From 09411eba33cd0f9cdc819fa00c0f4fb28735e796 Mon Sep 17 00:00:00 2001
From: "Steven R. Loomis" <srl295@gmail.com>
Date: Tue, 31 Oct 2023 17:41:53 -0500
Subject: [PATCH] =?UTF-8?q?feat(common):=20ldml=20update=20to=20WIP=20cldr?=
 =?UTF-8?q?=20data=20=F0=9F=99=80?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- bn-bengali with a basic test
- fr-optimise
- note: disabling bengali reorders for now, not working properly due to tertiary reordering feat(core): tertiary reorders 🙀  #9707

For: #9916
---
 core/tests/unit/ldml/keyboards/meson.build    |   4 +-
 .../ldml-keyboards/techpreview/3.0/bn.xml     | 172 ++++++++++++++
 .../techpreview/3.0/fr-t-k0-optimise-test.xml |  16 ++
 .../techpreview/3.0/fr-t-k0-optimise.xml      | 200 ++++++++++++++++
 .../techpreview/3.0/fr-t-k0-test-test.xml     |  22 ++
 .../techpreview/3.0/fr-t-k0-test.xml          | 213 ++++++++++++++++++
 .../techpreview/test/bn-test.xml              |  26 +++
 7 files changed, 652 insertions(+), 1 deletion(-)
 create mode 100644 resources/standards-data/ldml-keyboards/techpreview/3.0/bn.xml
 create mode 100644 resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise-test.xml
 create mode 100644 resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise.xml
 create mode 100644 resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test-test.xml
 create mode 100644 resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test.xml
 create mode 100644 resources/standards-data/ldml-keyboards/techpreview/test/bn-test.xml

diff --git a/core/tests/unit/ldml/keyboards/meson.build b/core/tests/unit/ldml/keyboards/meson.build
index 89077fabac..f4b7f23a29 100644
--- a/core/tests/unit/ldml/keyboards/meson.build
+++ b/core/tests/unit/ldml/keyboards/meson.build
@@ -10,8 +10,10 @@
 tests_from_cldr = [
   'ja-Latn',
   'pt-t-k0-abnt2',
-  'fr-t-k0-azerty',
+  # 'fr-t-k0-optimise', (not yet)
+  # 'fr-t-k0-test',
   'pcm',
+  'bn',
 ]
 
 tests_without_testdata = [
diff --git a/resources/standards-data/ldml-keyboards/techpreview/3.0/bn.xml b/resources/standards-data/ldml-keyboards/techpreview/3.0/bn.xml
new file mode 100644
index 0000000000..fc04ec3cbc
--- /dev/null
+++ b/resources/standards-data/ldml-keyboards/techpreview/3.0/bn.xml
@@ -0,0 +1,172 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE keyboard3 SYSTEM "../dtd/ldmlKeyboard3.dtd">
+<keyboard3 locale="bn" conformsTo="techpreview">
+    <!--
+        History:
+          Based on
+    https://github.com/keymanapp/ldml-keyboards-dev/blob/master/keyboards/sil-bengali/bn-t-k0-cldr-phonetic.ldml
+     -->
+    <locales>
+        <locale id="as" />
+        <locale id="syl" />
+    </locales>
+    <version number="1.3.0" />
+    <info name="SIL Bengali-Assamese Phonetic" indicator="bn" layout="QWERTY" />
+
+    <keys>
+        <import base="cldr" path="techpreview/keys-Zyyy-punctuation.xml" />
+        <import base="cldr" path="techpreview/keys-Zyyy-currency.xml" />
+
+        <!-- key names based on https://www.loc.gov/catdir/cpso/romanization/bengali.pdf + Assamese for /wa/ -->
+
+        <!-- UNSHIFTED KEYS -->
+        <!-- E: (top) row -->
+        <key id="candrabindu" output="\u{0981}" /> <!-- n̐  -->
+        <key id="1" output="১" />
+        <key id="2" output="২" />
+        <key id="3" output="৩" />
+        <key id="4" output="৪" />
+        <key id="5" output="৫" />
+        <key id="6" output="৬" />
+        <key id="7" output="৭" />
+        <key id="8" output="৮" />
+        <key id="9" output="৯" />
+        <key id="0" output="০" />
+        <!-- hyphen -->
+        <!-- equal -->
+
+        <!-- D: -->
+        <key id="more" output="\m{q}" /> <!-- 'additional characters' -->
+        <key id="wa" output="ৱ" /> <!-- Assamese-->
+        <key id="e" output="\u{09C7}" />
+        <key id="ra" output="র" />
+        <key id="ta" output="ত" />
+        <key id="sha" output="য" />
+        <key id="u" output="\u{09C1}" />
+        <key id="i" output="\u{09BF}" />
+        <key id="o" output="\u{09CB}" />
+        <key id="pa" output="প" />
+
+        <!-- C: -->
+        <key id="ā" output="\u{09BE}" />
+        <key id="sa" output="স" />
+        <key id="da" output="দ" />
+        <key id="ṭa" output="ট" />
+        <key id="ga" output="গ" />
+        <key id="ha" output="হ" />
+        <key id="ja" output="জ" />
+        <key id="ka" output="ক" />
+        <key id="la" output="ল" />
+
+        <!-- B: -->
+        <key id="śa" output="শ" />
+        <key id="hasant" output="\u{09CD}" />
+        <key id="ca" output="চ" />
+        <key id="ḍa" output="ড" />
+        <key id="ba" output="ব" />
+        <key id="na" output="ন" />
+        <key id="ma" output="ম" />
+        <!-- comma -->
+        <key id="dahri" output="।" />
+
+
+        <!-- SHIFTED KEYS -->
+        <key id="ṃ" output="\u{0982}" />
+        <!-- exclam, etc -->
+
+        <!-- gap -->
+        <key id="ña" output="ঞ" />
+        <key id="ai" output="\u{09C8}" />
+        <key id="ṛ" output="\u{09C3}" />
+        <key id="tha" output="থ" />
+        <key id="ẏa" output="য়" /> <!-- Missing in Keyman version of file-->
+        <key id="ū" output="\u{09C2}" />
+        <key id="ī" output="\u{09C0}" />
+        <key id="au" output="\u{09CC}" />
+        <key id="pha" output="ফ" />
+
+        <key id="lengthener" output="\m{A}" />
+        <key id="sha" output="ষ" />
+        <key id="dha" output="ধ" />
+        <key id="ṭha" output="ঠ" />
+        <key id="gha" output="ঘ" />
+        <!-- gap -->
+        <key id="jha" output="ঝ" />
+        <key id="kha" output="খ" />
+        <!-- gap -->
+
+        <!-- gap -->
+        <key id="vis-hasant" output="\m{X}" />
+        <key id="cha" output="ছ" />
+        <key id="ḍha" output="ঢ" />
+        <key id="bha" output="ভ" />
+        <key id="ṇa" output="ণ" />
+        <key id="ṅa" output="ঙ" />
+        <!-- less-than -->
+        <key id="nukta" output="\u{09BC}" />
+        <!-- question -->
+
+
+    </keys>
+    <layers formId="us">
+        <layer modifiers="none">
+            <row keys="candrabindu 1 2 3 4 5 6 7 8 9 0 hyphen equal" />
+            <row keys="more wa e ra ta sha u i o pa open-square close-square backslash" />
+            <row keys="ā sa da ṭa ga ha ja ka la semi-colon apos" />
+            <row keys="śa hasant ca ḍa ba na ma comma dahri slash" />
+            <row keys="space" />
+        </layer>
+        <layer modifiers="shift">
+            <row
+                keys="ṃ bang at hash dollar percent caret amp asterisk open-paren close-paren underscore plus" />
+            <row keys="gap ña ai ṛ tha ẏa ū ī au pha open-curly close-curly pipe" />
+            <row keys="lengthener sha dha ṭha gha gap jha kha gap colon double-quote" />
+            <row keys="gap vis-hasant cha ḍha bha ṇa ṅa open-angle nukta question" />
+            <row keys="space" />
+        </layer>
+    </layers>
+
+    <transforms type="simple">
+        <transformGroup>
+            <transform from="\u{09C7}\m{A}" to="\u{09CC}" /> <!-- E + lengthener = AU -->
+            <!-- <transform from="\u{09C7}\u{09BE}" to="\u09CB" /> --> <!-- E + O = O This is handled by normalization. -->
+
+            <!-- these suport the 'q' key -->
+            <transform from="\m{q}:" to="\u{0983}" />
+            <transform from="\m{q}L" to="ৡ" />
+            <transform from="\m{q}।" to="॥" />
+            <transform from="\m{q}ড" to="ড়" />
+            <transform from="\m{q}ঢ" to="ঢ়" />
+            <transform from="\m{q}ত" to="ৎ" />
+            <transform from="\m{q}য" to="য়" />
+            <transform from="\m{q}র" to="ৰ" />
+            <transform from="\m{q}ল" to="ঌ" />
+            <transform from="\m{q}\u{09BE}" to="অ" />
+            <transform from="\m{q}\u{09BF}" to="ই" />
+            <transform from="\m{q}\u{09C0}" to="ঈ" />
+            <transform from="\m{q}\u{09C1}" to="উ" />
+            <transform from="\m{q}\u{09C2}" to="ঊ" />
+            <transform from="\m{q}\u{09C3}" to="ৠ" />
+            <transform from="\m{q}\u{09C7}" to="এ" />
+            <transform from="\m{q}\u{09C8}" to="ঐ" />
+            <transform from="\m{q}\u{09CB}" to="ও" />
+            <transform from="\m{q}\u{09CC}" to="ঔ" />
+            <transform from="\m{q}\m{A}" to="আ" />
+            <transform from="\m{q}\m{X}" to="\u{09CD}\u{200C}" /> <!-- virama + zwnj-->
+        </transformGroup>
+          <!-- TODO: document these -->
+          <!-- TODO-LDML: not working, probably due to #9707 -->
+          <!-- <transformGroup>
+            <reorder from="\u{09BC}" tertiary="3"/>
+            <reorder from="\u{09CD}[\u{0980}\u{0985}-\u{098C}\u{098F}\u{0990}\u{0993}-\u{09A8}\u{09AA}-\u{09B0}\u{09B2}\u{09B6}-\u{09B9}\u{09BD}\u{09DC}\u{09DD}\u{09DF}-\u{09E1}\u{09E6}-\u{09F1}\u{09FC}]" order="10" tertiaryBase="true"/>
+            <reorder from="\u{09CD}[\u{200C}\u{200D}][\u{0980}\u{0985}-\u{098C}\u{098F}\u{0990}\u{0993}-\u{09A8}\u{09AA}-\u{09B0}\u{09B2}\u{09B6}-\u{09B9}\u{09BD}\u{09DC}\u{09DD}\u{09DF}-\u{09E1}\u{09E6}-\u{09F1}\u{09FC}]" order="10" tertiaryBase="true"/>
+            <reorder from="\u{09CD}" order="120" tertiaryBase="true"/>
+            <reorder from="[\u{09BF}\u{09C7}\u{09C8}]" order="60"/>
+            <reorder from="[\u{09C1}-\u{09C4}\u{09E2}\u{09E3}]" order="70"/>
+            <reorder from="[\u{09BE}\u{09C0}\u{09CB}\u{09CC}]" order="75"/>
+            <reorder from="\u{0981}" order="85"/>
+            <reorder from="[\u{0982}\u{0983}]" order="95"/>
+            <reorder from="\u{09FE}" order="117"/>
+        </transformGroup> -->
+    </transforms>
+</keyboard3>
diff --git a/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise-test.xml b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise-test.xml
new file mode 100644
index 0000000000..593b8bd2e7
--- /dev/null
+++ b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise-test.xml
@@ -0,0 +1,16 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE keyboardTest3 SYSTEM "../dtd/ldmlKeyboardTest3.dtd">
+<keyboardTest3 conformsTo="techpreview">
+    <info keyboard="fr-t-k0-optimise.xml" author="Team Keyboard" name="fr-test" />
+    <repertoire name="simple-repertoire" chars="[a b c d e \u{22}]" type="simple" /> <!-- verify that these outputs are all available from simple keys on any layer, for all form factors -->
+    <repertoire name="chars-repertoire" chars="[á é ó]" type="gesture" /> <!-- verify that these outputs are all available from simple or gesture keys on any layer, for touch -->
+    <tests name="key-tests">
+        <test name="key-test">
+            <startContext to="abc\u0022..."/>
+            <!-- tests by pressing key ids -->
+            <keystroke key="s"/>
+            <check result="abc\u0022...s" />
+            <!-- TODO -->
+        </test>
+    </tests>
+</keyboardTest3>
diff --git a/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise.xml b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise.xml
new file mode 100644
index 0000000000..8db073cc76
--- /dev/null
+++ b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-optimise.xml
@@ -0,0 +1,200 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE keyboard3 SYSTEM "../dtd/ldmlKeyboard3.dtd">
+<!--
+  New French
+-->
+<keyboard3 locale="fr-t-k0-optimise" conformsTo="techpreview">
+	<version number="1.0.0" />
+	<info author="Team Keyboard" name="French AZERTY optimisé" layout="AZERTY" indicator="FR" />
+
+	<displays>
+    <display output="\u0300" display="${grave}" />
+    <display output="\m{acute}" display="´" />
+    <display output="\m{grave}" display="`" />
+    <display output="\m{umlaut}" display="¨" />
+    <display output="\m{caret}" display="^" />
+    <display output="\m{tilde}" display="~" />
+  </displays>
+
+	<keys>
+		<import base="cldr" path="techpreview/keys-Zyyy-punctuation.xml" />
+		<import base="cldr" path="techpreview/keys-Zyyy-currency.xml" />
+
+		<!-- deadkeys -->
+		<key id="mark-acute" output="\m{acute}" />
+		<key id="mark-breve" output="\m{breve}" />
+		<key id="mark-caret" output="\m{caret}" />
+		<key id="mark-caron" output="\m{caron}" />
+		<key id="mark-cedilla" output="\m{cedilla}" />
+		<key id="mark-circumflex" output="\m{circumflex}" />
+		<key id="mark-comma" output="\m{comma}" />
+		<key id="mark-currency" output="\m{currency}" />
+		<key id="mark-dotabove" output="\m{dotabove}" />
+		<key id="mark-dotbelow" output="\m{dotbelow}" />
+		<key id="mark-doubleacute" output="\m{doubleacute}" />
+		<key id="mark-doublegrave" output="\m{doublegrave}" />
+		<key id="mark-euro" output="\m{euro}" />
+		<key id="mark-grave" output="\m{grave}" />
+		<key id="mark-macron" output="\m{macron}" />
+		<key id="mark-ogonek" output="\m{ogonek}" />
+		<key id="mark-ring" output="\m{ring}" />
+		<key id="mark-scientific" output="\m{scientific}" />
+		<key id="mark-solidus" output="\m{solidus}" />
+		<key id="mark-stroke" output="\m{stroke}" />
+		<key id="mark-submacron" output="\m{submacron}" />
+		<key id="mark-tilde" output="\m{tilde}" />
+		<key id="mark-umlaut" output="\m{umlaut}" />
+
+
+		<!-- extra keys -->
+		<key id="e-grave" output="è" />
+		<key id="e-acute" output="é" />
+		<key id="c-cedilla" output="ç" />
+		<key id="a-acute" output="á" />
+		<key id="a-grave" output="à" />
+		<key id="e-caret" output="ê" />
+
+		<!-- extra symbols -->
+		<key id="bullet" output="•" />
+		<key id="umlaut" output="¨" />
+		<key id="sub-2" output="₂" />
+		<key id="super-2" output="²" longPressKeyIds="sub-2" />
+		<key id="en-dash" output="–" />
+		<key id="plus-minus" output="±" />
+		<key id="vulgar-half" output="½" />
+		<key id="vulgar-quarter" output="¼" />
+		<key id="ellipsis" output="…" />
+		<key id="open-apos" output="‘" />
+		<key id="close-apos" output="’" />
+		<key id="open-g" output="«" />
+		<key id="close-g" output="»" />
+
+		<key id="theta" output="θ" /> <!-- "for Romani" -->
+		<key id="Theta" output="ϴ" /> <!-- "for Romani" -->
+		<key id="tm" output="™" />
+		<key id="open-double" output="“" />
+		<key id="close-double" output="’" />
+		<key id="oe" output="œ" />
+		<key id="Oe" output="Œ" />
+		<key id="registered" output="®" />
+		<key id="minus-sign" output="−" />
+		<key id="ss" output="ß" />
+		<key id="ininity" output="∞" />
+		<key id="division" output="÷" />
+		<key id="less-equal" output="≤" />
+		<key id="greater-equal" output="≥" />
+		<key id="copy" output="©" />
+		<key id="inverse-question" output="¿" />
+		<key id="inverse-bang" output="¡" />
+		<key id="middle-dot" output="•" />
+		<key id="A-grave" output="À" />
+		<key id="E-acute" output="É" />
+		<key id="E-caret" output="Ê" />
+		<key id="em-dash" output="—" />
+		<key id="open-single" output="’" />
+		<key id="close-single" output="’" />
+		<key id="u-grave" output="Ù" />
+		<key id="radix" output="√" />
+		<key id="C-cedilla" output="Ç" />
+		<key id="not-equal" output="±" />
+
+		<key id="infinity" output="∞" />
+		<key id="E-grave" output="È" />
+		<key id="U-grave" output="Ù" />
+		<key id="nb-hyphen" output="‑" />
+		<key id="multiplication" output="×" />
+		<key id="ezh" output="ʒ" />
+		<key id="Ezh" output="Ʒ" />
+		<key id="ae" output="æ" />
+		<key id="Ae" output="Æ" />
+		<key id="permille" output="‰" />
+		<key id="dbl-dagger" output="‡" />
+		<key id="dagger" output="†" />
+		<key id="asym-equal" output="≃" />
+		<key id="Ss" output="ẞ" />
+	</keys>
+
+	<layers formId="iso">
+		<layer modifiers="none">
+      <row keys="at a-grave e-acute e-grave e-caret open-paren close-paren open-apos close-apos open-g close-g apos mark-caret" />
+			<row keys="a z e r t y u i o p hyphen plus" />
+			<row keys="q s d f g h j k l m slash asterisk" />
+			<row keys="open-angle w x c v b n period comma colon semi-colon" />
+			<row keys="space" />
+		</layer>
+
+		<layer modifiers="shift">
+			<row keys="hash 1 2 3 4 5 6 7 8 9 0 double-quote mark-umlaut" />
+			<row keys="A Z E R T Y U I O P en-dash plus-minus" />
+			<row keys="Q S D F G H J K L M backslash vulgar-half" />
+			<row keys="close-angle W X C V B N question bang ellipsis equal" />
+			<row keys="space" />
+		</layer>
+
+		<layer modifiers="alt">
+			<row keys="mark-breve section mark-acute mark-grave amp open-square close-square mark-macron underscore open-double close-double degree mark-caron"/>
+			<row keys="ae pound euro registered open-curly close-curly u-grave mark-dotabove oe percent minus-sign dagger"/>
+			<row keys="theta ss dollar mark-currency mark-scientific mark-euro gap mark-solidus pipe infinity division multiplication"/>
+			<row keys="less-equal ezh copy c-cedilla mark-cedilla mark-stroke mark-tilde inverse-question inverse-bang middle-dot asym-equal"/>
+			<row keys="space" />
+		</layer>
+
+		<layer modifiers="alt shift">
+			<row keys="mark-circumflex A-grave E-acute E-grave E-caret mark-doubleacute mark-doublegrave gap em-dash open-single close-single mark-ring gap"/>
+			<row keys="Ae gap gap gap tm gap U-grave mark-dotbelow Oe permille nb-hyphen dbl-dagger"/>
+			<row keys="Theta Ss gap gap gap mark-submacron gap gap gap gap radix vulgar-quarter"/>
+			<row keys="greater-equal Ezh gap C-cedilla mark-ogonek gap gap gap mark-comma gap not-equal"/>
+			<row keys="space" />
+		</layer>
+	</layers>
+
+	<variables>
+		<!-- spacing accents as string variables -->
+		<string id="grave"  value="`" />
+		<string id="caret"  value="^" />
+		<string id="umlaut" value="¨" />
+		<string id="tilde"  value="~" />
+
+		<!-- sets representing vowels and accented vowels.  -->
+		<!-- There's an extra space between the lower and upper case for visual separation -->
+		<!-- but, there are only 10 items in the following four sets -->
+		<set id="vowel"       value="a e i o u  A E I O U" />
+		<set id="graveVowel"  value="à è ì ò ù  À È Ì Ò Ù" />
+		<set id="caretVowel"  value="â ê î ô û  Â Ê Î Ô Û" />
+		<set id="umlautVowel" value="ä ë ï ö ü  Ä Ë Ï Ö Ü" />
+
+		<!-- a set containing all spacing accents -->
+		<set id="spacing_accent" value="${grave} ${caret} ${umlaut} ${tilde}" />
+	</variables>
+
+	<transforms type="simple">
+		<transformGroup>
+			<!-- use sets for those that fit neatly into sets -->
+			<transform from="${grave}($[vowel])"  to="$[1:graveVowel]" />
+			<transform from="${caret}($[vowel])"  to="$[1:caretVowel]" />
+			<transform from="${umlaut}($[vowel])" to="$[1:umlautVowel]" />
+
+			<!-- y also takes umlaut (in this sample) -->
+			<transform from="${umlaut}y" to="ÿ" />
+
+			<!-- tilde is on a subset, not all vowels, + n -->
+			<transform from="${tilde}a" to="ã" />
+			<transform from="${tilde}A" to="Ã" />
+			<transform from="${tilde}n" to="ñ" />
+			<transform from="${tilde}N" to="Ñ" />
+			<transform from="${tilde}o" to="õ" />
+			<transform from="${tilde}O" to="Õ" />
+
+			<!-- accent + space = spacing accent -->
+			<transform from="($[spacing_accent]) " to="$1" />
+		</transformGroup>
+		<transformGroup>
+			<!-- this is a reorder group -->
+			<!-- nod-Lana partial example -->
+			<reorder from="\u1A60" order="127" />
+			<reorder from="\u1A6B" order="42" />
+			<reorder from="[\u1A75-\u1A79]" order="55" />
+			<!-- ... partial example ... -->
+		</transformGroup>
+	</transforms>
+</keyboard3>
diff --git a/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test-test.xml b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test-test.xml
new file mode 100644
index 0000000000..c03ce54362
--- /dev/null
+++ b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test-test.xml
@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE keyboardTest3 SYSTEM "../dtd/ldmlKeyboardTest3.dtd">
+<keyboardTest3 conformsTo="techpreview">
+    <info keyboard="fr-t-k0-test.xml" author="Team Keyboard" name="fr-test" />
+    <repertoire name="simple-repertoire" chars="[a b c d e \u{22}]" type="simple" /> <!-- verify that these outputs are all available from simple keys on any layer, for all form factors -->
+    <repertoire name="chars-repertoire" chars="[á é ó]" type="gesture" /> <!-- verify that these outputs are all available from simple or gesture keys on any layer, for touch -->
+    <tests name="key-tests">
+        <test name="key-test">
+            <startContext to="abc\u{0022}..."/>
+            <!-- tests by pressing key ids -->
+            <keystroke key="s"/>
+            <check result="abc\u{0022}...s" />
+            <keystroke key="t"/>
+            <check result="abc\u{0022}...st" />
+            <keystroke key="u"/>
+            <check result="abc\u{0022}...stu" />
+            <!-- tests by specifying 'to' output char -->
+            <emit to="v"/>
+            <check result="abc\u{0022}...stuv" />
+        </test>
+    </tests>
+</keyboardTest3>
diff --git a/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test.xml b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test.xml
new file mode 100644
index 0000000000..76b35304ef
--- /dev/null
+++ b/resources/standards-data/ldml-keyboards/techpreview/3.0/fr-t-k0-test.xml
@@ -0,0 +1,213 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE keyboard3 SYSTEM "../dtd/ldmlKeyboard3.dtd">
+<!--
+    This file is part of the CLDR Keyboard Technical Preview.
+    This is a sample data file.
+    This file is subject to change.
+    Please see https://cldr.unicode.org/index/keyboard-workgroup for the latest information.
+
+	Also NOTE: this is really a test keyboard.  CLDR-12026 will be for the real new azerty keyboard
+-->
+<keyboard3 locale="fr-t-k0-test" conformsTo="techpreview">
+	<locales>
+		<locale id="br" /> <!-- example of including Breton -->
+	</locales>
+	<!-- 'conformsTo' gives CLDR spec conformance. Distinguishes from prior
+		keyboard formats -->
+	<!-- 'version' element is now optional -->
+	<version number="1.0.0" />
+	<info name="French Test AZERTY" author="Team Keyboard" layout="AZERTY" indicator="FR" />
+
+	<displays>
+		<!-- Note: displays is only used for keycap presentation -->
+		<!-- this example is not required for this keyboard as we use the spacing
+			modifiers -->
+		<display output="\u{0300}" display="${grave}" /> <!-- display combining grave as modifier letter grave ˋ -->
+		<!-- Note: We discussed why the existing displayMap was used for display.
+			rather than adding something closer to the key layout. 1. This way we can
+			tell the renderer what to do. Could be double diacritics, spacing issues,
+			etc. 2. We expect that keys have output. Otherwise the input content is malformed.
+			3. Don't Repeat Yourself : multiple keys (i.e. on different symbol layers)
+			may have the same display. -->
+
+		<!-- Example of a display mapping for a key by id -->
+		<display keyId="symbol" display="@" />
+		<display keyId="numeric" display="123" />
+		<displayOptions baseCharacter="x" />
+	</displays>
+
+	<keys>
+		<import base="cldr" path="techpreview/keys-Zyyy-punctuation.xml" />
+		<import base="cldr" path="techpreview/keys-Zyyy-currency.xml" />
+
+		<!-- switch keys -->
+		<key id="shift" layerId="shift" />
+		<key id="numeric" layerId="numeric" />
+		<key id="symbol" layerId="symbol" />
+		<key id="base" layerId="base" />
+
+		<!--
+			TODO: need discussion
+		<key id="bksp" gap="true" /> -->
+		<key id="extra" gap="true" />
+		<!--
+			TODO: need discussion
+		<key id="enter" to="\u{000A}" />
+		-->
+		<key id="enter" gap="true" />
+
+		<!-- extra keys -->
+		<key id="u-grave" output="ü" />
+		<key id="e-grave" output="é" /> <!-- 2 -->
+		<key id="e-acute" output="è" /> <!-- 7 -->
+		<key id="c-cedilla" output="ç" /> <!-- 9 -->
+		<key id="a-grave" output="à"  />
+		<key id="a-acute" output="á" />
+		<key id="a-caret" output="â" />
+		<key id="a-umlaut" output="ä" />
+		<key id="a-tilde" output="ã" />
+		<key id="a-ring" output="å" />
+		<key id="a-caron" output="ā" />
+
+		<key id="A-grave" output="À" /> <!-- 0 -->
+		<key id="A-acute" output="Á" />
+		<key id="A-caret" output="Â" />
+		<key id="A-umlaut" output="Ä" />
+		<key id="A-tilde" output="Ã" />
+		<key id="A-ring" output="Å" />
+		<key id="A-caron" output="Ā" />
+
+
+		<!-- extra symbols -->
+		<key id="bullet" output="•" />
+		<key id="umlaut" output="¨" />
+		<key id="super-2" output="²" multiTapKeyIds="sub-2 2" /> <!-- taps produce: ² ₂ 2 -->
+		<key id="sub-2" output="₂" />
+
+		<!-- note that a-caret is the default-->
+		<key id="a" flickId="a" output="a"
+			longPressKeyIds="a-grave a-caret a-acute a-umlaut a-tilde a-ring a-caron"
+			longPressDefaultKeyId="a-caret" />
+		<!-- note that A-caret is the default-->
+		<key id="A" flickId="b" output="A"
+			longPressKeyIds="A-grave A-caret A-acute A-umlaut a-tilde A-ring A-caron"
+			longPressDefaultKeyId="A-caret" />
+	</keys>
+
+	<flicks>
+		<flick id="b">
+			<flickSegment directions="nw" keyId="A-grave" />
+			<flickSegment directions="nw se" keyId="A-acute" />
+			<flickSegment directions="e" keyId="A-caron" />
+			<flickSegment directions="s" keyId="numeric" /> <!-- layer shifting flick-->
+		</flick>
+		<flick id="a">
+			<flickSegment directions="nw" keyId="a-grave" />
+			<flickSegment directions="nw se" keyId="a-acute" />
+			<flickSegment directions="e" keyId="a-caron" />
+		</flick>
+	</flicks>
+
+	<layers formId="iso">
+		<!-- in DTD: required if conformsTo ≥ 41 -->
+		<layer modifiers="none">
+			<row
+				keys="super-2 amp e-grave double-quote apos open-paren hyphen e-acute underscore c-cedilla a-acute close-paren equal" />
+			<row keys="a z e r t y u i o p caret dollar" />
+			<row keys="q s d f g h j k l m u-grave asterisk" />
+			<row keys="open-angle w x c v b n comma semi-colon colon bang" />
+			<row keys="space" />
+		</layer>
+
+		<layer modifiers="shift">
+			<row keys="1 2 3 4 5 6 7 8 9 0 degree plus" />
+			<row keys="A Z E R T Y U I O P umlaut pound" />
+			<row keys="Q S D F G H J K L M percent micro" />
+			<row keys="close-angle W X C V B N question period slash section" />
+			<row keys="space" />
+		</layer>
+	</layers>
+
+	<layers formId="touch" minDeviceWidth="150">
+		<!-- optional attribute for min physical device size -->
+		<layer id="base">
+			<row keys="a z e r t y u i o p" />
+			<row keys="q s d f g h j k l m" />
+			<row keys="shift gap w x c v b n gap" /> <!--TODO:  + bksp -->
+			<row keys="numeric extra space enter" />
+		</layer>
+
+		<layer id="shift">
+			<row keys="A Z E R T Y U I O P" />
+			<row keys="Q S D F G H J K L M" />
+			<row keys="base W X C V B N" />  <!--TODO:  + bksp -->
+			<row keys="numeric extra space enter" />
+		</layer>
+
+		<layer id="numeric">
+			<row keys="1 2 3 4 5 6 7 8 9 0" />
+			<row
+				keys="hyphen slash colon semi-colon open-paren close-paren dollar amp at double-quote" />
+			<row keys="symbol period comma question bang double-quote" /> <!--TODO:  + bksp -->
+			<row keys="base extra space enter" />
+		</layer>
+
+		<layer id="symbol">
+			<row
+				keys="open-square close-square open-curly close-curly hash percent caret asterisk plus equal" />
+			<row keys="underscore backslash pipe tilde open-angle close-angle euro pound yen bullet" />
+			<row keys="numeric period comma question bang double-quote" /> <!--TODO:  + bksp -->
+			<row keys="base extra space enter" />
+		</layer>
+	</layers>
+
+	<variables>
+		<!-- spacing accents as string variables -->
+		<string id="grave"  value="`" />
+		<string id="caret"  value="^" />
+		<string id="umlaut" value="¨" />
+		<string id="tilde"  value="~" />
+
+		<!-- sets representing vowels and accented vowels.  -->
+		<!-- There's an extra space between the lower and upper case for visual separation -->
+		<!-- but, there are only 10 items in the following four sets -->
+		<set id="vowel"       value="a e i o u  A E I O U" />
+		<set id="graveVowel"  value="à è ì ò ù  À È Ì Ò Ù" />
+		<set id="caretVowel"  value="â ê î ô û  Â Ê Î Ô Û" />
+		<set id="umlautVowel" value="ä ë ï ö ü  Ä Ë Ï Ö Ü" />
+
+		<!-- a set containing all spacing accents -->
+		<set id="spacing_accent" value="${grave} ${caret} ${umlaut} ${tilde}" />
+	</variables>
+
+	<transforms type="simple">
+		<transformGroup>
+			<!-- use sets for those that fit neatly into sets -->
+			<transform from="${grave}($[vowel])"  to="$[1:graveVowel]" />
+			<transform from="${caret}($[vowel])"  to="$[1:caretVowel]" />
+			<transform from="${umlaut}($[vowel])" to="$[1:umlautVowel]" />
+
+			<!-- y also takes umlaut (in this sample) -->
+			<transform from="${umlaut}y" to="ÿ" />
+
+			<!-- tilde is on a subset, not all vowels, + n -->
+			<transform from="${tilde}a" to="ã" />
+			<transform from="${tilde}A" to="Ã" />
+			<transform from="${tilde}n" to="ñ" />
+			<transform from="${tilde}N" to="Ñ" />
+			<transform from="${tilde}o" to="õ" />
+			<transform from="${tilde}O" to="Õ" />
+
+			<!-- accent + space = spacing accent -->
+			<transform from="($[spacing_accent]) " to="$1" />
+		</transformGroup>
+		<transformGroup>
+			<!-- this is a reorder group -->
+			<!-- nod-Lana partial example -->
+			<reorder from="\u{1A60}" order="127" />
+			<reorder from="\u{1A6B}" order="42" />
+			<reorder from="[\u{1A75}-\u{1A79}]" order="55" />
+			<!-- ... partial example ... -->
+		</transformGroup>
+	</transforms>
+</keyboard3>
diff --git a/resources/standards-data/ldml-keyboards/techpreview/test/bn-test.xml b/resources/standards-data/ldml-keyboards/techpreview/test/bn-test.xml
new file mode 100644
index 0000000000..ba0d7d8430
--- /dev/null
+++ b/resources/standards-data/ldml-keyboards/techpreview/test/bn-test.xml
@@ -0,0 +1,26 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE keyboardTest3 SYSTEM "../dtd/ldmlKeyboardTest3.dtd">
+<keyboardTest3 conformsTo="techpreview">
+    <info keyboard="bn.xml" author="Team Keyboard" name="bn-test" />
+    <tests name="tests">
+        <test name="au">
+            <startContext to=""/>
+            <keystroke key="ka"/>
+            <keystroke key="e"/>
+            <keystroke key="lengthener"/>
+            <check result="কৌ" /> <!-- ka + AU -->
+        </test>
+        <test name="greetings">
+          <startContext to="" />
+          <keystroke key="śa"/>
+          <keystroke key="u"/>
+          <keystroke key="bha"/>
+          <keystroke key="e"/>
+          <keystroke key="ca"/>
+          <keystroke key="hasant"/>
+          <keystroke key="cha"/>
+          <keystroke key="ā"/>
+          <check result="শুভেচ্ছা"/>
+        </test>
+    </tests>
+</keyboardTest3>