Dictionary Grep/Search

Note: the wildcard character is a dot: ”.”, not a question mark. To specify any number of characters, say ”.*” rather than “*”.

We've added Ajax Search! as a new option on this form. Ajax should allow faster searches for words since the entire page is not retrieved, just the word list that results from your search. Let us know if you have problems. / webmaster@puzzlers.org

See our FAQ Document.

For more powerful searching, you can try our Dictionary Grep/Search - Advanced.

The dictionaries available at the moment are:

• NI2 words – single words from NI2
• NI2 compound forms – compound forms from NI2
• OSPD – the official Scrabble dictionary
• UK cryptics – a dictionary created for British cryptics
• UK cryptics, no accents – the British cryptics list with accented letters removed
• Shorter OED– the Shorter Oxford English Dictionary
• Enable – the Enable Scrabble word list
• Moby words – single words from the Moby list
• Moby compound forms – compound forms from the Moby list
• FILE main list – main dictionary from the FILE project
• FILE to-do list – to-do list from the FILE project
• Unix system dictionary – dictionary included with Unix
• Roger King's dictionary – dictionary obtained from Roger King
• Public Brand Software – word list obtained by Evan Antworth
• Center for Research in Lexicography – a dictionary credited to the CRL
• Carnegie Mellon list – originally a phonetic dictionary, created by CMU
• Roget's Thesaurus list – from a 1913 Roget's Thesaurus
• Orchy list – from orchy.com
• Merriam-Webster 9th Collegiate list – words from the MW 9th collegiate
• 1st Unabridged – 1st unabridged dictionary; no source information
• 2nd Unabridged – 2nd unabridged dictionary; no source information
• Pocket – pocket dictionary; no source information
• 1st anonymous list – anonymous word list; no source information
• 2nd anonymous list – anonymous word list; no source information

There is a page where you can read about and download these dictionaries/wordlists. You should at least look at the background information about these lists before deciding to trust them.

The search string can just be a plain string, such as “gai”, which will return “gaiety”, “assegai”, and “regain”, among other words, but it can also be a regular expression. Regular expressions are powerful tools for searching patterns. I don't have space here for a full explanation of regular expressions, but I hope to provide support for questions about them soon. If you know how to use them, go ahead. If not, here are a few references:

• Daz wrote a ”first draft of a regular expression primer” that he sent to npl-folk.
• A reference page for regular expressions at Cornell that may help you get started
• Lucifer created some examples that refer to specific flat types; you may find them useful.

Single quotes are ignored in the pattern. This is to avoid a security issue. Quotes appear in very few entries in any case, and can still be matched, either by a dot, or by a circumlocution such as [^a-zA-z \-\!], which will match odd characters.

The UK cryptics list, in addition, contains some special characters, namely these: Åàäâåáçéêèîïñöôóûùü. The “no accents” version in the list converts these to their normal plain text equivalents.

The case sensitivity option should be self explanatory. The third choice permits you to logically reverse the sense of the test: if you specify a search string of “e” and “don't match” then you will only get words without an “e” in them.

For more powerful searching, you can try our Dictionary Grep/Search - Advanced.

You can specify the word length to return. The length is tested after any conversions caused by query type, above, so with “consonants only” selected, “tit for tat” will only be found if the word length is set to 6.

The result always shows you how many words match. To speed things up, however, you may sometimes not want to see the actual words; for example if you're trying to determine frequency counts. You can specify “count of matching words” to avoid generating the word list. This is particularly useful if you might otherwise return lists of hundreds of thousands of words.