How can I use regex to copy each word of a string into a hash?

cei has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: How can I use regex to copy each word of a string into a hash? by Ovid (Cardinal) on Jan 16, 2001 at 07:12 UTC
Are you sure you don't mean an array? How would you break the words up into keys and values? If you just want to stuff values into the keys of the hash (with '1' as the value), the basic form looks like this: `my $sentence = 'This is a test'; my %hash = map { $_, 1 } ($sentence =~ /\w+/g);` [download] If you need a count, you could do this: `my $sentence = 'This is a test'; my %hash; for ($sentence =~ /\w+/g) { $hash{ $_ }++; }` [download]	[reply] [d/l] [select]
Re: Re: How can I use regex to copy each word of a string into a hash? by cei (Monk) on Jan 16, 2001 at 23:42 UTC
A frequency count is exactly the direction I was headed. Perl Cookbook gave an example for "Processing Every Word in a File", but it wasn't clear to me how to apply it to a string instead of a file. thanks	[reply]
Re: How can I use regex to copy each word of a string into a hash? by I0 (Priest) on Jan 16, 2001 at 11:43 UTC
I'm not quite sure what you're asking for. Did you want something like this? `@hash{$string =~ /(\w['\w-]*)/g} = ();`	[reply] [d/l]
Re: How can I use regex to copy each word of a string into a hash? by ColonelPanic (Friar) on Jan 16, 2001 at 18:29 UTC
Keep in mind that `split` will generally be a more efficient solution, whenever it is possible. When's the last time you used duct tape on a duct? --Larry Wall	[reply] [d/l]
Re: Re: How can I use regex to copy each word of a string into a hash? by cei (Monk) on Jan 16, 2001 at 23:45 UTC
True, but if I'm trying to index english text, the oddities of punctuation may make a split impractical. Unless you've got a suggestion on an iterative split?	[reply]
If you just want words... by tedv (Pilgrim) on Jan 17, 2001 at 01:37 UTC
If you just want words, why bother with punctuation at all? Just do a massive `s/\W+/ /g` on the string beforehand and you'll get a big list of words, separated by spaces. I suppose those damn apostrophies will cause you pain, and you want "it's" to differ from "its". It's unclear whether or not capitalization matters-- is "BASIC" a different word from "basic"? What about "Smith" versus "smith"? Anyway, I'd probably write something like this: `local $/ = undef; $_ = <MY_FILE>; my %hash = (); $hash{$_}++ foreach split /[^\w']+/; # Change $_ to lc if case matters` [download] -Ted	[reply] [d/l] [select]
How can I use "How can I use regex to copy each word of a string into a hash?"? by frankus (Priest) on Jan 16, 2001 at 16:05 UTC
I'm really boring. I take bits of code for my repository; this looks useful for repeated word searching in a string^, by using a cache. I'm off to write functionality (and benchmark) so the first search uses regex: subsequent searches the hash. ^Provided your only need to test for existence. -- Brother Frankus.	[reply]