graphics/ocropus: Updated for version 0.6_20120920.

Now a python only application

Signed-off-by: Matteo Bernardini <ponce@slackbuilds.org>

8 files changed, 44 insertions, 118 deletions

diff --git a/graphics/ocropus/README b/graphics/ocropus/README
index 9aeb9fc26e..50b1f3c306 100644
--- a/graphics/ocropus/README
+++ b/graphics/ocropus/README
@@ -5,3 +5,6 @@ natural language modeling, and multi-lingual capabilities.
 The system is being developed with the generous support from Google and 
 other organizations; the primary developers are at the IUPR Research 
 Group at the DFKI Research Center.
+
+Note: the tarball of the sources is nearly 400 megs, so be patient when
+downloading/building.
diff --git a/graphics/ocropus/no_display.patch b/graphics/ocropus/no_display.patch
new file mode 100644
index 0000000000..2fdae89fce
--- /dev/null
+++ b/graphics/ocropus/no_display.patch
@@ -0,0 +1,15 @@
+See https://code.google.com/p/ocropus/issues/detail?id=365
+
+diff -Naur ocropus-20120920.orig/ocropy/setup.py ocropus-20120920/ocropy/setup.py
+--- ocropus-20120920.orig/ocropy/setup.py	2012-09-20 06:48:34.000000000 +0200
++++ ocropus-20120920/ocropy/setup.py	2012-09-20 11:16:24.784307573 +0200
+@@ -4,6 +4,9 @@
+ from distutils.core import setup, Extension, Command
+ from distutils.command.install_data import install_data
+ 
++import matplotlib
++matplotlib.use('Agg')
++
+ from ocrolib import default
+ modeldir = "models/"
+ modelfiles = default.installable
diff --git a/graphics/ocropus/ocrodata-env.diff b/graphics/ocropus/ocrodata-env.diff
deleted file mode 100644
index 04cfd5d9af..0000000000
--- a/graphics/ocropus/ocrodata-env.diff
+++ /dev/null
@@ -1,15 +0,0 @@
-Description: Respect the OCRODATA environment variable for all lua scripts.
-Author: Jakub Wilk <jwilk@debian.org>
-
-Index: ocropus-0.3.1/ocroscript/ocrotoplevel.cc
-===================================================================
---- ocropus-0.3.1.orig/ocroscript/ocrotoplevel.cc	2009-11-26 18:47:54.000000000 +0100
-+++ ocropus-0.3.1/ocroscript/ocrotoplevel.cc	2009-11-26 18:47:54.000000000 +0100
-@@ -471,6 +471,7 @@
-     lua_call(L, 0, 0);
- 
-     // handle OCRODATA environment variable as a directory
-+    if(getenv("OCRODATA")) ocroscripts = getenv("OCRODATA");
-     lua_pushstring(L, ocrodata);
-     lua_setglobal(L, "ocrodata");
- 
diff --git a/graphics/ocropus/ocropus.SlackBuild b/graphics/ocropus/ocropus.SlackBuild
index e8c2ce60b0..b3aa2b7e00 100644
--- a/graphics/ocropus/ocropus.SlackBuild
+++ b/graphics/ocropus/ocropus.SlackBuild
@@ -23,12 +23,11 @@
 # ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 PRGNAM=ocropus
-VERSION=${VERSION:-0.3.1}
+VERSION=${VERSION:-0.6_20120920}
+SRCVER=${SRCVER:-20120920}
 BUILD=${BUILD:-1}
 TAG=${TAG:-_SBo}
 
-DIRVER=${DIRVER:-0.3}
-
 if [ -z "$ARCH" ]; then
   case "$( uname -m )" in
     i?86) ARCH=i486 ;;
@@ -61,45 +60,34 @@ set -e
 rm -rf $PKG
 mkdir -p $TMP $PKG $OUTPUT
 cd $TMP
-rm -rf $PRGNAM-$DIRVER
-tar xvf $CWD/$PRGNAM-$VERSION.tar.gz
-cd $PRGNAM-$DIRVER
+rm -rf $PRGNAM-$SRCVER
+tar xvf $CWD/$PRGNAM-$SRCVER.tar.?z*
+cd $PRGNAM-$SRCVER
 chown -R root:root .
 chmod -R u+w,go+r-w,a-s .
 
-# Debian patch to fix hardcoded /usr/local paths in some source files
-patch -p1 < $CWD/usr-local.diff
-# Debian patch to fix behaviour of the OCRODATA environment variable
-patch -p1 < $CWD/ocrodata-env.diff
+# We don't need a DISPLAY, we're just packaging
+patch -p1 < $CWD/no_display.patch
+
+# Fix some paths
+sed -i "s|/usr/local/share|/usr/share|" \
+  ocropy/Notebooks/ocropus-steps.ipynb \
+  ocropy/ocrolib/default.py \
+  ocropy/ocrolib/common.py
 
-CFLAGS="$SLKCFLAGS" \
-CXXFLAGS="$SLKCFLAGS" \
-./configure \
-  --prefix=/usr \
-  --sysconfdir=/etc \
-  --localstatedir=/var \
-  --libdir=/usr/lib${LIBDIRSUFFIX} \
-  --mandir=/usr/man \
-  --docdir=/usr/doc/$PRGNAM-$VERSION \
-  --with-tesseract=/usr \
-  --with-iulib=/usr \
-  --without-fst \
-  --without-SDL \
-  --without-leptonica \
-  --build=$ARCH-slackware-linux
+( cd ocropy
+python setup.py install --root=$PKG )
 
-make
-make install DESTDIR=$PKG
+# move models in a subfolder
+mkdir -p $PKG/usr/share/models
+mv $PKG/usr/share/$PRGNAM/* $PKG/usr/share/models
+mv $PKG/usr/share/models $PKG/usr/share/$PRGNAM/
 
 find $PKG | xargs file | grep -e "executable" -e "shared object" | grep ELF \
   | cut -f 1 -d : | xargs strip --strip-unneeded 2> /dev/null || true
 
-# Add Debian's manpage
-mkdir -p $PKG/usr/man/man1
-gzip -9c $CWD/ocroscript.1 > $PKG/usr/man/man1/ocroscript.1.gz
-
 mkdir -p $PKG/usr/doc/$PRGNAM-$VERSION
-cp -a CHANGES COPYING DIRS INSTALL README $PKG/usr/doc/$PRGNAM-$VERSION
+cp -a fraktur-boxes historic-newspaper uw3-500 $PKG/usr/doc/$PRGNAM-$VERSION
 cat $CWD/$PRGNAM.SlackBuild > $PKG/usr/doc/$PRGNAM-$VERSION/$PRGNAM.SlackBuild
 
 mkdir -p $PKG/install
diff --git a/graphics/ocropus/ocropus.info b/graphics/ocropus/ocropus.info
index c25b9cc4e0..4fea27c62f 100644
--- a/graphics/ocropus/ocropus.info
+++ b/graphics/ocropus/ocropus.info
@@ -1,10 +1,10 @@
 PRGNAM="ocropus"
-VERSION="0.3.1"
-HOMEPAGE="http://sites.google.com/site/ocropus/"
-DOWNLOAD="http://ocropus.googlecode.com/files/ocropus-0.3.1.tar.gz"
-MD5SUM="2a1b66419ae69ef031d5e6269db15bb5"
+VERSION="0.6_20120920"
+HOMEPAGE="https://code.google.com/p/ocropus/"
+DOWNLOAD="http://ponce.cc/slackware/sources/repo/ocropus-20120920.tar.xz"
+MD5SUM="a61133bdb989e4a812dd130024830c0f"
 DOWNLOAD_x86_64=""
 MD5SUM_x86_64=""
-REQUIRES="iulib tesseract"
+REQUIRES="matplotlib pytables python-magick scipy"
 MAINTAINER="Pierre Cazenave"
 EMAIL="pwcazenave < at > gmail {dot} com"
diff --git a/graphics/ocropus/ocroscript.1 b/graphics/ocropus/ocroscript.1
deleted file mode 100644
index d8087203f7..0000000000
--- a/graphics/ocropus/ocroscript.1
+++ /dev/null
@@ -1,43 +0,0 @@
-.TH ocroscript 1 "June 06, 2008"
-.SH NAME
-ocropus \- command line OCR tool
-.SH SYNOPSIS
-.B ocroscript
-.RI "<script> <arguments>"
-.SH DESCRIPTION
-You can see a list of all available commands by looking in the $OCROSCRIPTS
-(/usr/share/ocropus/scripts/ by default) path.
-.PP
-The \(oqrecognize\(cq script uses tesseract for recognition and sends the html-based hOCR
-ouput to stdout. Tesseract is probably the most mature text recognizer within
-OCRopus at the moment. Natively, Tesseract doesn't do layout analysis, but
-combined with OCRopus, it makes for a pretty good OCR system:
-.RS
-$ ocroscript recognize page.png > page.html
-.RE
-.PP
-Here is a brief summary of the remaining command line commands available.
-You will need to look at the script to see what the command line arguments are:
-.TP
-degrade.lua
-Simple document image degradation
-.TP
-hocr-to-text.lua
-Convert hOCR output to plain text.
-.TP
-line-clean.lua
-Given a line image, remove marginal noise and fix some other problems.
-.TP
-sauvola.lua
-Perform Sauvola thresholding.
-.SH SEE ALSO
-.BR tesseract (1),
-.br
-.PP
-.UR http://code.google.com/p/ocropus/w/list
-.UE
-.SH AUTHOR
-ocroscript was written by Thomas Breuel.
-.PP
-This manual page was written by Jeffrey Ratcliffe <Jeffrey.Ratcliffe@gmail.com>,
-for the Debian project (but may be used by others).
diff --git a/graphics/ocropus/slack-desc b/graphics/ocropus/slack-desc
index 00aef62c44..5dcb605a1e 100644
--- a/graphics/ocropus/slack-desc
+++ b/graphics/ocropus/slack-desc
@@ -16,4 +16,4 @@ ocropus: The system is being developed with the generous support from Google
 ocropus: and other organizations; the primary developers are at the IUPR 
 ocropus: Research Group at the DFKI Research Center.
 ocropus: 
-ocropus: http://sites.google.com/site/ocropus/
+ocropus: https://code.google.com/p/ocropus/
diff --git a/graphics/ocropus/usr-local.diff b/graphics/ocropus/usr-local.diff
deleted file mode 100644
index 0a17478e45..0000000000
--- a/graphics/ocropus/usr-local.diff
+++ /dev/null
@@ -1,22 +0,0 @@
-Description:
-  Use /usr/share/ocropus/scripts/ and /usr/share/ocropus/ as defaults for
-  OCROSCRIPTS and OCRODATA.
-Author: Jakub Wilk <jwilk@debian.org>
-
-Index: ocropus-0.3.1/ocroscript/ocrotoplevel.cc
-===================================================================
---- ocropus-0.3.1.orig/ocroscript/ocrotoplevel.cc	2009-11-26 16:56:18.000000000 +0100
-+++ ocropus-0.3.1/ocroscript/ocrotoplevel.cc	2009-11-26 17:16:32.000000000 +0100
-@@ -68,10 +68,10 @@
- 
- // FIXME the Jamfile isn't passing this flag, so for now, this is a workaround
- #ifndef OCROSCRIPTS
--#define OCROSCRIPTS "/usr/local/share/ocropus/scripts/"
-+#define OCROSCRIPTS "/usr/share/ocropus/scripts/"
- #endif
- #ifndef OCRODATA
--#define OCRODATA "/usr/local/share/ocropus/"
-+#define OCRODATA "/usr/share/ocropus/"
- #endif
- 
- const char *ocroscripts = OCROSCRIPTS;