Initial import

2 files changed, 44 insertions, 0 deletions

diff --git a/.SRCINFO b/.SRCINFO
new file mode 100644
index 000000000000..9817a8498789
--- /dev/null
+++ b/.SRCINFO
@@ -0,0 +1,20 @@
+pkgbase = ucto
+	pkgdesc = An advanced rule-based (regular-expression) and unicode-aware tokenizer for various languages. Tokenization is an essential first step in any NLP pipeline.
+	pkgver = 0.8.0
+	pkgrel = 1
+	url = http://ilk.uvt.nl/ucto
+	arch = i686
+	arch = x86_64
+	license = GPL3
+	makedepends = libtool
+	makedepends = autoconf
+	depends = ticcutils>=0.7
+	depends = icu
+	depends = libxml2
+	depends = libfolia>=0.13
+	options = !libtool
+	source = http://software.ticc.uvt.nl/ucto-0.8.0.tar.gz
+	md5sums = 67750bc54be80bdc9bb47acf8e5aa616
+
+pkgname = ucto
+
diff --git a/PKGBUILD b/PKGBUILD
new file mode 100644
index 000000000000..d6e8997fe74a
--- /dev/null
+++ b/PKGBUILD
@@ -0,0 +1,24 @@
+# Maintainer: Maarten van Gompel <proycon at anaproy dot nl>
+pkgname=ucto
+pkgver=0.8.0
+pkgrel=1
+pkgdesc="An advanced rule-based (regular-expression) and unicode-aware tokenizer for various languages. Tokenization is an essential first step in any NLP pipeline."
+arch=('i686' 'x86_64')
+license=('GPL3')
+depends=('ticcutils>=0.7' 'icu' 'libxml2' 'libfolia>=0.13')
+makedepends=('libtool' 'autoconf')
+options=(!libtool)
+url="http://ilk.uvt.nl/ucto"
+source=("http://software.ticc.uvt.nl/$pkgname-$pkgver.tar.gz")
+md5sums=('67750bc54be80bdc9bb47acf8e5aa616')
+
+build() {
+  cd $srcdir/$pkgname-$pkgver
+  ./configure --prefix=/usr --sysconfdir=/etc --localstatedir=/var
+  make
+}
+
+package() {
+  cd $srcdir/$pkgname-$pkgver
+  make DESTDIR=$pkgdir install
+}