Add working examples of full-text-search of SQLite3

[KEINOS]

This PR is a part of #299 which is an example of using Kagome as a tokenizer for full-text-search, a.k.a. FTS4, of SQLite3.

[ikawaha]

sugoi!

[KEINOS]

F.Y.I.

This PR was inspired while reading the below book by @mattn san.

• p.372, 9.2 "データーベース登録プログラム", "Go言語プログラミングエッセンス エンジニア選書"
  • Written by: Mattn
  • Published: 2023/3/9 (技術評論社)
  • ISBN: 4297134195 / 978-4297134198
  • ASIN: B0BVZCJQ4F / https://amazon.co.jp/dp/4297134195

[KEINOS]

I just noticed that the below two query at line 76 and 77 in main.go, the column name words and content should be in reverse for the usage.

_, err = db.Exec(`
        CREATE TABLE IF NOT EXISTS contents_fts(docid INTEGER PRIMARY KEY AUTOINCREMENT, words TEXT);
        CREATE VIRTUAL TABLE IF NOT EXISTS fts USING fts4(content, tokenize=unicode61 "remove_diacritics=2");
    `)

I will 追いPush the changes.

[KEINOS]

Done!

@ikawaha Plz review it when you got free time.

[mattn]

sugoi!

[KEINOS]

@ikawaha Sorry, I should've PRed to the develop branch. Won't happen again. 🙏

[ikawaha]

@KEINOS Never mind, I just forgot to create a develop branch :p. PRs are always welcome!

[ikawaha]

@KEINOS LGTM 👍

It would be better if the example sentences were easy to understand the results of morphological analysis. Please review the PR for https://github.com/KEINOS/Fork_kagome/pull/3 and merge if you would like.

[KEINOS]

@ikawaha

I merged your PR, thank you!

Indeed, it got way better to understand about the analysis. I added some other search words to clarify more. Please merge if it LGTY!

[ikawaha]

Thank you!