Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Upgraded tesseract to version 3.04.01. Created separate langage package called tesseract-data #328

Closed
wants to merge 1 commit into from
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions graphics/tesseract-data/DESCR
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
Language files for tesseract.
26 changes: 26 additions & 0 deletions graphics/tesseract-data/Makefile
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
# $NetBSD: Makefile,v 1.16 2015/10/07 11:26:22 fhajny Exp $
PKGNAME= tesseract-data-3.04.00
DISTNAME= tessdata-${PKGVERSION_NOREV}
PKGREVISION= 1
CATEGORIES= graphics

DISTFILES= ${DISTNAME}.tar.gz


MASTER_SITES= ${MASTER_SITE_GITHUB:=tesseract-ocr/}
GITHUB_PROJECT=tessdata
GITHUB_TAG= ${PKGVERSION_NOREV}

MAINTAINER= [email protected]
HOMEPAGE= https://github.com/tesseract-ocr/tessdata
COMMENT= Commercial quality open source OCR engine
LICENSE= apache-2.0

WRKSRC= ${WRKDIR}/tessdata-${PKGVERSION_NOREV}
NO_BUILD= yes

do-install:
mkdir -p ${DESTDIR}/opt/local/share/tessdata/
mv ${WRKDIR}/${DISTNAME}/* ${DESTDIR}/opt/local/share/tessdata/

.include "../../mk/bsd.pkg.mk"
159 changes: 159 additions & 0 deletions graphics/tesseract-data/PLIST
Original file line number Diff line number Diff line change
@@ -0,0 +1,159 @@
@comment $NetBSD$
share/tessdata/afr.traineddata
share/tessdata/amh.traineddata
share/tessdata/ara.cube.bigrams
share/tessdata/ara.cube.fold
share/tessdata/ara.cube.lm
share/tessdata/ara.cube.nn
share/tessdata/ara.cube.params
share/tessdata/ara.cube.size
share/tessdata/ara.cube.word-freq
share/tessdata/ara.traineddata
share/tessdata/asm.traineddata
share/tessdata/aze.traineddata
share/tessdata/aze_cyrl.traineddata
share/tessdata/bel.traineddata
share/tessdata/ben.traineddata
share/tessdata/bod.traineddata
share/tessdata/bos.traineddata
share/tessdata/bul.traineddata
share/tessdata/cat.traineddata
share/tessdata/ceb.traineddata
share/tessdata/ces.traineddata
share/tessdata/chi_sim.traineddata
share/tessdata/chi_tra.traineddata
share/tessdata/chr.traineddata
share/tessdata/cym.traineddata
share/tessdata/dan.traineddata
share/tessdata/dan_frak.traineddata
share/tessdata/deu.traineddata
share/tessdata/deu_frak.traineddata
share/tessdata/dzo.traineddata
share/tessdata/ell.traineddata
share/tessdata/eng.cube.bigrams
share/tessdata/eng.cube.fold
share/tessdata/eng.cube.lm
share/tessdata/eng.cube.nn
share/tessdata/eng.cube.params
share/tessdata/eng.cube.size
share/tessdata/eng.cube.word-freq
share/tessdata/eng.tesseract_cube.nn
share/tessdata/eng.traineddata
share/tessdata/enm.traineddata
share/tessdata/epo.traineddata
share/tessdata/equ.traineddata
share/tessdata/est.traineddata
share/tessdata/eus.traineddata
share/tessdata/fas.traineddata
share/tessdata/fin.traineddata
share/tessdata/fra.cube.bigrams
share/tessdata/fra.cube.fold
share/tessdata/fra.cube.lm
share/tessdata/fra.cube.nn
share/tessdata/fra.cube.params
share/tessdata/fra.cube.size
share/tessdata/fra.cube.word-freq
share/tessdata/fra.tesseract_cube.nn
share/tessdata/fra.traineddata
share/tessdata/frk.traineddata
share/tessdata/frm.traineddata
share/tessdata/gle.traineddata
share/tessdata/glg.traineddata
share/tessdata/grc.traineddata
share/tessdata/guj.traineddata
share/tessdata/hat.traineddata
share/tessdata/heb.traineddata
share/tessdata/hin.cube.bigrams
share/tessdata/hin.cube.fold
share/tessdata/hin.cube.lm
share/tessdata/hin.cube.nn
share/tessdata/hin.cube.params
share/tessdata/hin.cube.word-freq
share/tessdata/hin.tesseract_cube.nn
share/tessdata/hin.traineddata
share/tessdata/hrv.traineddata
share/tessdata/hun.traineddata
share/tessdata/iku.traineddata
share/tessdata/ind.traineddata
share/tessdata/isl.traineddata
share/tessdata/ita.cube.bigrams
share/tessdata/ita.cube.fold
share/tessdata/ita.cube.lm
share/tessdata/ita.cube.nn
share/tessdata/ita.cube.params
share/tessdata/ita.cube.size
share/tessdata/ita.cube.word-freq
share/tessdata/ita.tesseract_cube.nn
share/tessdata/ita.traineddata
share/tessdata/ita_old.traineddata
share/tessdata/jav.traineddata
share/tessdata/jpn.traineddata
share/tessdata/kan.traineddata
share/tessdata/kat.traineddata
share/tessdata/kat_old.traineddata
share/tessdata/kaz.traineddata
share/tessdata/khm.traineddata
share/tessdata/kir.traineddata
share/tessdata/kor.traineddata
share/tessdata/kur.traineddata
share/tessdata/lao.traineddata
share/tessdata/lat.traineddata
share/tessdata/lav.traineddata
share/tessdata/lit.traineddata
share/tessdata/mal.traineddata
share/tessdata/mar.traineddata
share/tessdata/mkd.traineddata
share/tessdata/mlt.traineddata
share/tessdata/msa.traineddata
share/tessdata/mya.traineddata
share/tessdata/nep.traineddata
share/tessdata/nld.traineddata
share/tessdata/nor.traineddata
share/tessdata/ori.traineddata
share/tessdata/osd.traineddata
share/tessdata/pan.traineddata
share/tessdata/pol.traineddata
share/tessdata/por.traineddata
share/tessdata/pus.traineddata
share/tessdata/ron.traineddata
share/tessdata/rus.cube.fold
share/tessdata/rus.cube.lm
share/tessdata/rus.cube.nn
share/tessdata/rus.cube.params
share/tessdata/rus.cube.size
share/tessdata/rus.cube.word-freq
share/tessdata/rus.traineddata
share/tessdata/san.traineddata
share/tessdata/sin.traineddata
share/tessdata/slk.traineddata
share/tessdata/slk_frak.traineddata
share/tessdata/slv.traineddata
share/tessdata/spa.cube.bigrams
share/tessdata/spa.cube.fold
share/tessdata/spa.cube.lm
share/tessdata/spa.cube.nn
share/tessdata/spa.cube.params
share/tessdata/spa.cube.size
share/tessdata/spa.cube.word-freq
share/tessdata/spa.traineddata
share/tessdata/spa_old.traineddata
share/tessdata/sqi.traineddata
share/tessdata/srp.traineddata
share/tessdata/srp_latn.traineddata
share/tessdata/swa.traineddata
share/tessdata/swe.traineddata
share/tessdata/syr.traineddata
share/tessdata/tam.traineddata
share/tessdata/tel.traineddata
share/tessdata/tgk.traineddata
share/tessdata/tgl.traineddata
share/tessdata/tha.traineddata
share/tessdata/tir.traineddata
share/tessdata/tur.traineddata
share/tessdata/uig.traineddata
share/tessdata/ukr.traineddata
share/tessdata/urd.traineddata
share/tessdata/uzb.traineddata
share/tessdata/uzb_cyrl.traineddata
share/tessdata/vie.traineddata
share/tessdata/yid.traineddata
6 changes: 6 additions & 0 deletions graphics/tesseract-data/distinfo
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
$NetBSD: distinfo,v 1.12 2015/11/03 21:34:21 agc Exp $

SHA1 (tessdata-3.04.00.tar.gz) = 6ea24cccf0e823da98589ccc75d51f0950618236
RMD160 (tessdata-3.04.00.tar.gz) = 0a3c3b3c127b6031e2e037d78e3a6f159fb9e869
SHA512 (tessdata-3.04.00.tar.gz) = 4fbb66137c729e16c7a9e35b09916a45c1bb5ec5a7002a22647e0b10975362cb44c6d6c0c997baf25866f78749ec2d4a86317ec3fb664bd963243e230516d162
Size (tessdata-3.04.00.tar.gz) = 499088801 bytes
33 changes: 14 additions & 19 deletions graphics/tesseract/Makefile
Original file line number Diff line number Diff line change
@@ -1,14 +1,17 @@
# $NetBSD: Makefile,v 1.17 2016/01/06 10:46:53 adam Exp $

DISTNAME= tesseract-ocr-3.02.02
PKGNAME= ${DISTNAME:S/-ocr//}
PKGREVISION= 2
# $NetBSD: Makefile,v 1.16 2015/10/07 11:26:22 fhajny Exp $
DISTNAME= tesseract-3.04.01
PKGREVISION= 1
CATEGORIES= graphics
MASTER_SITES= https://tesseract-ocr.googlecode.com/files/
DISTFILES+= ${DISTNAME}.tar.gz

DISTFILES= ${DISTNAME}.tar.gz


MASTER_SITES= ${MASTER_SITE_GITHUB:=tesseract-ocr/}
GITHUB_PROJECT=tesseract
GITHUB_TAG= ${PKGVERSION_NOREV}

MAINTAINER= [email protected]
HOMEPAGE= http://code.google.com/p/tesseract-ocr/
HOMEPAGE= https://github.com/tesseract-ocr/tesseract
COMMENT= Commercial quality open source OCR engine
LICENSE= apache-2.0

Expand All @@ -17,22 +20,14 @@ USE_LIBTOOL= yes
USE_TOOLS+= gmake pax
GNU_CONFIGURE= yes
CONFIGURE_ENV+= LIBLEPT_HEADERSDIR=${BUILDLINK_PREFIX.leptonica}/include
MAKE_ENV+= LANGS=${TESSERACT_LANGS:Q}

WRKSRC= ${WRKDIR}/tesseract-ocr
WRKSRC= ${WRKDIR}/tesseract-${PKGVERSION_NOREV}

INSTALLATION_DIRS= libexec share/doc/tesseract share/tesseract

TESSERACT_LANGS= afr ara aze bel ben bul cat ces chi_sim chi_tra chr \
dan deu ell eng enm epo equ est eus fin fra frk frm \
glg grc heb hin hrv hun ind isl ita jpn kan kor lav \
lit mal mkd mlt msa nld nor pol por rus slk slv spa \
sqi srp swa swe tam tel tgl tha tur ukr vie
.for l in ${TESSERACT_LANGS}
DISTFILES+= tesseract-ocr-3.02.${l}.tar.gz
.endfor

LIBS.SunOS+= -lsocket -lnsl

DEPENDS+=tesseract-data>=3.04.00:../tesseract-data

.include "../../graphics/leptonica/buildlink3.mk"
.include "../../mk/bsd.pkg.mk"
Loading