search

tani / textlint-rule-spelling

A textlint rule for spellings of languages as much as possible
GNU General Public License v3.0
8 stars 5 forks source link

"Spelling" splits words arbitrarily to flag errors #6

Open edwardwoodward opened 5 years ago

edwardwoodward commented 5 years ago

Using this rule to check text files, I get lots of "errors" for correctly spelled words because "Spelling" is dividing words into 2 parts at arbitrary points. For example: "using" is flagged as "usi" and "ng" The same word can be split in different ways, for example "option" is flagged as "opti" in one location and as "optio" in another. Is there any reason for this?

tani commented 5 years ago

textline-rule-spelling depends on nspell to find misspelled words. So this library doesn't contain such routines to define what word is. You should doubt the bug of nspell because of that reason. Could you try behavior of nspell, or show me the example code for this issue. Thanks! https://github.com/fgborges/textlint-rule-spelling/blob/af32dff5268c9cf05fb50c78045a6d6382ebd096/src/index.js#L46

tani commented 5 years ago

Sorry, I'm wrong. The definition of what word is is here. https://github.com/fgborges/textlint-rule-spelling/blob/af32dff5268c9cf05fb50c78045a6d6382ebd096/src/index.js#L34

edwardwoodward commented 5 years ago

Thanks for the replies. As I said on the other thread, I have no programming experience, so, sorry, I can't help with anything to do with coding. I've looked at the explanation of regular expressions in JavaScript, but still don't understand what Line 34 means. I am checking textlint to see if it can help our non-native English speakers when handling English manuals and other documents. If it is helpful, I will continue adding issues to this and other rules.