TERMUX_PKG_HOMEPAGE=https://github.com/tesseract-ocr/tesseract
TERMUX_PKG_DESCRIPTION="Tesseract is probably the most accurate open source OCR engine available"
-TERMUX_PKG_VERSION=3.04.01
+TERMUX_PKG_VERSION=3.05.00
TERMUX_PKG_DEPENDS="libtool, libuuid, leptonica"
TERMUX_PKG_SRCURL=https://github.com/tesseract-ocr/tesseract/archive/${TERMUX_PKG_VERSION}.tar.gz
+TERMUX_PKG_SHA256=3fe83e06d0f73b39f6e92ed9fc7ccba3ef734877b76aa5ddaaa778fac095d996
TERMUX_PKG_FOLDERNAME=tesseract-${TERMUX_PKG_VERSION}
termux_step_pre_configure() {
export LIBLEPT_HEADERSDIR=${TERMUX_PREFIX}/include/leptonica
cd $TERMUX_PKG_SRCDIR
- perl -p -i -e 's|ADD_RT, true|ADD_RT, false|g' configure.ac
+ perl -p -i -e 's|ADD_RT], true|ADD_RT], false|g' configure.ac
./autogen.sh
}
rm -f eng.*
for f in cube.{bigrams,fold,lm,nn,params,size,word-freq} tesseract_cube.nn traineddata; do
f=eng.$f
- termux_download https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/$f $f
+ # From the tessdata README: "These language data files only work with
+ # Tesseract 4. They are based on the sources in tesseract-ocr/langdata on GitHub.
+ # Get language data files for Tesseract 3.04 or 3.05 from the 3.04 tree."
+ termux_download \
+ https://raw.githubusercontent.com/tesseract-ocr/tessdata/3.04.00/$f \
+ $f
done
}