Module:String2/sandbox: Difference between revisions

Revision as of 09:09, 9 June 2021

Revision as of 01:27, 2 September 2021

Line 1:

local getArgs = require('Module:Arguments').getArgs

local p = {}

Line 292 ⟶ 291:

end

--[[--------------------------< H Y P H E N _ T O _ D A S H >--------------------------------------------------

--[[

Truncate a string

Converts a hyphen to a dash under certain conditions. The hyphen must separate

Implements Template:Trunc, which is similar to string.sub, but when second

like items; unlike items are returned unmodified. These forms are modified:

argument is missing or is not a number, return entire string

letter - letter (A - B)

digit - digit (4-5)

digit separator digit - digit separator digit (4.1-4.5 or 4-1-4-5)

letterdigit - letterdigit (A1-A5) (an optional separator between letter and

digit is supported – a.1-a.5 or a-1-a-5)

digitletter - digitletter (5a - 5d) (an optional separator between letter and

digit is supported – 5.a-5.d or 5-a-5-d)

any other forms are returned unmodified.

str may be a comma- or semicolon-separated list

Usage:

{{#invoke:String2|trunc|abc123|3}}

Parameters:

args[1]: string to truncate

args[2]: number of characters to keep

]]

function p.trunchyphen_to_dash(frame str )

local args = getArgs(frame,{parentFirst=true})

str = str:gsub ('&[nm]dash;', {['&ndash;'] = '–', ['&mdash;'] = '—'}); -- replace &mdash; and &ndash; entities with their characters; semicolon mucks up the text.split

if not args[1] then

str = str:gsub ('&#45;', '-'); -- replace HTML numeric entity with hyphen character

return ""

str = str:gsub ('&nbsp;', ' '); -- replace &nbsp; entity with generic keyboard space character

local out = {};

local list = mw.text.split (str, '%s*[,;]%s*'); -- split str at comma or semicolon separators if there are any

for _, item in ipairs (list) do -- for each item in the list

item = mw.text.trim(item) -- trim whitespace

if mw.ustring.match (item, '^%w*[%.%-]?%w+%s*[%-–—]%s*%w*[%.%-]?%w+$') then -- if a hyphenated range or has endash or emdash separators

if item:match ('^%a+[%.%-]?%d+%s*%-%s*%a+[%.%-]?%d+$') or -- letterdigit hyphen letterdigit (optional separator between letter and digit)

item:match ('^%d+[%.%-]?%a+%s*%-%s*%d+[%.%-]?%a+$') or -- digitletter hyphen digitletter (optional separator between digit and letter)

item:match ('^%d+[%.%-]%d+%s*%-%s*%d+[%.%-]%d+$') or -- digit separator digit hyphen digit separator digit

item:match ('^%d+%s*%-%s*%d+$') or -- digit hyphen digit

item:match ('^%a+%s*%-%s*%a+$') then -- letter hyphen letter

item = item:gsub ('(%w*[%.%-]?%w+)%s*%-%s*(%w*[%.%-]?%w+)', '%1–%2'); -- replace hyphen, remove extraneous space characters

else

item = mw.ustring.gsub (item, '%s*[–—]%s*', '–'); -- for endash or emdash separated ranges, replace em with en, remove extraneous whitespace

end

end

table.insert (out, item); -- add the (possibly modified) item to the output table

end

if not args[2] then

return table.concat (out, ', '); -- concatenate the output table into a comma separated string

return args[1]

end

length = tonumber(args[2])

function p.hyphen2dash( frame )

if not length then

return local str = frame.args[1] or ''

return p.hyphen_to_dash(str)

end

return mw.ustring.sub(args[1],1,length)

end