From 95bc6ba0ca5056274ccc81608919de22032084ad Mon Sep 17 00:00:00 2001
From: Arttu <Blizzara@users.noreply.github.com>
Date: Wed, 3 Jul 2024 19:49:38 +0200
Subject: [PATCH] feat: add "initcap" function (#656)

Initcap is like
"[title](https://github.com/Blizzara/substrait/blob/70d1eb71623ca0754157dd5d87348bae51d420c4/extensions/functions_string.yaml#L1023)",
but while "title" defines articles
 are not to be capitalized, initcap capitalizes also articles.

Initcap is supported by basically all DB systems, including Spark and
DataFusion
---
 extensions/functions_string.yaml | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/extensions/functions_string.yaml b/extensions/functions_string.yaml
index 2e0f174db..3acdb48ae 100644
--- a/extensions/functions_string.yaml
+++ b/extensions/functions_string.yaml
@@ -1048,6 +1048,34 @@ scalar_functions:
           char_set:
             values: [ UTF8, ASCII_ONLY ]
         return: "fixedchar<L1>"
+  -
+    name: initcap
+    description: >-
+      Capitalizes the first character of each word in the input string, including articles,
+      and lowercases the rest. Implementation should follow the utf8_unicode_ci collations
+      according to the Unicode Collation Algorithm described at http://www.unicode.org/reports/tr10/.
+    impls:
+      - args:
+          - value: "string"
+            name: "input"
+        options:
+          char_set:
+            values: [ UTF8, ASCII_ONLY ]
+        return: "string"
+      - args:
+          - value: "varchar<L1>"
+            name: "input"
+        options:
+          char_set:
+            values: [ UTF8, ASCII_ONLY ]
+        return: "varchar<L1>"
+      - args:
+          - value: "fixedchar<L1>"
+            name: "input"
+        options:
+          char_set:
+            values: [ UTF8, ASCII_ONLY ]
+        return: "fixedchar<L1>"
   -
     name: char_length
     description: >-