From 16a23fa9628ed4248a21a2d1f9d0093ee45d893f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=BCrgen=20Buchm=C3=BCller?= <pullmoll@t-online.de>
Date: Mon, 20 Feb 2017 20:50:14 +0100
Subject: [PATCH] tesseract-ocr: update to 3.05.00

---
 srcpkgs/tesseract-ocr/template | 93 ++++++++++++++++------------------
 1 file changed, 45 insertions(+), 48 deletions(-)

diff --git a/srcpkgs/tesseract-ocr/template b/srcpkgs/tesseract-ocr/template
index b61afb49cb1..85178fbd84f 100644
--- a/srcpkgs/tesseract-ocr/template
+++ b/srcpkgs/tesseract-ocr/template
@@ -1,13 +1,13 @@
 # Template file for 'tesseract-ocr'
 pkgname=tesseract-ocr
-version=3.04.01
-revision=2
+version=3.05.00
+revision=1
 wrksrc=tesseract-${version}
-_tessdata_ver=6c9657715d38cb44aea9135605860b1b61b0e275
+_tessdata_ver=8bf2e7ad08db9ca174ae2b0b3a7498c9f1f71d40
 build_style=gnu-configure
 configure_args="LIBLEPT_HEADERSDIR=${XBPS_CROSS_BASE}/usr/include"
-hostmakedepends="automake libtool leptonica"
-makedepends="pango-devel leptonica-devel"
+hostmakedepends="automake libtool pkg-config leptonica"
+makedepends="cairo-devel pango-devel leptonica-devel icu-devel"
 short_desc="Tesseract Open Source OCR engine"
 maintainer="Jürgen Buchmüller <pullmoll@t-online.de>"
 license="Apache-2.0"
@@ -16,8 +16,8 @@ distfiles="
  https://github.com/tesseract-ocr/tesseract/archive/${version}.tar.gz>${pkgname}-${version}.tar.gz
  https://github.com/tesseract-ocr/tessdata/archive/${_tessdata_ver}.tar.gz>tessdata-${_tessdata_ver}.tar.gz"
 checksum="
- 57f63e1b14ae04c3932a2683e4be4954a2849e17edd638ffe91bc5a2156adc6a
- 9da120aa47fc5f2780fb3bf6a346b502607e2b087d1f6857c6d55d5d003ec567"
+ 3fe83e06d0f73b39f6e92ed9fc7ccba3ef734877b76aa5ddaaa778fac095d996
+ fd5cfdcada2a83aea8c855fbf916e1b34882659f796c10449c78d1385264d5a2"
 
 # Create a package for one specific language $1
 pkg_lang() {
@@ -35,26 +35,23 @@ post_extract() {
 }
 pre_configure() {
 	cd ${wrksrc}
-	# Some include paths are not set in training/Makefile.am ?
-	sed -i training/Makefile.am \
-		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/glib-2.0 \\\\" \
-		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/lib/glib-2.0/include \\\\" \
-		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/pango-1.0 \\\\" \
-		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/cairo \\\\" \
-		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/freetype2 \\\\"
-	./autogen.sh
-	case "$XBPS_TARGET_MACHINE" in
-		*-musl) # Don't re-define off_t for musl libc
-			sed -i ${wrksrc}/ccutil/scanutils.cpp \
-				-e "/typedef long off_t;/d"
-	esac
+	# Some include paths are not set in training/Makefile.am
+#	sed -i training/Makefile.am \
+#		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/glib-2.0 \\\\" \
+#		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/lib/glib-2.0/include \\\\" \
+#		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/pango-1.0 \\\\" \
+#		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/cairo \\\\" \
+#		-e "/AM_CPPFLAGS += \\\\/a -I${XBPS_CROSS_BASE}/usr/include/freetype2 \\\\"
+	NOCONFIGURE=1 ./autogen.sh
+}
+post_build() {
+	make ${makejobs} training
 }
 post_install() {
 	# Rename binary to avoid conflict with tesseract package
 	mv ${DESTDIR}/usr/bin/tesseract{,-ocr}
 	mv ${DESTDIR}/usr/share/man/man1/tesseract{,-ocr}.1
 	vdoc ChangeLog
-	vdoc ReleaseNotes
 	vdoc README.md
 	vdoc testing/eurotext.tif
 	vdoc testing/phototest.tif
@@ -65,33 +62,33 @@ tesseract-ocr-tools_package() {
 	short_desc+=" - Tools (development)"
 	pkg_install() {
 		local f
-		make -C training DESTDIR=${DESTDIR} install
-		vmkdir usr/share/tesseract
-		vmkdir usr/share/man/man1
-		vmkdir usr/share/man/man5
-		# Copy shell scripts
-		for f in language-specific.sh tesstrain.sh tesstrain_utils.sh; do
-			if [ -e ${wrksrc}/training/${f} ]; then
-				cp -a ${wrksrc}/training/${f} \
-					${PKGDESTDIR}/usr/share/tesseract
-			fi
-		done
-		# Move binaries and manual pages
-		for f in ambiguous_words classifier_tester cntraining combine_tessdata \
-			dawg2wordlist mftraining set_unicharset_properties \
-			shapeclustering text2image unicharambigs unicharset \
-			unicharset_extractor wordlist2dawg; do
-			if [ -e ${DESTDIR}/usr/bin/${f} ]; then
-				mv ${DESTDIR}/usr/bin/${f} \
-					${PKGDESTDIR}/usr/share/tesseract
-			fi
-			for n in 1 5; do
-				if [ -e ${DESTDIR}/usr/share/man/man${n}/${f}.${n} ]; then
-					mv ${DESTDIR}/usr/share/man/man${n}/${f}.${n} \
-						${PKGDESTDIR}/usr/share/man/man${n}
-				fi
-			done
-		done
+		make DESTDIR=${PKGDESTDIR} training-install
+#		vmkdir usr/share/tesseract
+#		vmkdir usr/share/man/man1
+#		vmkdir usr/share/man/man5
+#		# Copy shell scripts
+#		for f in language-specific.sh tesstrain.sh tesstrain_utils.sh; do
+#			if [ -e ${wrksrc}/training/${f} ]; then
+#				cp -a ${wrksrc}/training/${f} \
+#					${PKGDESTDIR}/usr/share/tesseract
+#			fi
+#		done
+#		# Move binaries and manual pages
+#		for f in ambiguous_words classifier_tester cntraining combine_tessdata \
+#			dawg2wordlist mftraining set_unicharset_properties \
+#			shapeclustering text2image unicharambigs unicharset \
+#			unicharset_extractor wordlist2dawg; do
+#			if [ -e ${DESTDIR}/usr/bin/${f} ]; then
+#				mv ${DESTDIR}/usr/bin/${f} \
+#					${PKGDESTDIR}/usr/share/tesseract
+#			fi
+#			for n in 1 5; do
+#				if [ -e ${DESTDIR}/usr/share/man/man${n}/${f}.${n} ]; then
+#					mv ${DESTDIR}/usr/share/man/man${n}/${f}.${n} \
+#						${PKGDESTDIR}/usr/share/man/man${n}
+#				fi
+#			done
+#		done
 	}
 }
 tesseract-ocr-devel_package() {