termux-packages/packages/tesseract/build.sh

29 lines
1.2 KiB
Bash
Raw Normal View History

2016-02-06 13:57:02 +01:00
TERMUX_PKG_HOMEPAGE=https://github.com/tesseract-ocr/tesseract
TERMUX_PKG_DESCRIPTION="Tesseract is probably the most accurate open source OCR engine available"
2016-03-27 13:21:42 +02:00
TERMUX_PKG_VERSION=3.04.01
2016-02-06 13:57:02 +01:00
TERMUX_PKG_DEPENDS="libtool, libuuid, leptonica"
TERMUX_PKG_SRCURL=https://github.com/tesseract-ocr/tesseract/archive/${TERMUX_PKG_VERSION}.tar.gz
TERMUX_PKG_FOLDERNAME=tesseract-${TERMUX_PKG_VERSION}
termux_step_pre_configure() {
2016-03-27 13:21:42 +02:00
# http://blog.matt-swain.com/post/26419042500/installing-tesseract-ocr-on-mac-os-x-lion
2016-02-06 13:57:02 +01:00
export LIBLEPT_HEADERSDIR=${TERMUX_PREFIX}/include/leptonica
cd $TERMUX_PKG_SRCDIR
perl -p -i -e 's|ADD_RT, true|ADD_RT, false|g' configure.ac
./autogen.sh
}
termux_step_post_make_install() {
2016-03-27 13:21:42 +02:00
# download english trained data
2016-02-06 13:57:02 +01:00
cd "${TERMUX_PREFIX}/share/tessdata"
rm -f eng.*
2016-05-31 11:52:40 +02:00
for f in cube.{bigrams,fold,lm,nn,params,size,word-freq} tesseract_cube.nn traineddata; do
f=eng.$f
# From the tessdata README: "These language data files only work with
# Tesseract 4. They are based on the sources in tesseract-ocr/langdata on GitHub.
# Get language data files for Tesseract 3.04 or 3.05 from the 3.04 tree."
termux_download https://raw.githubusercontent.com/tesseract-ocr/tessdata/3.04.00/$f $f
2016-05-31 11:52:40 +02:00
done
2016-02-06 13:57:02 +01:00
}