Jericho HTML Parser is a simple but powerful java library allowing

analysis and manipulation of parts of an HTML document, including
some common server-side tags, while reproducing verbatim any
unrecognised or invalid HTML.

It also provides high-level HTML form manipulation functions.

WWW:	http://jerichohtml.sourceforge.net/doc/index.html

PR:		ports/124770
Submitted by:	Marcin Cieslak <saper at SYSTEM.PL>

4 files changed, 78 insertions, 0 deletions

diff --git a/www/jericho-html/Makefile b/www/jericho-html/Makefile
new file mode 100644
index 000000000000..4dc195625119
--- /dev/null
+++ b/www/jericho-html/Makefile
@@ -0,0 +1,52 @@
+# New ports collection makefile for:	jerichohtml
+# Date created:		2008-06-17
+# Whom:			Marcin Cieslak <saper@SYSTEM.PL>
+#
+# $FreeBSD$
+#
+
+PORTNAME=	jericho-html
+PORTVERSION=	2.5
+CATEGORIES=	www java
+MASTER_SITES=	SF
+MASTER_SITE_SUBDIR=	${PORTNAME:S,-,,}
+
+MAINTAINER=	saper@SYSTEM.PL
+COMMENT=	A java library to analyse and manipulate HTML
+
+USE_ZIP=	yes
+USE_JAVA=	1.3+
+
+INTERFACES:=	"compile-time-dependencies/slf4j-api-1.4.3.jar:\
+		compile-time-dependencies/commons-logging-api-1.1.jar:\
+		compile-time-dependencies/log4j-api-1.2.14.jar"
+
+PORTDOCS=	api
+PLIST_FILES+=	%%JAVAJARDIR%%/${PORTNAME}.jar
+
+do-build:
+	(cd ${WRKSRC} &&  ${RM} -rf classes/* && ${JAVAC}         \
+		-classpath ${INTERFACES:S, ,,g}                   \
+		-d classes src/java/au/id/jericho/lib/html/*.java \
+		src/java/au/id/jericho/lib/html/nodoc/*.java)
+	${JAR} -cf ${WRKSRC}/lib/${PORTNAME}.jar                  \
+		 -C ${WRKSRC}/classes .
+.if !defined(NOPORTDOCS)
+	(cd ${WRKSRC} && ${RM} -rf doc/* && ${JAVADOC} -quiet     \
+		-windowtitle "Jericho HTML Parser ${PORTVERSION}" \
+		-classpath ${INTERFACES:S, ,,g}:src/java:classes  \
+		-use -d ${WRKSRC}/doc/api                         \
+		-subpackages au.id.jericho.lib.html               \
+		-exclude au.id.jericho.lib.html.nodoc             \
+		-noqualifier au.id.jericho.lib.html               \
+		-group "Core package" au.id.jericho.lib.html)
+.endif
+
+do-install:
+	${INSTALL_DATA} ${WRKSRC}/lib/${PORTNAME}.jar ${JAVAJARDIR}
+.if !defined(NOPORTDOCS)
+	${MKDIR} ${DOCSDIR}
+	(cd ${WRKSRC}/doc && ${FIND} api | ${CPIO} -pdmu ${DOCSDIR})
+.endif
+
+.include <bsd.port.mk>
diff --git a/www/jericho-html/distinfo b/www/jericho-html/distinfo
new file mode 100644
index 000000000000..280531763604
--- /dev/null
+++ b/www/jericho-html/distinfo
@@ -0,0 +1,3 @@
+MD5 (jericho-html-2.5.zip) = 64306d0eb82608e50496a680b319182d
+SHA256 (jericho-html-2.5.zip) = 212b9e8b72f9787dfafd046e8716f0d04365afcd3f4d2fb293e69d5b90e456b4
+SIZE (jericho-html-2.5.zip) = 1456664
diff --git a/www/jericho-html/files/patch-encoding b/www/jericho-html/files/patch-encoding
new file mode 100644
index 000000000000..3396a1440a73
--- /dev/null
+++ b/www/jericho-html/files/patch-encoding
@@ -0,0 +1,15 @@
+--- src/java/au/id/jericho/lib/html/StreamEncodingDetector.java.orig	2008-06-17 21:01:53.890292905 +0200
++++ src/java/au/id/jericho/lib/html/StreamEncodingDetector.java	2008-06-17 21:02:43.940300330 +0200
+@@ -203,9 +203,9 @@
+ 		// Assume the more likely case of four 8-bit characters <= U+00FF.

+ 		// Check whether it fits some common EBCDIC strings that might be found at the start of a document:

+ 		if (b1==0x4C) { // first character is EBCDIC '<' (ASCII 'L'), check a couple more characters before assuming EBCDIC encoding:

+-			if (b2==0x6F && b3==0xA7 && b4==0x94) return setEncoding(EBCDIC,"default EBCDIC encoding (<?xml...> detected)"); // first four bytes are "<?xm" in EBCDIC ("Lo§”" in Windows-1252)

+-			if (b2==0x5A && b3==0xC4 && b4==0xD6) return setEncoding(EBCDIC,"default EBCDIC encoding (<!DOCTYPE...> detected)"); // first four bytes are "<!DO" in EBCDIC ("LZÄÖ" in Windows-1252)

+-			if ((b2&b3&b4&0x80)!=0) return setEncoding(EBCDIC,"default EBCDIC-compatible encoding (HTML element detected)"); // all of the 3 bytes after the '<' have the high-order bit set, indicating EBCDIC letters such as "<HTM" ("LÈãÔ" in Windows-1252), or "<htm" ("Lˆ£”" in Windows-1252)

++			if (b2==0x6F && b3==0xA7 && b4==0x94) return setEncoding(EBCDIC,"default EBCDIC encoding (<?xml...> detected)"); // first four bytes are "<?xm" in EBCDIC 
++			if (b2==0x5A && b3==0xC4 && b4==0xD6) return setEncoding(EBCDIC,"default EBCDIC encoding (<!DOCTYPE...> detected)"); // first four bytes are "<!DO" in EBCDIC 
++			if ((b2&b3&b4&0x80)!=0) return setEncoding(EBCDIC,"default EBCDIC-compatible encoding (HTML element detected)"); // all of the 3 bytes after the '<' have the high-order bit set, indicating EBCDIC letters such as "<HTM" or "<htm" 
+ 			// although this is not an exhaustive check for EBCDIC, it is safer to assume a more common preliminary encoding if none of these conditions are met.

+ 		}

+ 		// Now confident that it is not EBCDIC, but some other 8-bit encoding.

diff --git a/www/jericho-html/pkg-descr b/www/jericho-html/pkg-descr
new file mode 100644
index 000000000000..b38a2d771bd5
--- /dev/null
+++ b/www/jericho-html/pkg-descr
@@ -0,0 +1,8 @@
+Jericho HTML Parser is a simple but powerful java library allowing
+analysis and manipulation of parts of an HTML document, including
+some common server-side tags, while reproducing verbatim any
+unrecognised or invalid HTML. 
+
+It also provides high-level HTML form manipulation functions.
+
+WWW:	http://jerichohtml.sourceforge.net/doc/index.html