Daftar kata bahasa Indonesia - sumber dan lisensi
Indonesian word list - origin and licence
=========================================

Source:    hunspell-id (Indonesian spelling dictionary, version 2022.09.21), hunspell-id Authors
           (Ammar Shadiq, Andika Triwidada, Arno Brevoort, Benitius Brevoort, Kurniadi, Viko Adi Rahmawan,
           Volker Mueller, Yurizal Susanto; https://github.com/shuLhan/hunspell-id),
           as shipped in the LibreOffice dictionaries, commit 32b006a2c22a4ac7e8ed3f03346f7b3d85a970a4:
           https://raw.githubusercontent.com/LibreOffice/dictionaries/32b006a2c22a4ac7e8ed3f03346f7b3d85a970a4/id/id_ID.dic
             SHA-256 775ff17a52801f3d2ee80120952fb44cc9bac6c3cf61740de775e439851a5803
           .../id/id_ID.aff
             SHA-256 c625d5b237a489c452cf1f9c666600103e8093667dff89e7030a24217995dc79
           .../id/README-dict.adoc (here: README-hunspell-id-original.adoc),
           .../id/LICENSE-dict (GNU LGPL version 3 text, here unchanged: LICENSE-dict)
Retrieved: 2026-09-28
Licence:   "GNU Lesser General Public License as published by the Free Software Foundation, either
           version 3 of the License, or (at your option) any later version" (README-dict.adoc, section
           "Lisensi", Copyright (C) 2004-2022 hunspell-id Authors). The LGPLv3 is the GPLv3 plus
           additional permissions, so the data may be passed on under the GPLv3.

Processing (scripts/build-wordlist.mjs, function idFormen, run: node scripts/build-wordlist.mjs id):
- id_ID.aff (FLAG long, ISO-8859-1) is expanded with its prefix and suffix rules: prefixes
  (me-, di-, ber-, ter-, pe-, per-, se-, ke- ...), suffixes (-an, -i, -kan) and circumfixes
  (ke-...-an, pe-...-an, me-...-kan, se-...-nya, ke-...-nya; flag A1 = CIRCUMFIX: prefix and suffix
  only together), NEEDAFFIX A2 (the stem alone is not a word), cross products of prefix and suffix,
  prefixes licensed by a suffix continuation class
- without clitics: the particles -lah/-kah and the pronouns -ku/-mu/-nya (classes l0, n0, nl, o0 and
  the variants -annya/-kanlah ... of a0/i0/k0) are not counted as words of their own (rumahku,
  bukunya, makanlah), except a hand-checked list of lexicalised adverbs (ID_NYA: akhirnya, biasanya,
  misalnya, rupanya, tampaknya ...)
- lower-case words a-z only (proper names and acronyms such as Jakarta or ABRI are written with
  capitals in id_ID.dic; reduplications with hyphen such as anak-anak are left out), 2-20 letters,
  at least one vowel (no abbreviations such as km, dll)
- only forms that occur in wordfreq (id) - wordfreq 3.1.1 has only the "small" Indonesian list,
  so all forms have Zipf >= about 3 (see README_wordfreq.txt)
- proper names that id_ID.dic also lists in lower case removed: country names from CLDR
  (Intl.DisplayNames 'id') and a hand-checked list (ID_NAMEN: indonesia, jawa, islam, amerika ...;
  derived forms such as keislaman stay). id_ID.dic also lists many person, place and brand names in
  lower case (henry, linda, angga, glodok, tiktok); these were found by reviewing the stems of the
  finished list that are not in the LibreOffice Indonesian thesaurus (th_id_ID_v2, used only for this
  review, not in the build) and added to ID_NAMEN
- slurs and the coarsest vulgar words removed together with their forms (ID_RAUS)
Result:    15,288 word forms in <length>.txt.
Wordle:    wordle.txt = forms with exactly 5 letters a-z (like Katla, github.com/pveyes/katla:
           keyboard a-z, "kata valid 5 huruf sesuai KBBI") -> 2,316 words.
Limits:    wordfreq does not know upper and lower case; KBBI words that are also common names
           (haris, salim) stay. Derived forms missing in id_ID.dic (e.g. diproduksi, digelar) are
           not generated.

Licence of the lists in this folder (<length>.txt, wordle.txt): GPL-3.0-or-later
(COPYING-GPL-3.0.txt), because they combine hunspell-id (LGPL-3.0-or-later) with wordfreq data
(CC BY-SA 4.0).