带Ruby的小写BBCode标签

Question

I'm trying to covert a forum where BBCode tag are upcase. I need to write a parser to downcase my tag, from [QUOTE] to [quote], from [/QUOTE] to [/quote], etc etc.

I write this:

string.gsub(/#\[(.*?)\]/, ' \1'.downcase)

but of course it doesn't work!

How can I fix it?

Answer 1

• Your # is wrong.
• You don't need to use capture (by putting () ). You can refer to the entire match. [ , ] , / will remain as is by downcase , so no harm including them. In fact, your regex already possibly includes / in the capture, so it does not make sense to exclude just [ and ] from the capture.
• Your '\\1'.downcase did not work because that is equivalent to '\\1' . To perform a method on the match, you need a block.
• I assume your .*? in the regex intends to capture nested brackets correctly, but that works only half way. If you had "[foo [bar] baz]" , then by \\[(.*?)\\] , you can avoid matching "[foo [bar] baz]" and "[bar] baz]" , but it still mathces "[foo [bar]" . So .*? is not meaningful.

Considering the points above, you can do the following if you need to consider nested brackets:

string.gsub(/\[[^\[\]]+\]/, &:downcase)

Otherwise,

string.gsub(/\[.+\]/, &:downcase)

Answer 2

You can modify the match by using the block version of gsub .

 s.gsub(/\[(.+?)\]/) { |match| match.downcase }

or the more compact version

 s.gsub(/\[(.+?)\]/, &:downcase)

Also note there was a # that was preventing the regexp to match.

Example:

s = "from [QUOTE] to [quote], from [/QUOTE] to [/quote]"
s.gsub(/\[(.*?)\]/, &:downcase)
 => "from [quote] to [quote], from [/quote] to [/quote]"