From 1669aa7eb3b1885a1a20d257d1c133bc66f7d6ae Mon Sep 17 00:00:00 2001 From: Tobias Klaus Date: Mon, 16 Oct 2017 22:52:15 +0200 Subject: [PATCH] adds pdfsandwich --- app-text/pdfsandwich/ChangeLog | 59 +++++++++++++++++++ app-text/pdfsandwich/ChangeLog-2015 | 58 ++++++++++++++++++ app-text/pdfsandwich/Manifest | 5 ++ app-text/pdfsandwich/metadata.xml | 25 ++++++++ app-text/pdfsandwich/pdfsandwich-0.1.6.ebuild | 37 ++++++++++++ 5 files changed, 184 insertions(+) create mode 100644 app-text/pdfsandwich/ChangeLog create mode 100644 app-text/pdfsandwich/ChangeLog-2015 create mode 100644 app-text/pdfsandwich/Manifest create mode 100644 app-text/pdfsandwich/metadata.xml create mode 100644 app-text/pdfsandwich/pdfsandwich-0.1.6.ebuild diff --git a/app-text/pdfsandwich/ChangeLog b/app-text/pdfsandwich/ChangeLog new file mode 100644 index 0000000..bab2c21 --- /dev/null +++ b/app-text/pdfsandwich/ChangeLog @@ -0,0 +1,59 @@ +# ChangeLog for app-text/pdfsandwich +# Copyright 1999-2017 Gentoo Foundation; Distributed under the GPL v2 +# (auto-generated from git log) + +*pdfsandwich-0.1.4 (09 Aug 2015) + + 09 Aug 2015; Robin H. Johnson +metadata.xml, + +pdfsandwich-0.1.4.ebuild: + proj/gentoo: Initial commit + + This commit represents a new era for Gentoo: + Storing the gentoo-x86 tree in Git, as converted from CVS. + + This commit is the start of the NEW history. + Any historical data is intended to be grafted onto this point. + + Creation process: + 1. Take final CVS checkout snapshot + 2. Remove ALL ChangeLog* files + 3. Transform all Manifests to thin + 4. Remove empty Manifests + 5. Convert all stale $Header$/$Id$ CVS keywords to non-expanded Git $Id$ + 5.1. Do not touch files with -kb/-ko keyword flags. + + Signed-off-by: Robin H. Johnson + X-Thanks: Alec Warner - did the GSoC 2006 migration + tests + X-Thanks: Robin H. Johnson - infra guy, herding this + project + X-Thanks: Nguyen Thai Ngoc Duy - Former Gentoo + developer, wrote Git features for the migration + X-Thanks: Brian Harring - wrote much python to improve + cvs2svn + X-Thanks: Rich Freeman - validation scripts + X-Thanks: Patrick Lauer - Gentoo dev, running new 2014 + work in migration + X-Thanks: Michał Górny - scripts, QA, nagging + X-Thanks: All of other Gentoo developers - many ideas and lots of paint on + the bikeshed + + 24 Aug 2015; Justin Lecher metadata.xml: + Use https by default + + Convert all URLs for sites supporting encrypted connections from http to + https + + Signed-off-by: Justin Lecher + + 24 Aug 2015; Mike Gilbert metadata.xml: + Revert DOCTYPE SYSTEM https changes in metadata.xml + + repoman does not yet accept the https version. + This partially reverts eaaface92ee81f30a6ac66fe7acbcc42c00dc450. + + Bug: https://bugs.gentoo.org/552720 + + 24 Jan 2016; Michał Górny metadata.xml: + Set appropriate maintainer types in metadata.xml (GLEP 67) + diff --git a/app-text/pdfsandwich/ChangeLog-2015 b/app-text/pdfsandwich/ChangeLog-2015 new file mode 100644 index 0000000..fdcc64b --- /dev/null +++ b/app-text/pdfsandwich/ChangeLog-2015 @@ -0,0 +1,58 @@ +# ChangeLog for app-text/pdfsandwich +# Copyright 1999-2017 Gentoo Foundation; Distributed under the GPL v2 +# $Header: /var/cvsroot/gentoo-x86/app-text/pdfsandwich/ChangeLog,v 1.11 2015/07/31 07:46:01 tomka Exp $ + +*pdfsandwich-0.1.4 (31 Jul 2015) + + 31 Jul 2015; Thomas Kahle +pdfsandwich-0.1.4.ebuild, + -pdfsandwich-0.1.2-r1.ebuild, -pdfsandwich-0.1.2.ebuild: + bump to 0.1.4, remove old + + 09 Jun 2015; Justin Lecher metadata.xml: + Updating remote-id in metadata.xml + +*pdfsandwich-0.1.2-r1 (07 Aug 2014) + + 07 Aug 2014; Thomas Kahle +pdfsandwich-0.1.2-r1.ebuild, + -pdfsandwich-0.0.7.ebuild, -pdfsandwich-0.0.8.ebuild: + EAPI bump, add missing dep on *magick (bug 519282), remove old + +*pdfsandwich-0.1.2 (29 Jul 2014) + + 29 Jul 2014; Thomas Kahle +pdfsandwich-0.1.2.ebuild: + bump to 0.1.2 + +*pdfsandwich-0.0.8 (21 Apr 2013) + + 21 Apr 2013; Thomas Kahle +pdfsandwich-0.0.8.ebuild: + Bump to 0.0.8 + +*pdfsandwich-0.0.7 (14 Jan 2013) + + 14 Jan 2013; Thomas Kahle +pdfsandwich-0.0.7.ebuild, + -pdfsandwich-0.0.3.ebuild, -pdfsandwich-0.0.6.ebuild: + bump to 0.0.7 + +*pdfsandwich-0.0.6 (10 Oct 2012) + + 10 Oct 2012; Thomas Kahle +pdfsandwich-0.0.6.ebuild: + Bump to 0.0.6 + + 13 Sep 2011; Thomas Kahle -pdfsandwich-0.0.2.ebuild, + pdfsandwich-0.0.3.ebuild: + update cuneiform dep, remove old, thanks Halcy0n + +*pdfsandwich-0.0.3 (27 Oct 2010) + + 27 Oct 2010; Thomas Kahle +pdfsandwich-0.0.3.ebuild: + bump to 0.0.3 + + 22 Oct 2010; Thomas Kahle pdfsandwich-0.0.2.ebuild: + Ebuild-QA + +*pdfsandwich-0.0.2 (22 Oct 2010) + + 22 Oct 2010; Thomas Kahle +pdfsandwich-0.0.2.ebuild, + +metadata.xml: + New ebuild: app-text/pdfsandwich, fixes bug 321885, original ebuild + by Tobias Elze. diff --git a/app-text/pdfsandwich/Manifest b/app-text/pdfsandwich/Manifest new file mode 100644 index 0000000..ae0a4c2 --- /dev/null +++ b/app-text/pdfsandwich/Manifest @@ -0,0 +1,5 @@ +DIST pdfsandwich-0.1.6.tar.bz2 17031 SHA256 96831eb191bcd43e730dcce169d5c14b47bba0b6cd5152a8703e3b573013a2a2 SHA512 fa122bb2cac678555a7abe9b3c720baf608197d60593b8fdaaf7009837228b30d5ab74fac1c51c68e0e75ae155b0a774077745e5eb157b951c634529fe6146f9 WHIRLPOOL 02f38e9031bf98d085b42e5fb0369cb7b8f14adf3005f56db81b1c8c48dd4a6e9a97848e36eab606475349be4fd5ca21c14f32d6358e15e94240e4ca2c19f913 +EBUILD pdfsandwich-0.1.6.ebuild 970 SHA256 31e52d93880ab0f86aec47a27b2abd5362b6c7a85ae9a575103b6ec7743ae5ef SHA512 3505739fec5f3d76a7d51c0c9f4f4cf992faa59e23779dfd0ebf677246da0a794ca18af304a5631e7fc0984bd698bfe69f6c3a012389cdc943a543ac0fe4b263 WHIRLPOOL 293f1bbd0ef56e30dc2dc30809dcee8cbbd106a20ac3562dfb77c865f302386ff180d4d9b19156c1fdafd5433904c7e71ff5657e1ce69a86779ee6009407394b +MISC ChangeLog 2188 SHA256 ee5dae685f3f6b62ed269902da785d53c6f0bc465867d0f1587e7d4a5e893f79 SHA512 441efdd15a62c58bb10863773d1855fb888c0cc3531849e3e97ef897c7d100dc5114fe0e2633047e22586a35d157bb1ba7b91bad3110a12d0d56c5186ac4e9fd WHIRLPOOL 4357762a53f83c3795c62ca50f7f229713651caa5d8eeb00793d2656eae0e517fa7d45c0bae652e15ed9281beffe3d9057d72fdf4902062b4e29ba64bfa3fd3a +MISC ChangeLog-2015 1879 SHA256 042aacf9cd5653fa9742930dcfaf43d8064dabf9b181a6ed3baf2060a186ebdc SHA512 e78796f8be7bf24e6f0942116bcb02dadb59484e74836c3d1bcf40d0578a7285da033b60193e62cac4a1e2303263ad4302300651c337f6cd0e7fb34b0e234061 WHIRLPOOL 7536baf0c008841cf5346854fbbab09dcbfe6d79d5016fa2d633438ddf78c919c7aeabd55bdfb53d0298b02c4b6feaccb94a0763183f5f1a4e367f82bd18ea38 +MISC metadata.xml 986 SHA256 63d97f486fa755d40887c5c547b3398b26da828b3049449b3aacfeff55975ec2 SHA512 69efd526eeb24fb1176b23678775d36987ad0db32f6f4a56e760eb5fb27cf3b42604441374521ff7f5d7a061c43e3ede2e08372421d1e7f7753a5f3e160b8171 WHIRLPOOL fda9100da8958a678552bbf05a3e1307f82a956311f2ace0d315937ae60f15ec9582b4b9562ddc5d48a46276d1656c0258024774d62a253d573f7c94f49ec8c5 diff --git a/app-text/pdfsandwich/metadata.xml b/app-text/pdfsandwich/metadata.xml new file mode 100644 index 0000000..8c10b27 --- /dev/null +++ b/app-text/pdfsandwich/metadata.xml @@ -0,0 +1,25 @@ + + + + + tomka@gentoo.org + + +pdfsandwich generates "sandwich" OCR pdf files, i.e. pdf files which +contain only images (no text) will be processed by optical character +recognition (OCR) and the text will be added to each page invisibly +"behind" the images. + +pdfsandwich is a command line tool which is supposed to be useful to +OCR scanned books or journals. It is able to recognize the page layout +even for multicolumn text. + +Essentially, pdfsandwich is a wrapper script which calls the following +binaries: convert, cuneiform, gs, and hocr2pdf. It is known to run on +Unix systems and has been tested on Linux and MacOS X. It supports +parallel processing on multiprocessor systems. + + + pdfsandwich + + diff --git a/app-text/pdfsandwich/pdfsandwich-0.1.6.ebuild b/app-text/pdfsandwich/pdfsandwich-0.1.6.ebuild new file mode 100644 index 0000000..21bf32d --- /dev/null +++ b/app-text/pdfsandwich/pdfsandwich-0.1.6.ebuild @@ -0,0 +1,37 @@ +# Copyright 1999-2017 Gentoo Foundation +# Distributed under the terms of the GNU General Public License v2 + +EAPI=5 + +DESCRIPTION="generator of sandwich OCR pdf files" +HOMEPAGE="http://www.tobias-elze.de/pdfsandwich" +SRC_URI="mirror://sourceforge/pdfsandwich/${P}.tar.bz2" + +LICENSE="GPL-2" +KEYWORDS="~amd64 ~x86" +SLOT="0" +IUSE="png" + +RDEPEND=">=app-text/tesseract-3.00 + media-gfx/exact-image + app-text/unpaper + app-text/ghostscript-gpl + virtual/imagemagick-tools[png?]" +DEPEND="sys-apps/gawk + >=dev-lang/ocaml-3.10[ocamlopt]" + +src_prepare() { + sed -i "/^OCAMLOPTFLAGS/s/$/ -ccopt \"\$(CFLAGS) \$(LDFLAGS)\"/" Makefile || die + sed -i "s/install -s/install/" Makefile || die +} + +src_install() { + emake DESTDIR="${D}" install +} + +pkg_postinst() { + elog "pdfsandwich relies on the tesseract library for OCR." + elog "Consequently language support is determined by tesseract's" + elog "language support which in turn is controlled via the LINGUAS" + elog "variable in make.conf." +}