1 TERMUX_PKG_HOMEPAGE
=https
://github.com
/tesseract-ocr
/tesseract
2 TERMUX_PKG_DESCRIPTION
="Tesseract is probably the most accurate open source OCR engine available"
3 TERMUX_PKG_VERSION
=3.04.01
4 TERMUX_PKG_DEPENDS
="libtool, libuuid, leptonica"
5 TERMUX_PKG_SRCURL
=https
://github.com
/tesseract-ocr
/tesseract
/archive
/${TERMUX_PKG_VERSION}.
tar.gz
6 TERMUX_PKG_SHA256
=57f63e1b14ae04c3932a2683e4be4954a2849e17edd638ffe91bc5a2156adc6a
7 TERMUX_PKG_FOLDERNAME
=tesseract-
${TERMUX_PKG_VERSION}
9 termux_step_pre_configure
() {
10 # http://blog.matt-swain.com/post/26419042500/installing-tesseract-ocr-on-mac-os-x-lion
11 export LIBLEPT_HEADERSDIR
=${TERMUX_PREFIX}/include
/leptonica
14 perl
-p
-i
-e
's|ADD_RT, true|ADD_RT, false|g' configure.ac
18 termux_step_post_make_install
() {
19 # download english trained data
20 cd "${TERMUX_PREFIX}/share/tessdata"
22 for f
in cube.
{bigrams
,fold,lm
,nn
,params
,size
,word-freq
} tesseract_cube.nn traineddata
; do
24 # From the tessdata README: "These language data files only work with
25 # Tesseract 4. They are based on the sources in tesseract-ocr/langdata on GitHub.
26 # Get language data files for Tesseract 3.04 or 3.05 from the 3.04 tree."
27 termux_download https
://raw.githubusercontent.com
/tesseract-ocr
/tessdata
/3.04.00/$f $f