added wv (formerly mswordview) - a command line util to convert MSWord docs to something

readable.

4 files changed, 218 insertions, 0 deletions

diff --git a/app-text/wv/ChangeLog b/app-text/wv/ChangeLog
new file mode 100644
index 000000000000..aed381911e34
--- /dev/null
+++ b/app-text/wv/ChangeLog
@@ -0,0 +1,11 @@
+# ChangeLog for app-text/wv
+# Copyright 2002 Gentoo Technologies, Inc.; Distributed under the GPL
+# /space/gentoo/cvsroot/gentoo-x86/dev-lang/swig/ChangeLog,v 1.1 2002/02/01 21:53:11 gbevin Exp
+
+*wv-0.7.1 (8 Feb 2002)
+  
+  19 Apr 2002; G.Shapovalov <georges@its.caltech.edu> all_files :
+
+  initial release. This is wv, formerly known as mswordview. This is a utility (or set of) which
+  can convert MSWord files to different formats from clear text, html, tex, etc. to ps, pdf.
+
diff --git a/app-text/wv/files/digest-wv-0.7.1 b/app-text/wv/files/digest-wv-0.7.1
new file mode 100644
index 000000000000..9ce34dc9b76a
--- /dev/null
+++ b/app-text/wv/files/digest-wv-0.7.1
@@ -0,0 +1 @@
+MD5 f4bf6e9efd6c6eccaad02d3f315beb75 wv-0.7.1.tar.gz 1259264
diff --git a/app-text/wv/files/wv-0.7.1-rvt.patch b/app-text/wv/files/wv-0.7.1-rvt.patch
new file mode 100644
index 000000000000..080bcf8f1675
--- /dev/null
+++ b/app-text/wv/files/wv-0.7.1-rvt.patch
@@ -0,0 +1,155 @@
+diff -Naur wv/escher.c wv.mod/escher.c
+--- wv/escher.c	Wed Apr 11 01:27:47 2001
++++ wv.mod/escher.c	Wed Jan 16 13:33:08 2002
+@@ -215,8 +215,7 @@
+ 		count +=
+ 		    wvGetSpgrContainer (&
+ 					(item->spgrcontainer
+-					 [item->no_spgrcontainer - 1]),
+-&amsofbh, fd);
++					 [item->no_spgrcontainer - 1]), &amsofbh, fd);
+ 		break;
+ 	    default:
+ 		count += wvEatmsofbt (&amsofbh, fd);
+@@ -303,8 +302,7 @@
+ 		count +=
+ 		    wvGetSpgrContainer (&
+ 					(item->spgrcontainer
+-					 [item->no_spgrcontainer - 1]),
+-&amsofbh, fd);
++					 [item->no_spgrcontainer - 1]), &amsofbh, fd);
+ 		break;
+ 	    default:
+ 		count += wvEatmsofbt (&amsofbh, fd);
+@@ -684,10 +682,14 @@
+ U32
+ wvEatmsofbt (MSOFBH * amsofbh, wvStream * fd)
+ {
++    /* Ries (rvt@dds.nl)
++       Changed read to bound to 32bit instead of 8bit 
++       MS prolly does 32 bit bounderaries ??? */
+     U32 i;
+-    for (i = 0; i < amsofbh->cbLength; i++)
+-	read_8ubit (fd);
+-    return (amsofbh->cbLength);
++    U32 check=(amsofbh->cbLength + 3) >> 2;
++    for (i = 0; i < check; i++)
++	read_32ubit (fd);
++    return (check << 2);
+ }
+ 
+ void
+diff -Naur wv/picf.c wv.mod/picf.c
+--- wv/picf.c	Wed Jan  2 20:33:34 2002
++++ wv.mod/picf.c	Wed Jan 16 13:33:08 2002
+@@ -218,13 +218,19 @@
+ 		break;
+ 	    default:
+ 		{
+-		    U32 len = entry - 2;
++		    U32 lene2 = entry - 2;
+ 		    U32 i;
+ 		    wvTrace (
+-			     ("len is %d, predict end of %x\n", len,
++			     ("lene2 is %d, predict end of %x\n", len,
+ 			      wvStream_tell (fd) + (entry - 2) * 2));
+-		    for (i = 0; i < len; i++)
++		    /* RIES (rvt@dds.nl)
++		       prolly a dirty patch because I check count
++		       everytime it's incremnented against lene2.
++		       This seems twork very well I tried it on around 15.000
++		       word documents and it seems to work! */
++		    for (i = 0; i < lene2; i++)
+ 		      {
++		          if ( (count + 1) >= len) return (count);
+ 			  test = read_16ubit (fd);
+ 			  if ((i == 0)
+ 			      && ((test == 0x0f43) || (test == 0x0b41)))
+@@ -232,11 +238,13 @@
+ 				wvTrace (
+ 					 ("Found a Bitmap, Will strip header and return with bitmap data\n"));
+ 				count += 2;
++    		        	if ( (count + 1) >= len) return (count);
+ 				pad = test;
+ 				test = read_32ubit (fd);	/*0x00cc0020 */
+ 				if (test != 0x00cc0020)
+ 				    wvTrace (("Old Graphic\n"));
+ 				count += 4;
++    		        	if ( (count + 1) >= len) return (count);
+ 
+ 				if (pad == 0x0f43)
+ 				  {
+@@ -244,20 +252,26 @@
+ 				      if (test != 0x0000)
+ 					  wvTrace (("Old Graphic\n"));
+ 				      count += 2;
++			              if ( (count + 1) >= len) return (count);
+ 				  }
+ 
+ 				read_16ubit (fd);	/*width */
+ 				count += 2;
++    		        	if ( (count + 1) >= len) return (count);
+ 				read_16ubit (fd);	/*height */
+ 				count += 2;
++    		        	if ( (count + 1) >= len) return (count);
+ 				test = read_32ubit (fd);	/*0x00000000L */
+ 				if (test != 0x00000000L)
+ 				    wvTrace (("Old Graphic\n"));
+ 				count += 4;
++    		        	if ( (count + 1) >= len) return (count);
+ 				read_16ubit (fd);	/*width */
+ 				count += 2;
++    		        	if ( (count + 1) >= len) return (count);
+ 				read_16ubit (fd);	/*height */
+ 				count += 2;
++    		        	if ( (count + 1) >= len) return (count);
+ 				test = read_32ubit (fd);	/*0x00000000L */
+ 				if (test != 0x00000000L)
+ 				    wvTrace (("Old Graphic\n"));
+@@ -265,6 +279,7 @@
+ 				return (count);
+ 			    }
+ 			  count += 2;
++	        	  if ( (count + 1) >= len) return (count);
+ 		      }
+ 		}
+ 		break;
+diff -Naur wv/wvText.in wv.mod/wvText.in
+--- wv/wvText.in	Sun Mar 11 20:00:24 2001
++++ wv.mod/wvText.in	Wed Jan 16 13:33:08 2002
+@@ -3,6 +3,7 @@
+ prefix=@prefix@
+ exec_prefix=@exec_prefix@
+ datadir=@datadir@
++tmpdir=/tmp
+ 
+ # argument checking
+ if [ ${#} -ne "2" ]; then
+@@ -29,23 +30,23 @@
+     fi
+ 
+     # intermediate file
+-    TMP_FILE="/tmp/wv$$.html"
++    TMP_FILE="wv$$.html"
+ 
+-    wvHtml "${1}" "${TMP_FILE}" >/dev/null 2>&1
++    wvHtml "${1}" --targetdir="${tmpdir}" "${TMP_FILE}" >/dev/null 2>&1
+     if [ ${?} -ne "0" ]; then
+ 	echo "Could not convert into HTML"
+ 	exit 1
+     fi
+ 
+     # lynx actually does quite well
+-    TERM=vt100 lynx -dump -force_html "${TMP_FILE}" > "${2}"
++    TERM=vt100 lynx -dump -force_html "${tmpdir}/${TMP_FILE}" > "${2}"
+     if [ ${?} -ne "0" ]; then
+ 	    echo "Could not convert into Text"
+ 	    exit 1
+     fi
+ 
+     # clean up
+-    rm -f ${TMP_FILE}
++    rm -f "${tmpdir}/${TMP_FILE}"
+ 
+ else
+     # fall back onto our cruddy output
diff --git a/app-text/wv/wv-0.7.1.ebuild b/app-text/wv/wv-0.7.1.ebuild
new file mode 100644
index 000000000000..55a054f68ea8
--- /dev/null
+++ b/app-text/wv/wv-0.7.1.ebuild
@@ -0,0 +1,51 @@
+# Copyright 1999-2001 Gentoo Technologies, Inc.
+# Distributed under the terms of the GNU General Public License
+# Maintainer: George Shapovalov <georges@cco.caltech.edu>
+# /space/gentoo/cvsroot/gentoo-x86/dev-libs/libwmf/libwmf-0.2.2.ebuild,v 1.2 2002/01/24 19:40:43 karltk Exp
+
+
+
+S=${WORKDIR}/${P}
+DESCRIPTION="tool for convertion of MSWord doc and rtf files to something readable"
+SRC_URI="http://prdownloads.sourceforge.net/wvware/wv-0.7.1.tar.gz"
+HOMEPAGE="http://www.wvware.com"
+
+DEPEND="virtual/glibc
+	>=sys-devel/gcc-2.95.2
+	>=media-libs/libwmf-0.2.2 
+	>=media-libs/freetype-2.0.1
+	sys-libs/zlib
+	media-libs/libpng"
+	
+RDEPEND="media-gfx/imagemagick"		
+
+rc_unpack() {
+        unpack ${A}
+
+        cd ${S}
+		patch -p1 < ${FILESDIR}/wv-0.7.1-rvt.patch
+}
+
+
+src_compile() {
+	
+	unset CXXFLAGS
+	unset CFLAGS
+
+	./configure \
+		--host=${CHOST} \
+		--prefix=/usr \
+		--infodir=/usr/share/info \
+		--mandir=/usr/share/man \
+		--with-docdir=/usr/share/doc/${PF} || die "./configure failed"
+
+	emake || die
+}
+
+src_install () {
+    make \
+        prefix=${D}/usr \
+        mandir=${D}/usr/share/man \
+        infodir=${D}/usr/share/info \
+        install || die
+}