Created
August 14, 2023 16:04
-
-
Save rkatic/33d6ece5eb37632459bf2a8a5aae7cea to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
const HR = ` | |
# sintaksa: regularni_izraz [dio_za_dodat] | |
# prva grupa u regularnom izrazu je i dio koji se brise | |
PFX (je|ni|ne|ho|naj|pra|is) | |
PFX (ne) i | |
SFX (i|mo|te|u) am | |
SFX (e|i|u|o) a | |
SFX (om|oj|og|im|em) a | |
SFX (ima|emu|oga|ome|omu|oma|oje|ije|jema) a | |
SFX (ea) u | |
SFX (e|emo|ste) u | |
SFX [ai](ma) | |
SFX (ci|aka) ka | |
SFX . ti | |
SFX (h|o|smo|ste|jah|jaše|jasmo|jaste|jahu) ti | |
SFX (jeh|ješe|jesmo|jeste|jehu) iti | |
SFX (la|lo|li|le) ti | |
SFX (a|u|e|i|m) | |
SFX (om|ovi|ova|ove|ima) | |
SFX (ovima) | |
SFX (a|u) o | |
SFX (čiju|čima|či) ko | |
SFX (ala) lo | |
SFX (ima) o | |
SFX (la|lo|li|le) ao | |
SFX (m|a|mo|ju|te) ti | |
SFX (na|ne|nu|no|ni) an | |
SFX (nim|noj|nom|nog|nima|nih) an | |
SFX (noga|nomu|nome) an | |
SFX (ju|ma) | |
SFX (oma|oje) i | |
SFX (če|ci|cima) k | |
SFX (a|mo|te) | |
SFX (gu) ći | |
SFX (žeš|žemo|žete) ći | |
SFX (e) i | |
SFX (žeš|žem|žemo|žete|že|žu) zati | |
` | |
function rootGeneratorFor(definition) { | |
const rules = [] | |
for (const line of definition.split('\n')) { | |
const [cmd, pat, add = ''] = line.trim().split(/\s+/g, 3) | |
if (cmd === 'PFX') { | |
rules.push([true, new RegExp(pat === '.' ? '.' : `^(?:${pat}).`), add]) | |
} | |
if (cmd === 'SFX') { | |
rules.push([false, new RegExp(pat === '.' ? '.' : `.(?:${pat})$`), add]) | |
} | |
} | |
return function * genRoots(term) { | |
// yield term | |
const rightOps = [] | |
const leftOps = [] | |
for (const [isPfx, re, add] of rules) { | |
re.lastIndex = 0 | |
const m = term.match(re) | |
if (!m) continue | |
const rm = (m[1] || '').length | |
if (isPfx) { | |
yield add + term.slice(rm) | |
leftOps.push([rm, add]) | |
} else { | |
yield term.slice(0, term.length - rm) + add | |
rightOps.push([rm, add]) | |
} | |
} | |
for (const [rmLeft, addLeft] of leftOps) { | |
const maxRmRight = term.length - rmLeft - 1 | |
for (const [rmRight, addRight] of rightOps) { | |
if (rmRight <= maxRmRight) { | |
yield `${addLeft}${term.slice(rmLeft, term.length - rmRight)}${addRight}` | |
} | |
} | |
} | |
} | |
} | |
const genHrRoots = rootGeneratorFor(HR) | |
console.log([...genHrRoots('prakodovima')]) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment