after[1] = string.sub(str, 1, i+1)
before_after[2] = string.sub(str, i+2)
return i
end
else
if not (match_with_offset(':', str, m) or match_with_offset('<', str, m) or match_with_offset('%-{', str, m)) then
return false
end
if m[0] == '-{' then
state = COLON_STATE.LC
lcLevel = lcLevel + 1
i = m[1] + 1
else
i = m[1] - 1
end
end
elseif state == COLON_STATE.LC then
if not (match_with_offset('%-{', str, m, i+1) or match_with_offset('}%-', str, m, i+1)) then
break
end
if m[0] == '-{' then
i = m[1] + 1
lcLevel = lcLevel + 1
elseif m[0] == '}-' then
i = m[1] + 1
lcLevel = lcLevel - 1
if lcLevel == 0 then
state = COLON_STATE.TEXT
end
end
elseif state == CONON_STATE.TAG then
if c == '>' then
ltLevel = ltLevel + 1
state = COLON_STATE.TEXT
elseif c == '/' then
state = COLON_STATE.TAGSLASH
end
elseif state == COLON_STATE.TAGSTART then
if c == '/' then
state = COLON_STATE.CLOSETAG
elseif c == '!' then
state = COLON_STATE.COMMENT
elseif c == '>' then
state = COLON_STATE.TEXT
else
state = COLON_STATE.TAG
end
elseif state == COLON_STATE.CLOSETAG then
if c == '>' then
if ltLevel > 0 then
ltLevel = ltLevel - 1
end
state = COLON_STATE.TEXT
end
elseif state == COLON_STATE.TAGSLASH then
if c == '-' then
state = COLON_STATE.COMMENTDASH
else
state = COLON_STATE.COMMENT
end
elseif state == COLON_STATE.COMMENTDASH then
if c == '>' then
state = COLON_STATE.TEXT
else
state = COLON_STATE.COMMENT
end
end
end
return false
end
-- the Module
local p = {}
local getArgs = require('Module:Arguments').getArgs
function p.main(frame)
local args = getArgs(frame, { removeBlanks = false, frameOnly = true })
local rawText = args[1] or ''
rawText = mw.text.unstripNoWiki(rawText)
rawText = mw.text.decode(rawText)
rawText = frame:preprocess(rawText)
processedText = rawText
processedText = parseTables(processedText)
processedText = parseBlockLevel(processedText)
processedText = string.gsub(processedText, '\n', '')
if string.sub(processedText, 1, 4) == '<th>' then
processedText = string.sub(processedText, 5)
end
if string.sub(processedText, -5, -1) == '</th>' then
processedText = string.sub(processedText, 1, -6)
end
return processedText
end
return p
模块文档