Module:String

[查看 | 编辑 | 历史 | 清除缓存]文档页面跳转至代码 ↴

这个模块提供了基本的字符串操作函数。

此处提供的大多数函数都可以使用命名参数、未命名参数或混合参数调用。 如果使用命名参数，Mediawiki将自动从参数中删除任何前导或尾随空格。取决于预期用途，保留或删除此类空白各有优劣的。

全局参数

ignore_errors

如果设置为“true”或“1”，则任何错误条件都将导致返回空字符串而不是错误消息。

error_category

如果发生错误，则指定要包含在错误消息中的分类的名称。默认分类为Template:Clc。

no_category

如果设置为“true”或 1，则在生成错误时不会添加任何类别。

此模块的单元测试可在Module:String/testcases进行。

len

此函数返回目标字符串的长度。

用法：

{{#invoke:String|len|target_string}}

或

{{#invoke:String|len|s= target_string }}

参数：

s

目标字符串

示例：

• {{#invoke:String|len| abcdefghi }} → 11
• {{#invoke:String|len|s= abcdefghi }} → 9

sub

此函数返回指定索引处目标字符串的子字符串。

用法：

{{#invoke:String|sub|target_string|start_index|end_index}}

或

{{#invoke:String|sub|s= target_string |i= start_index |j= end_index }}

参数：

s

返回此字符串的子集

i

要返回的子字符串的第一个索引，默认为1。

j

要返回的字符串的最后一个索引，默认为最后一个字符。

字符串的第一个字符的索引为1。如果i或j是负值，则将其解析为通过从字符串末尾计数来选择字符。因此，值-1与选择字符串的最后一个字符相同。

如果请求的索引超出给定字符串的范围，则会报告错误。

示例：

• "{{#invoke:String|sub| abcdefghi }}" → " abcdefghi "
• "{{#invoke:String|sub|s= abcdefghi }}" → "abcdefghi"
• "{{#invoke:String|sub| abcdefghi | 3 }}" → "bcdefghi "
• "{{#invoke:String|sub|s= abcdefghi |i= 3 }}" → "cdefghi"
• "{{#invoke:String|sub| abcdefghi | 3 | 4 }}" → "bc"
• "{{#invoke:String|sub|s= abcdefghi |i= 3 |j= 4 }}" → "cd"

sublength

此函数实现了{{Str sub old}}的功能，保留仅为维持旧的模板。

用法：

{{#invoke:String|sublength|s= 目标字符串 |i= 开始索引 |len= 长度 }}

参数：

s

字符串

i

返回的子串的开始索引。第一个个字符的索引为0。

len

返回的字符串的长度，默认为最后一个字符。

示例：

• {{#invoke:String|sublength|s= abcdefghi }} → abcdefghi
• {{#invoke:String|sublength|s= abcdefghi |i= 3 }} → defghi
• {{#invoke:String|sublength|s= abcdefghi |i= 3 |len= 4 }} → defg

match

此函数从源字符串返回一个匹配指定模式的子字符串。

用法：

{{#invoke:String|match|源字符串|模式字符串|开始索引|匹配数量|纯标记|无匹配输出}}

或者

{{#invoke:String|match|s= 源字符串 |pattern= 模式字符串 |start= 开始索引|match= 匹配数量|plain= 纯标记 |nomatch= 无匹配输出 }}

参数：

s

在这个字符串内搜索

pattern

需要在字符串内搜索的模式或者字符串

start

源字符串内的索引，从此索引开始搜索。字符串的第一个字符的索引为1。默认为1。

match

在一些情况下，可能需要在单个字符串中进行多次匹配。此参数指定了返回哪一个匹配，其中第一个匹配就是match=1。如果指定了负数，那么返回的匹配从最后一个匹配开始数的。因此match=-1表达返回最后一个匹配。默认为1。

plain

pattern是否应该理解为纯字符串，而非正则表达式。默认为false。

nomatch

如果没有找到匹配，返回nomatch的值，而不是一个错误。

如果match_number或start_index超出了查询的字符串的范围，此函数会产生错误。如果没有找到匹配，也会产生错误。

如果添加了参数ignore_errors=true，那么错误会被禁止，失败时返回的是一个空字符串。

匹配是Lua的一种正则表达式，更多信息可以参考：

• Scribunto patterns
• Scribunto Unicode string patterns

示例：

• {{#invoke:String|match| abc123def456 |%d+}} → 123
• {{#invoke:String|match|s= abc123def456 |pattern= %d+ }} → 123
• {{#invoke:String|match| abc123def456 |%d+|6}} → 23
• {{#invoke:String|match|s= abc123def456 |pattern= %d+ |start= 6 }} → 3
• {{#invoke:String|match|s= abc123def456 |pattern= %d+ |start= 6 |match= 2 }} → 456
• {{#invoke:String|match|s= abc123%d+ |pattern= %d+ }} → 123
• {{#invoke:String|match|s= abc123%d+ |pattern= %d+ |plain= true }} → %d+
• {{#invoke:String|match|s= abc |pattern= %d }} →
• {{#invoke:String|match|s= abc |pattern= %d |nomatch= No numeric characters in string }} → No numeric characters in string
• {{#invoke:String|match|s= abc |pattern= %d |ignore_errors= true }} →
• {{#invoke:String|match|s= 0012001200 |pattern= 0*(%d*) }} → 12001200

pos

返回在目标字符串中的位置pos处的单个字符。

用法：

{{#invoke:String|pos|目标字符串|索引值}}

或

{{#invoke:String|pos|target= 目标字符串 |pos= 索引值 }}

参数：

目标

在此字符串内搜索

pos

需要返回的字符所在的索引。

第一个字符的索引值为1。

如果为负值，则此函数会从字符串的最后面开始倒数。也就是说，pos=-1相当于返回最后一个字符。

如果此值为0，或者超出了字符串的长度，则会返回错误。

示例：

• {{#invoke:String|pos| abcdefghi | 4 }} → c
• {{#invoke:String|pos|target= abcdefghi |pos= 4 }} → d

str_find

此函数与{{str_find}}的行为重复，包含了其所有的技巧，仅为支持存在的模板，但是对于新的代码和模板不推荐使用。新的代码建议直接使用find函数。

返回源字符串中匹配目的的第一个索引。索引是从1开始数的，如果源字符串中找不到目标字符串，则返回-1。

注意：如果目标字符串是空白或者缺失，则函数返回值为1，这可能不是预期的行为，必须要单独考虑。

用法：

{{#invoke:String|str_find|源字符串|目标字符串}}

或

{{#invoke:String|str_find|source= 源字符串 |target= 目标字符串 }}

参数：

source

在此字符串内搜索

target

在源内搜索的字符串

示例：

• {{#invoke:String|str_find| abc123def }} → 1
• {{#invoke:String|str_find|source= abc123def }} → 1
• {{#invoke:String|str_find| abc123def |123}} → 5
• {{#invoke:String|str_find|source= abc123def |target= 123 }} → 4
• {{#invoke:String|str_find| abc123def |not}} → -1

find

此函数在一个字符串内搜索另一个字符串。

用法：

{{#invoke:String|find|源字符串|目标字符串|开始索引|纯标记}}

或

{{#invoke:String|find|source= 源字符串 |target= 目标字符串 |start= 开始索引 |plain= 纯标记 }}

参数：

source

在此字符串内搜索

target

在源字符串内搜索的字符串

start

在源字符串的此位置索引版主开始搜索。默认为1

plain

布尔值标记，表示目的应该被理解为纯文本，而不是Scribunto ustring模式，默认为true

此函数返回了在索引>"start"，其中"source"中可以持续到"target"。索引是从1开始数的。如果没有找到目标字符串，函数返回0。如果索引或者目的缺失或者是空的，此函数也会返回0。

此函数应该可以安全地操作UTF-8字符串。

示例：

• {{#invoke:String|find|abc123def|12}} → 4
• {{#invoke:String|find|source=abc123def|target=12}} → 4
• {{#invoke:String|find|source=abc123def|target=pqr}} → 0
• {{#invoke:String|find| abc123def |123}} → 5
• {{#invoke:String|find|source= abc123def |target= 123 }} → 4
• {{#invoke:String|find|source=abc123def|target=%d |start=3 |plain=false }} → 4

与未命名的参数使用时，两端的空格会被保留且被计入：

• {{#invoke:String|find| abc123def |c|false}} → 5
• {{#invoke:String|find|source= abc123def |target=c|plain=false}} → 3
• {{#invoke:string|find|abc 123 def|%s|plain=false}} → 4

检测字符串是否存在：

• Template:Pf → Didn't find needle

replace（gsub）

此函数在一个字符串内替换到目的字符串或者模式。此函数是通过调用mw.ustring.gsub实现的。

用法：

{{#invoke:String|replace|源字符串|模式字符串|替换字符串|替换次数|纯标记}}

或

{{#invoke:String|replace|source= 源字符串 |pattern= 模式字符串 |replace= 替换字符串 |count= 替换次数 |plain= 纯标记 }}

参数：

source

在此字符串内搜索

pattern

在源内查找此字符串或者模式

replace

使用此字符串替换

count

需要替换的次数，默认为all

plain

布尔值标记，表示目的应该被理解为纯文本，而不是Scribunto ustring模式，默认为true

示例：

• "{{#invoke:String|replace| abc123def456 |123|XYZ}}" → " abcXYZdef456 "
• "{{#invoke:String|replace|source= abc123def456 |pattern= 123 |replace= XYZ }}" → "abcXYZdef456"
• "{{#invoke:String|replace| abc123def456 |%d+|XYZ|1|false}}" → " abcXYZdef456 "
• "{{#invoke:String|replace|source= abc123def456 |pattern= %d+ |replace= XYZ |count=1 |plain= false }}" → "abcXYZdef456"
• "{{#invoke:String|replace|source= abc123def456 |pattern= %d+ |replace= XYZ |plain= false }}" → "abcXYZdefXYZ"
• {{#invoke:String|replace|source= 0012001200 |pattern= ^0* |plain= false }} → 12001200

rep

将一个字符串重复n次，其本质为string.rep。

用法：

{{#invoke:String|rep|源|次数}}

参数：

源

需要重复的字符串

次数

重复的次数

示例：

• "{{#invoke:String|rep|hello|3}}" → "hellohellohello"
• "{{#invoke:String|rep| hello | 3 }}" → " hello hello hello "

[查看 | 编辑 | 历史 | 清除缓存]上述文档是从Module:String/doc引用的。

--[[  

This module is intended to provide access to basic string functions.

Most of the functions provided here can be invoked with named parameters, 
unnamed parameters, or a mixture.  If named parameters are used, Mediawiki will 
automatically remove any leading or trailing whitespace from the parameter.  
Depending on the intended use, it may be advantageous to either preserve or
remove such whitespace.

Global options
    ignore_errors: If set to 'true' or 1, any error condition will result in 
        an empty string being returned rather than an error message.  
        
    error_category: If an error occurs, specifies the name of a category to 
        include with the error message.  The default category is  
        [Category:Errors reported by Module String].
        
    no_category: If set to 'true' or 1, no category will be added if an error
        is generated.
        
Unit tests for this module are available at Module:String/tests.
]]

local str = {}

--[[
len

This function returns the length of the target string.

Usage:
{{#invoke:String|len|target_string|}}
OR
{{#invoke:String|len|s=target_string}}

Parameters
    s: The string whose length to report

If invoked using named parameters, Mediawiki will automatically remove any leading or
trailing whitespace from the target string.  
]]
function str.len( frame )
    local new_args = str._getParameters( frame.args, {'s'} );
    local s = new_args['s'] or '';
    return mw.ustring.len( s )
end

--[[
sub

This function returns a substring of the target string at specified indices.

Usage:
{{#invoke:String|sub|target_string|start_index|end_index}}
OR
{{#invoke:String|sub|s=target_string|i=start_index|j=end_index}}

Parameters
    s: The string to return a subset of
    i: The fist index of the substring to return, defaults to 1.
    j: The last index of the string to return, defaults to the last character.
    
The first character of the string is assigned an index of 1.  If either i or j
is a negative value, it is interpreted the same as selecting a character by 
counting from the end of the string.  Hence, a value of -1 is the same as 
selecting the last character of the string.

If the requested indices are out of range for the given string, an error is 
reported.
]]
function str.sub( frame )
    local new_args = str._getParameters( frame.args, { 's', 'i', 'j' } );
    local s = new_args['s'] or '';
    local i = tonumber( new_args['i'] ) or 1;
    local j = tonumber( new_args['j'] ) or -1;
    
    local len = mw.ustring.len( s );

    -- Convert negatives for range checking
    if i < 0 then
        i = len + i + 1;
    end
    if j < 0 then
        j = len + j + 1;
    end
    
    if i > len or j > len or i < 1 or j < 1 then
        return str._error( '截取字符串索引脱离区间' );
    end
    if j < i then
        return str._error( '截取字符串指示脱离顺序' );
    end
    
    return mw.ustring.sub( s, i, j )
end

--[[
This function implements that features of {{str sub old}} and is kept in order
to maintain these older templates.
]]
function str.sublength( frame )
    local i = tonumber( frame.args.i ) or 0
    local len = tonumber( frame.args.len )
    return mw.ustring.sub( frame.args.s, i + 1, len and ( i + len ) )
end

--[[
match

This function returns a substring from the source string that matches a 
specified pattern.

Usage:
{{#invoke:String|match|source_string|pattern_string|start_index|match_number|plain_flag|nomatch_output}}
OR
{{#invoke:String|pos|s=source_string|pattern=pattern_string|start=start_index
    |match=match_number|plain=plain_flag|nomatch=nomatch_output}}

Parameters
    s: The string to search
    pattern: The pattern or string to find within the string
    start: The index within the source string to start the search.  The first
        character of the string has index 1.  Defaults to 1.
    match: In some cases it may be possible to make multiple matches on a single 
        string.  This specifies which match to return, where the first match is 
        match= 1.  If a negative number is specified then a match is returned 
        counting from the last match.  Hence match = -1 is the same as requesting
        the last match.  Defaults to 1.
    plain: A flag indicating that the pattern should be understood as plain
        text.  Defaults to false.
    nomatch: If no match is found, output the "nomatch" value rather than an error.

If invoked using named parameters, Mediawiki will automatically remove any leading or
trailing whitespace from each string.  In some circumstances this is desirable, in 
other cases one may want to preserve the whitespace.

If the match_number or start_index are out of range for the string being queried, then
this function generates an error.  An error is also generated if no match is found.
If one adds the parameter ignore_errors=true, then the error will be suppressed and 
an empty string will be returned on any failure.

For information on constructing Lua patterns, a form of [regular expression], see:

* http://www.lua.org/manual/5.1/manual.html#5.4.1
* http://www.mediawiki.org/wiki/Extension:Scribunto/Lua_reference_manual#Patterns
* http://www.mediawiki.org/wiki/Extension:Scribunto/Lua_reference_manual#Ustring_patterns

]]
function str.match( frame )
    local new_args = str._getParameters( frame.args, {'s', 'pattern', 'start', 'match', 'plain', 'nomatch'} );
    local s = new_args['s'] or '';
    local start = tonumber( new_args['start'] ) or 1;
    local plain_flag = str._getBoolean( new_args['plain'] or false );
    local pattern = new_args['pattern'] or '';
    local match_index = math.floor( tonumber(new_args['match']) or 1 );
    local nomatch = new_args['nomatch'];
    
    if s == '' then
        return str._error( '目标字符串是空的' );
    end
    if pattern == '' then
        return str._error( '模式字符串是空的' );
    end
    if math.abs(start) < 1 or math.abs(start) > mw.ustring.len( s ) then
        return str._error( '要求的起始点脱离区间' );
    end
    if match_index == 0 then
        return str._error( '匹配索引脱离区间' );
    end
    if plain_flag then
        pattern = str._escapePattern( pattern );
    end
    
    local result
    if match_index == 1 then
        -- Find first match is simple case
        result = mw.ustring.match( s, pattern, start )
    else
        if start > 1 then
            s = mw.ustring.sub( s, start );
        end
        
        local iterator = mw.ustring.gmatch(s, pattern);
        if match_index > 0 then
            -- Forward search
            for w in iterator do
                match_index = match_index - 1;
                if match_index == 0 then
                    result = w;
                    break;
                end
            end    
        else
            -- Reverse search
            local result_table = {};
            local count = 1;
            for w in iterator do
                result_table[count] = w;
                count = count + 1;
            end
            
            result = result_table[ count + match_index ];            
        end
    end        
    
    if result == nil then
        if nomatch == nil then
            return str._error( '找不到匹配' );
        else
            return nomatch;
        end
    else
        return result;
    end
end

--[[
pos

This function returns a single character from the target string at position pos.

Usage:
{{#invoke:String|pos|target_string|index_value}}
OR
{{#invoke:String|pos|target=target_string|pos=index_value}}

Parameters
    target: The string to search
    pos: The index for the character to return

If invoked using named parameters, Mediawiki will automatically remove any leading or
trailing whitespace from the target string.  In some circumstances this is desirable, in 
other cases one may want to preserve the whitespace.

The first character has an index value of 1.

If one requests a negative value, this function will select a character by counting backwards 
from the end of the string.  In other words pos = -1 is the same as asking for the last character.

A requested value of zero, or a value greater than the length of the string returns an error.
]]
function str.pos( frame )
    local new_args = str._getParameters( frame.args, {'target', 'pos'} );
    local target_str = new_args['target'] or '';
    local pos = tonumber( new_args['pos'] ) or 0;

    if pos == 0 or math.abs(pos) > mw.ustring.len( target_str ) then
        return str._error( '字符串索引脱离区间' );
    end    
    
    return mw.ustring.sub( target_str, pos, pos );
end

--[[
str_find

This function duplicates the behavior of {{str_find}}, including all of its quirks.
This is provided in order to support existing templates, but is NOT RECOMMENDED for 
new code and templates.  New code is recommended to use the "find" function instead.

Returns the first index in "source" that is a match to "target".  Indexing is 1-based,
and the function returns -1 if the "target" string is not present in "source".

Important Note: If the "target" string is empty / missing, this function returns a
value of "1", which is generally unexpected behavior, and must be accounted for
separatetly.
]]
function str.str_find( frame )
    local new_args = str._getParameters( frame.args, {'source', 'target'} );
    local source_str = new_args['source'] or '';
    local target_str = new_args['target'] or '';

    if target_str == '' then
        return 1;
    end    
    
    local start = mw.ustring.find( source_str, target_str, 1, true )
    if start == nil then
        start = -1
    end
    
    return start
end

--[[
find

This function allows one to search for a target string or pattern within another
string.

Usage:
{{#invoke:String|find|source_str|target_string|start_index|plain_flag}}
OR
{{#invoke:String|find|source=source_str|target=target_str|start=start_index|plain=plain_flag}}

Parameters
    source: The string to search
    target: The string or pattern to find within source
    start: The index within the source string to start the search, defaults to 1
    plain: Boolean flag indicating that target should be understood as plain
        text and not as a Lua style regular expression, defaults to true

If invoked using named parameters, Mediawiki will automatically remove any leading or
trailing whitespace from the parameter.  In some circumstances this is desirable, in 
other cases one may want to preserve the whitespace.

This function returns the first index >= "start" where "target" can be found 
within "source".  Indices are 1-based.  If "target" is not found, then this 
function returns 0.  If either "source" or "target" are missing / empty, this
function also returns 0.

This function should be safe for UTF-8 strings.
]]
function str.find( frame )
    local new_args = str._getParameters( frame.args, {'source', 'target', 'start', 'plain' } ); 
    local source_str = new_args['source'] or '';
    local pattern = new_args['target'] or '';
    local start_pos = tonumber(new_args['start']) or 1;
    local plain = new_args['plain'] or true;
        
    if source_str == '' or pattern == '' then
        return 0;
    end    
    
    plain = str._getBoolean( plain );

    local start = mw.ustring.find( source_str, pattern, start_pos, plain )
    if start == nil then
        start = 0
    end
    
    return start
end

--[[
replace

This function allows one to replace a target string or pattern within another
string.

Usage:
{{#invoke:String|replace|source_str|pattern_string|replace_string|replacement_count|plain_flag}}
OR
{{#invoke:String|replace|source=source_string|pattern=pattern_string|replace=replace_string|
   count=replacement_count|plain=plain_flag}}

Parameters
    source: The string to search
    pattern: The string or pattern to find within source
    replace: The replacement text
    count: The number of occurences to replace, defaults to all.
    plain: Boolean flag indicating that pattern should be understood as plain
        text and not as a Lua style regular expression, defaults to true 
]]
function str.replace( frame )
    local new_args = str._getParameters( frame.args, {'source', 'pattern', 'replace', 'count', 'plain' } ); 
    local source_str = new_args['source'] or '';
    local pattern = new_args['pattern'] or '';
    local replace = new_args['replace'] or '';
    local count = tonumber( new_args['count'] );
    local plain = new_args['plain'] or true;
        
    if source_str == '' or pattern == '' then
        return source_str;
    end    
    plain = str._getBoolean( plain );

    if plain then
        pattern = str._escapePattern( pattern );
        replace = mw.ustring.gsub( replace, "%%", "%%%%" ); --Only need to escape replacement sequences.
    end
    
    local result;

    if count ~= nil then
        result = mw.ustring.gsub( source_str, pattern, replace, count );
    else
        result = mw.ustring.gsub( source_str, pattern, replace );
    end        

    return result;
end

--[[ 
    simple function to pipe string.rep to templates.
]]

function str.rep( frame )
    local repetitions = tonumber( frame.args[2] )
    if not repetitions then 
        return str._error( 'function rep expects a number as second parameter, received "' .. ( frame.args[2] or '' ) .. '"' )
    end
    return string.rep( frame.args[1] or '', repetitions )
end


function str.split(inputstr, sep, no_pattern, ignore_null)
	--#invoke 支援
	if type(inputstr) == type({table}) then
		if not getArgs then getArgs = require('Module:Arguments').getArgs end
		args = getArgs(inputstr, {parentFirst=true})
		for arg_name, arg_value in pairs( args ) do
			if arg_name == 1 or arg_name == '1' or arg_name == "str" or arg_name == "inputstr" or arg_name == "input" then
				input_str = arg_value
			elseif arg_name == 2 or arg_name == '2' or arg_name == "sep" or arg_name == "separator" then
				separ = arg_value
			elseif arg_name == 3 or arg_name == '3' or arg_name == "no_pattern" or arg_name == "no pattern" then
				no_pattern_flag = arg_value
			elseif arg_name == 4 or arg_name == '4' or arg_name == "ignore_null" or arg_name == "ignore null" then
				ignore_null_flag = arg_value
			elseif arg_name == 5 or arg_name == '5' or arg_name == "format" then
				format = arg_value or "*{{{1}}}\n";
			end
		end
		if not yesno then yesno = require('Module:Yesno') end
		no_pattern_flag = yesno( no_pattern_flag or 'yes' )
		ignore_null_flag = yesno( ignore_null_flag or 'no' )
		is_invoke = true
		format = mw.ustring.gsub(format or "*{{{1}}}\n", "%{%{%{.-%}%}%}", "%%s" );
		it = mw.ustring.find(format, "%%s", 1)
		if it == nil then format = format .. "%s" end
		format = mw.ustring.gsub(format, "\\n", "\n")
	else
		input_str = inputstr
		separ = sep
		no_pattern_flag = no_pattern
		ignore_null_flag = ignore_null
		is_invoke = false
	end
	input_str = input_str or '' 
	separ = separ or "%s"
	if no_pattern_flag == nil then no_pattern_flag = true end
	if ignore_null_flag == nil then ignore_null_flag = false end

	length = mw.ustring.len(input_str)
	--split函數起點
	if no_pattern_flag then
		separ = mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(mw.ustring.gsub(separ, 
			"%[", "%["), "%]", "%]"), "%{", "%{"), "%}", "%}"), "%%", "%%%%"), "%)", "%)"), "%-", "%-"),
			"%^", "%^"), "%$", "%$"), "%(", "%("), "%.", "%."), "%*", "%*"), "%+", "%+"), "%|", "%|");
	end
	iterator = 1 ; i = 1 ; flag = true
	result = {}
	separ_str_begin, separ_str_end = mw.ustring.find(input_str, separ, iterator)
	--
	debug1 = 1
	--
	while flag do
		debug1 = debug1 + 1
		if separ_str_begin == nil or iterator > length or debug1 >= 100 then
			separ_str_begin = 0
			separ_str_end = -2
			flag = false
		end
		if separ_str_end < separ_str_begin then separ_str_end = separ_str_begin end
		finded_str = mw.ustring.sub(input_str, iterator, separ_str_begin - 1)
		if not(mw.text.trim(finded_str) == '' and ignore_null_flag) then
			result[i] = finded_str
			i = i + 1
		end

		iterator = separ_str_end + 1
		separ_str_begin, separ_str_end = mw.ustring.find(input_str, separ, iterator)
	end
	if is_invoke then
		body = ''
		for i, result_str in pairs( result ) do
			body = body .. mw.ustring.gsub(format, "%%s", result_str)
		end
		return body
	end
	return result;
end

--[[
join

Join all non empty arguments together; the first argument is the separator.
Usage:
{{#invoke:String|join|sep|one|two|three}}
]]
function str.join(frame)
	local args = {}
	local sep
	for _, v in ipairs( frame.args ) do
		if sep then
			if v ~= '' then
				table.insert(args, v)
			end
		else
			sep = v
		end
	end
	return table.concat( args, sep or '' )
end

--[[
Helper function that populates the argument list given that user may need to use a mix of
named and unnamed parameters.  This is relevant because named parameters are not
identical to unnamed parameters due to string trimming, and when dealing with strings
we sometimes want to either preserve or remove that whitespace depending on the application.
]]
function str._getParameters( frame_args, arg_list )
    local new_args = {};
    local index = 1;
    local value;
    
    for i,arg in ipairs( arg_list ) do
        value = frame_args[arg]
        if value == nil then
            value = frame_args[index];
            index = index + 1;
        end
        new_args[arg] = value;
    end
    
    return new_args;
end        

--[[
Helper function to handle error messages.
]]
function str._error( error_str )
    local frame = mw.getCurrentFrame();
    local error_category = frame.args.error_category or '字符串模块报告的错误';
    local ignore_errors = frame.args.ignore_errors or false;
    local no_category = frame.args.no_category or false;
    
    if str._getBoolean(ignore_errors) then
        return '';
    end
    
    local error_str = '<strong class="error">字符串模块出错：' .. error_str .. '</strong>';
    if error_category ~= '' and not str._getBoolean( no_category ) then
        error_str = '[[Category:' .. error_category .. ']]' .. error_str;
    end        
    
    return error_str;
end

--[[
Helper Function to interpret boolean strings
]]
function str._getBoolean( boolean_str )
    local boolean_value;
    
    if type( boolean_str ) == 'string' then
        boolean_str = boolean_str:lower();
        if boolean_str == 'false' or boolean_str == 'no' or boolean_str == '0' 
                or boolean_str == '' then
            boolean_value = false;
        else
            boolean_value = true;
        end    
    elseif type( boolean_str ) == 'boolean' then
        boolean_value = boolean_str;
    else
        error( '布尔值找不到' );
    end    
    return boolean_value
end

--[[
Helper function that escapes all pattern characters so that they will be treated 
as plain text.
]]
function str._escapePattern( pattern_str )
    return mw.ustring.gsub( pattern_str, "([%(%)%.%%%+%-%*%?%[%^%$%]])", "%%%1" );
end

return str