From cc4776d29da5bf69dc662d1f1eb5696b3a5c40e8 Mon Sep 17 00:00:00 2001
From: Eric House <eehouse@eehouse.org>
Date: Fri, 24 Apr 2020 13:44:55 -0700
Subject: [PATCH] Populate an actual wordlist for Hungarian

Add Makefile filters to create a wordlist with about 42K words derived
from a github project (thanks to pointers from an informant. :-) Per
him, and contrary to how Catalan does it, double-letter-tile words
also appear in single-letter variants if the tiles allow.
---
 xwords4/dawg/Hungarian/Makefile  | 22 +++++++++++++++++++---
 xwords4/dawg/Hungarian/mklist.sh | 11 +++++++++++
 2 files changed, 30 insertions(+), 3 deletions(-)
 create mode 100755 xwords4/dawg/Hungarian/mklist.sh

diff --git a/xwords4/dawg/Hungarian/Makefile b/xwords4/dawg/Hungarian/Makefile
index 03547fbb4..e07355735 100644
--- a/xwords4/dawg/Hungarian/Makefile
+++ b/xwords4/dawg/Hungarian/Makefile
@@ -21,7 +21,7 @@ LANGCODE=hu_HU
 ENC = UTF-8
 TARGET_TYPE ?= WINCE
 
-DICTNOTE = "This wordlist contains the tile information for Hungarian but no words."
+DICTNOTE = "Derived from szotar/alap/ in https://github.com/laszlonemeth/magyarispell.git"
 
 LANG_SPECIAL_INFO = \
 		"CS Cs cS cs" /dev/null /dev/null \
@@ -34,9 +34,25 @@ LANG_SPECIAL_INFO = \
 
 include ../Makefile.langcommon
 
-# Empty dict
+# Filtering wordlist: When a word can be spelled with either a
+# double-letter tile or two single-letter tiles, it's in the list with
+# both spellings. That's what the longer sed expressions are doing
+# (emitting two words)
 $(XWLANG)Main.dict.gz:
-	echo -n "" | gzip -c > $@
+	cat hungarian_wordlist.txt \
+	| tr -d '\r' \
+	| tr [aábcdeéfghiíjklmnnyoóöőprtuúüűvzs] [AÁBCDEÉFGHIÍJKLMNNYOÓÖŐPRTUÚÜŰVZS] \
+	| grep -v '1\|2\|3\|4\|5\|6\|7' \
+	| sed -e 's,^\(.*\)CS\(.*\)$$,\11\2\n\1CS\2,g' \
+	| sed -e 's,GY,2,g' \
+	| sed -e 's,LY,3,g' \
+	| sed -e 's,NY,4,g' \
+	| sed -e 's,^\(.*\)SZ\(.*\)$$,\15\2\n\1SZ\2,g' \
+	| sed -e 's,TY,6,g' \
+	| sed -e 's,^\(.*\)ZS\(.*\)$$,\17\2\n\1ZS\2,g' \
+	| grep '^[1-7AÁBCDEÉFGHIÍJKLMNOÓÖŐPRSTUÚÜŰVZ]*$$' \
+	| tr '1234567' '\001\002\003\004\005\006\007' \
+	| gzip -c > $@
 
 # Everything but creating of the Main.dict file is inherited from the
 # "parent" Makefile.langcommon in the parent directory.
diff --git a/xwords4/dawg/Hungarian/mklist.sh b/xwords4/dawg/Hungarian/mklist.sh
new file mode 100755
index 000000000..7b207a3de
--- /dev/null
+++ b/xwords4/dawg/Hungarian/mklist.sh
@@ -0,0 +1,11 @@
+#!/bin/sh
+
+set -e -u
+
+# from: https://github.com/laszlonemeth/magyarispell.git
+DIR=/home/eehouse/dev/git/magyarispell/szotar/alap
+
+cat ${DIR}/fonev.1 ${DIR}/melleknev.1 ${DIR}/ige_alanyi.1 ${DIR}/ige_targy.1 ${DIR}/ragozatlan.2 |\
+	sed -e 's/#.*$//' -e 's/\[.*$//' -e 's/ .*$//' |\
+	grep -v '^$' |\
+	sort -u > hungarian_wordlist.txt