# Maintainer: Sergej Pupykin # Contributor: Andreas Hauser # Modified and included in Slint by Didier Spaier pkgname=tesseract pkgver=4.1.0 pkgrel=1slint slackdesc=\ ( #|-----handy-ruler------------------------------------------------------| "$pkgname (Tesseract Open Source OCR Engine)" "This package contains an OCR engine - libtesseract and a command line" "program - tesseract. Tesseract 4 adds a new neural net (LSTM) based " "OCR engine which is focused on line recognition, but also still" "supports the legacy Tesseract OCR engine of Tesseract 3 which works by" "recognizing character patterns." "Tesseract has unicode (UTF-8) support, and can recognize more than 100" "languages \"out of the box\". It supports various output formats:" "plain text, hOCR (HTML), PDF, invisible-text-only PDF, TSV." "The lead developer is Ray Smith. The maintainer is Zdenko Podobny." ) # url="https://github.com/tesseract-ocr/tesseract" source=(https://github.com/tesseract-ocr/tesseract/archive/$pkgver.tar.gz https://github.com/tesseract-ocr/tessdata/raw/bf82613055ebc6e63d9e3b438a5c234bfd638c93/osd.traineddata tesseract.1.asc) docs=(AUTHORS COPYING LICENSE ChangeLog) build() { cd $SRC/${pkgname}-$pkgver cp -f $SRC/tesseract.1.asc doc/ [ -x configure ] || ./autogen.sh [ -f Makefile ] || ./configure --prefix=/usr make make training make DESTDIR=$PKG install make DESTDIR=$PKG training-install mkdir -p $PKG/usr/share/tessdata install -Dm0644 $SRC/osd.traineddata $PKG/usr/share/tessdata/osd.traineddata mkdir -p $PKG/usr/doc/${pkgname}-$pkgver pandoc -s -f gfm README.md -o $PKG/usr/doc/${pkgname}-$pkgver/README.html }