英文:
Ruby string replace without capture group
问题
Ruby的String.sub(pat, txt)函数族执行替换操作，并且支持在txt中捕获组，即使pat是一个普通的字符串（至少它似乎支持\0来在替换文本中打印整个pat）。我不知道这是否是有意的，但如果txt是来自API输入的变量，那么它是不安全的。是否有替代的库函数可以用来执行字符串替换而不具有这个缺陷？或者有没有一种方式可以对txt进行转义。
英文:
Ruby's String.sub(pat, txt) family of functions performs substitution and supports capturing groups in txt even if pat is a plain string (at least it seems to support \0 to print all of pat in the replaced text). I don't know if this is intended, but is unsafe if txt is e.g. a variable taken as input from an API. Is there an alternative library function I can use to perform string replace without this drawback? Or alternatively some way to escape txt.
答案1
得分: 2
你的意思是这样的：
>> "mytext".sub("text", " \\0 替换为 \\` \\0")
=> "my text 替换为 my text"
你可以转义反斜杠：
>> "mytext".sub("text", " \\\\0 替换")
=> "my \\0 替换"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 替换"))
=> "my \\0 替换"
# 有多少个反斜杠？
>> "\\0".chars #=> ["\\", "0"] # 嗯.. 是的
>> "\\0".bytes #=> [92, 48]
# 一个 ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# 两个 ^ ^
或者使用一个块：
>> "mytext".sub("text") { " \\0 替换" }
=> "my \\0 替换"
或者也可以使用 []=：
>> t = "mytext"
=> "mytext"
>> t["text"] = " \\0 替换"
=> " \\0 替换"
>> t
=> "my \\0 替换"
请注意，在字符串替换中，诸如 $& 这样的字符组合被视为普通文本，而不是特殊的匹配变量。然而，您可以使用以下组合引用一些特殊的匹配变量：
\&和\0对应于$&，其中包含完全匹配的文本。\'对应于$'，其中包含匹配后的字符串。\\\`` 对应于$``，其中包含匹配前的字符串。\+对应于$+，其中包含最后一个捕获组。
https://docs.ruby-lang.org/en/3.2/String.html#class-String-label-Substitution+Methods
英文:
You mean like this:
>> "mytext".sub("text", " \\0 replaced by \\` \\0")
=> "my text replaced by my text"
You can escape backslashes:
>> "mytext".sub("text", " \\\>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
replaced")
=> "my \>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
replaced"))
=> "my \>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
replaced"
# how many backslashes?
>> "\>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
".chars #=> ["\\", "0"] # uhh.. yes
>> "\>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
".bytes #=> [92, 48]
# one ^
>> b_esc("\>> "mytext".sub("text", " \\\\0 replaced")
=> "my \\0 replaced"
def b_esc(str) = str.gsub("\\", "\\\\\\\\")
>> "mytext".sub("text", b_esc(" \\0 replaced"))
=> "my \\0 replaced"
# how many backslashes?
>> "\\0".chars #=> ["\\", "0"] # uhh.. yes
>> "\\0".bytes #=> [92, 48]
# one ^
>> b_esc("\\0").bytes #=> [92, 92, 48]
# two ^ ^
").bytes #=> [92, 92, 48]
# two ^ ^
Or use a block:
>> "mytext".sub("text") { " \\0 replaced" }
=> "my \\0 replaced"
Or maybe []=:
>> t = "mytext"
=> "mytext"
>> t["text"] = " \\0 replaced"
=> " \\0 replaced"
>> t
=> "my \\0 replaced"
> Note that within the string replacement, a character combination such as $& is treated as ordinary text, and not as a special match variable. However, you may refer to some special match variables using these combinations:
>
> \& and \0 correspond to $&, which contains the complete matched text.
> \' corresponds to $', which contains string after match.
> <code>\`</code> corresponds to <code>$`</code>, which contains string before match.
> \+ corresponds to $+, which contains last capture group.
https://docs.ruby-lang.org/en/3.2/String.html#class-String-label-Substitution+Methods
答案2
得分: 0
在我的情况下，最终选择了index + insert，因为单词总是以该模式结尾。如果不是这种情况，[]=就是我正在寻找的解决方案。
def prepend(txt, pat, word)
index = txt.index(pat)
return if index.nil?
txt.insert(index, word)
true
end
英文:
Ended up going with index + insert in my case since the word always ended with the pattern. []= was the solution I was looking for if this weren't the case.
def prepend(txt, pat, word)
index = txt.index(pat)
return if index.nil?
txt.insert(index, word)
true
end
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论