Some words with the letter Y appear under the letter I instead.

Examples:
NYET appears in between NIESHOUT and NIETZSCHE

ABIRRITATIVE is followed by
ABYS
ABYSM
ABYSMAL
[ etc ]
before continuing with
ABYSSUS
ABISTON
ABIT

XXV -> XXX -> Z -> ZA -> ZABAEAN

Words like YACHT, YACCA, YARTH etc all appear under I

Did you perhaps use the Frisian alphabet when sorting the list?

WHY DOES IT START WITH 2? 2 IS NOT A WORD. MY FRIEND AND I HAVE BEEN DRIVEN INSANE THESE LAST FEW DAYS TRYING TO FIGURE OUT A DICTIONARY DEFINITION OF 2 THAT WOULD PERMIT IT TO GO ON THIS LIST. PLEASE PLEASE PLEASE ANSWER QUICKLY. WE HAVE SUNK FAR TOO MUCH TIME INTO THIS AND WE NEED TO KNOW.

The example for words_dictionary.json is read_english_dictionary.py, but that file uses words_alpha.txt

According to merriam-webster, "Megafauna" is a word that is not in here at all, and should be added.

I noticed abbreviations and proper nouns are all lowercased unfortunately. Eg: iso.

There are words like "isn" "aren" "wouldn" - smells like a bug?

Hi! Thank you so much for this. It is such a huge help. Just want to know how to properly credit this to them or to you.

The word "cryptocurrency" exists in the English language (according to Merriam Webster dictionary) and is not in words_alpha.txt. This word should be added as it is a common word now.

Inital is in words_alpha.txt, which is incorrectly spelled. words_alpha.txt:inital

https://www.merriam-webster.com/dictionary/inital
https://www.merriam-webster.com/dictionary/initial
https://en.wiktionary.org/wiki/Special:Search?search=inital&go=Go&ns0=1

This should be fixed.

words.txt has 2 blank lines (#130 000 & #325 001)

Just a nit. The file has DOS line endings (Carriage Return [like typewriter] and Line Feed). You can save 370098-bytes by removing the Carriage Returns (e.g. '\r') leaving only '\n'. (you would also need to add a '\n' terminating the final line). This would make the file POSIX compliant (and it would still work just as well in windows for everything other than viewing in Notepad...) Either way works, it's just odd that such a file has DOS line endings.

There's dinos and dinosaur but no dino

jean-christophe
jean-pierre

Hello! I just need a big list of sentences, (1000+ words), and preferably imperative sentences, but both statements and questions are Ok. Thank you

beginning with words starting with 'i', there is a mixture of words starting with i and y. Did not detect any other errors.

All words make up one giant line without even spaces between each other.

The list suggests that the words elect. and mineral. are abbreviations, though I cannot verify this. There are also a lot of very long jargon abbreviations like anthrop. and anthropol. and palaeontol. It seems that many of these are borrowed from biology/medical and geology jargons. I'm not sure if they belong.

I see that in the file words_alpha there are the (wrong) words:
isnt
arent
wouldnt

and that these (right) words are not included:
isn't
aren't
wouldn't

Is this intentional?

Words starting with 'Y' are right after 'H' and mixed with 'I' for some reason.

Could make it more clear??

It's very handy to have this list.
It does have a few non-words in it, possibly due to OCR errors. A couple I caught were

• brainwashjng
• neritjc

As a person who is new to the DWYL Org/Community 🆕
I need to know how to contribute to the project effectively 💭
so that I can start my journey towards Doing What I Love with my Life! ❤️ ✅ 😂

Markdown:

_**Please read** our_
[**contributing guide**](https://github.com/dwyl/contributing)
(_thank you_!)

Note: these are line-separated but in the actual rendered page it's all one line.
see: https://github.com/dwyl/contributing/blob/master/CONTRIBUTING.md

@iteles we are getting quite a few PRs for this repo which I don't want to "reject",
but rather I want people to discuss their proposed "improvements" before submitting ...

I was scrolling through the list and I found "Ultra-englishUltra-french". I suppose this is not a word?

I have a test table open using Workbench, there is no way I can copy/paste the .txt file and auto populate the table?

The word list was originally used in a proof-of-concept auto-completion mini-project: https://github.com/nelsonic/ac but it's since been used by quite a few people.
Please give examples of how you are using the list in your project(s) so we can add them as suggestions in the readme. thanks!

So I just did a little bit of fooling around: I took the words_alpha.txt file and converted it into a minimal trie that can be JSON-serialized.

The result looks a bit like this:

To clarify: "_":1 marks a node that actually represents a word - there are nodes that are part of a larger word but do not represent words themselves - "antidisestablishmentarianism" is a word in this trie, but "antidisestablishmentarianis" (missing the last "m") is not.

Because of the verbosity of JSON, the resulting file is much larger than the original text file, but because it's so repetitive it compresses to a slightly smaller file: (could be relevant when served over http with gzipping enabled)

Combined with a very simple function we can really quickly check if a word is in the dictionary now:

// Yes, it's JavaScript, but it's not that hard to translate to Python code or other languages, right?
function inDictionary(word) {
   let node = words_trie;
   for (const char of word) {
      let nextNode = node[char];
      if (!nextNode) return false;
      node = nextNode;
   }
   return node._ === 1;
}

Here is the compressed file:
words_alpha_trie.tar.gz

Observable notebook where I did it here, plus a demo of the example function to test if a word is in the dictionary: https://observablehq.com/@jobleonard/efficiently-checking-if-a-word-is-in-a-dictionary

If there are no underscores in the regular words.txt file, then I think this approach should work there as well.

I thought it could be useful because, as stated, the compressed file is smaller than the text file, and in certain situations the trie is more interesting than the plain text file - so it might be more efficient to directly build it then

EDIT: I got a private email asking if it was OK to use this file. So in case anyone else is curious: since it's based on the words file in this repo, I'll just say it's released under the same unlicense license, and you're free to do whatever with it :)

Please remove "zoogrpahy" from the list since it's not a word.

words.txt doesn't contain lowercased "the"?

the lower case version don't exist in the text file.

Hi i just wanted to know if you can make a clean version of the file removing curses and such

Pls help me with python code

How adequate is this list to perform a letter frequency analysis? My only concern is the number of words . Can you provide me with a relative size comparison?

The word 'acceleratorh' is the word list, but I couldn't find that in any dictionary and am skeptical it is an actual word.

I want to use this English words list for my dictionary app, but I only want the word in its base form (no -ing, plural, past tense... form). Which file should I use?

I am thankful to you (owners and contributors) for this repo. I had made an application that using the words.txt database. The reason of submitting a PR, because I wanted to contribute something back. However I did not start right, because I had not read the contributing guidelines. My apology to all (owners and contributors) in this matter.

Recently I have made a repo for introducing my app LWW (Looking for Words within a Word). I wanted to submit a script in Python beforehand that create a simple result list of words out of a word using the words.txt database.

Since I had read the contributing guidelines, i wanted to start right and re-propose LWW to be contributed to this repo. However i had already created LWW repo in my own GitHub. Can you please advice me how to contribute LWW in this case?

I will explain the usage of LWW. This application is searching words in words.txt that will be cross-check with a word that has been given. It will compile and collected words in the words.txt that contain alphabets of the word (that has been given/typed). This app was made for fun, educational, and ideas for making words game.

Please visit my repo https://www.githubs.com/kakkarja/LWW for further understanding of the apps usage.

Once again, thank you for this repo. May you have all the success and wonderful time in your journey of life. Thank You 👍.

Could you create separate files for the types of words listed above?

In the following words:
he"ll
it"ll
hain"t

In your large list of English words, you start with 2, thus sending my friend and I on an absurd journey into linguistics and mathematics, trying to figure out why 2 is a word. There are no other single numbers on the list. Is it a title or tag for the list without formatting? Is it a necessary aspect in coding languages? Is it a mistake? Please get back to me ASAP so we can understand why a list of words begins with 2 and get some sleep. Thank you.

The file while downloading from Raw ends up being blocked as it mentions translating from Somali ?

It offends me deeply that "Motoki Shintani" is an error. It's not all Japanese that are emperialists.

Two errors I found (at least) in wordsalpha.txt and words_dictionary.json:

The two words "preinferred" and "preinferring" are combined into "preinferredpreinferring" which I suspect is not actually a word.
Similarly, they contain the word "snailfishessnailflower". "snailflower" is included as well, so in this case "snailfishes" just has an unnecessary suffix.

While working with words_alpha.txt in Python 3.6, I have faced slow loading and slow searching. I think there should be JSON formatted dictionary for better searching performance of all words.

Example :

# cat words_alpha.txt|grep ^ned                                        
ned
nedder
neddy
neddies
nederlands
# cat words.txt|grep ^ned
nedder
neddies
#

The documentation states that words_alpha.txt is a subset from words.txt which apparently is not the case as of now.

Clearly, "words_alpha.txt" is missing a lot of words from "words.txt" that only have letters, no numbers or symbols.
Why is that?
How was "words_alpha.txt" generated?

Is there any reason why the content of words_dictionary.json isn't a JSON array ?

For e.g instead of it being
{ "aalii": 1,
"aaliis": 1,
"aals": 1 } it could be ["aalii", "aaliis", "aals"]