htmlparser-cvs

[Derrick O. <der...@us...>]

Update of /cvsroot/htmlparser/htmlparser/bin
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv30655/htmlparser/bin

Added Files:
	beanybaby.cmd filterbuilder.cmd lexer.cmd linkextractor.cmd 
	parser.cmd sitecapturer sitecapturer.cmd stringextractor.cmd 
	thumbelina.cmd translate.cmd 
Removed Files:
	beanybaby.bat filterbuilder.bat lexer.bat linkextractor.bat 
	parser.bat stringextractor.bat thumbelina.bat translate.bat 
Log Message:
Documentation revamp part one.
Deprecated node decorators.
Added doSemanticAction for Text and Comment nodes.
Added missing sitecapturer scripts.
Fixed DOS batch files to work when called from any location.

--- parser.bat DELETED ---

--- stringextractor.bat DELETED ---

--- translate.bat DELETED ---

--- NEW FILE: parser.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/parser.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmlparser.jar org.htmlparser.Parser %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmlparser.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- NEW FILE: filterbuilder.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/filterbuilder.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_htmlparser_jar_error
if not exist %lib_path%filterbuilder.jar goto no_filterbuilder_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -Xmx256M -classpath %lib_path%filterbuilder.jar;%lib_path%htmlparser.jar org.htmlparser.parserapplications.filterbuilder.FilterBuilder %1
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_htmlparser_jar_error
echo Unable to find htmlparser.jar
goto end
:no_filterbuilder_jar_error
echo Unable to find filterbuilder.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- NEW FILE: sitecapturer.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/sitecapturer.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmlparser.jar org.htmlparser.parserapplications.SiteCapturer %1 %2 %3
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmlparser.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- NEW FILE: translate.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/translate.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmlparser.jar org.htmlparser.util.Translate %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmlparser.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- NEW FILE: lexer.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/lexer.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmllexer.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmllexer.jar org.htmlparser.lexer.Lexer %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmllexer.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- NEW FILE: linkextractor.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/linkextractor.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmlparser.jar org.htmlparser.parserapplications.LinkExtractor %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmlparser.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- NEW FILE: stringextractor.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/stringextractor.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmlparser.jar org.htmlparser.parserapplications.StringExtractor %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmlparser.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- beanybaby.bat DELETED ---

--- filterbuilder.bat DELETED ---

--- linkextractor.bat DELETED ---

--- NEW FILE: beanybaby.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/beanybaby.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmlparser.jar goto no_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -classpath %lib_path%htmlparser.jar org.htmlparser.beans.BeanyBaby %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_jar_error
echo Unable to find htmlparser.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end

--- lexer.bat DELETED ---

--- NEW FILE: sitecapturer ---
#! /bin/sh

if [ -z "$HTMLPARSER_HOME" ] ; then

  ## resolve links - $0 may be a link to the home
  PRG="$0"
  progname=`basename "$0"`
  saveddir=`pwd`

  # need this for relative symlinks
  dirname_prg=`dirname "$PRG"`
  cd "$dirname_prg"
  
  while [ -h "$PRG" ] ; do
    ls=`ls -ld "$PRG"`
    link=`expr "$ls" : '.*-> \(.*\)$'`
    if expr "$link" : '/.*' > /dev/null; then
	PRG="$link"
    else
	PRG=`dirname "$PRG"`"/$link"
    fi
  done
  
  HTMLPARSER_HOME=`dirname "$PRG"`/..

  cd "$saveddir"

  # make it fully qualified
  HTMLPARSER_HOME=`cd "$HTMLPARSER_HOME" && pwd`
fi

if [ -z "$JAVACMD" ] ; then 
  if [ -n "$JAVA_HOME"  ] ; then
    if [ -x "$JAVA_HOME/jre/sh/java" ] ; then 
      # IBM's JDK on AIX uses strange locations for the executables
      JAVACMD="$JAVA_HOME/jre/sh/java"
    else
      JAVACMD="$JAVA_HOME/bin/java"
    fi
  else
    JAVACMD=`which java 2> /dev/null `
    if [ -z "$JAVACMD" ] ; then 
        JAVACMD=java
    fi
  fi
fi
 
if [ ! -x "$JAVACMD" ] ; then
  echo "Error: JAVA_HOME is not defined correctly."
  echo "  We cannot execute $JAVACMD"
  exit 1
fi

if [ -n "$CLASSPATH" ] ; then
  LOCALCLASSPATH="$CLASSPATH"
fi

HTMLPARSER_LIB="${HTMLPARSER_HOME}/lib"

# add in the parser .jar file
if [ -z "$LOCALCLASSPATH" ] ; then
  LOCALCLASSPATH="${HTMLPARSER_LIB}/htmlparser.jar"
else
  LOCALCLASSPATH="${HTMLPARSER_LIB}/htmlparser.jar":"$LOCALCLASSPATH"
fi

# handle 1.1x JDKs
if [ -n "$JAVA_HOME" ] ; then
  if [ -f "$JAVA_HOME/lib/classes.zip" ] ; then
    LOCALCLASSPATH="$LOCALCLASSPATH:$JAVA_HOME/lib/classes.zip"
  fi
fi

"$JAVACMD" -classpath "$LOCALCLASSPATH" org.htmlparser.parserapplications.SiteCapturer "$@"


--- thumbelina.bat DELETED ---

--- NEW FILE: thumbelina.cmd ---
@echo off
rem HTMLParser Library $Name:  $ - A java-based parser for HTML
rem http://sourceforge.org/projects/htmlparser
rem Copyright (C) 2005 Derrick Oswald
rem
rem Revision Control Information
rem
rem $Source: /cvsroot/htmlparser/htmlparser/bin/thumbelina.cmd,v $
rem $Author: derrickoswald $
rem $Date: 2005/04/10 23:20:41 $
rem $Revision: 1.1 $
rem
rem This library is free software; you can redistribute it and/or
rem modify it under the terms of the GNU Lesser General Public
rem License as published by the Free Software Foundation; either
rem version 2.1 of the License, or (at your option) any later version.
rem
rem This library is distributed in the hope that it will be useful,
rem but WITHOUT ANY WARRANTY; without even the implied warranty of
rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
rem Lesser General Public License for more details.
rem
rem You should have received a copy of the GNU Lesser General Public
rem License along with this library; if not, write to the Free Software
rem Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
rem
setlocal enableextensions
if errorlevel 1 goto no_extensions_error
for %%i in (%0) do set cmd_path= %%~dpi
for /D %%i in (%cmd_path%..\lib\) do set lib_path=%%~dpi
if not exist %lib_path%htmllexer.jar goto no_htmllexer_jar_error
if not exist %lib_path%thumbelina.jar goto no_thumbelina_jar_error
for %%i in (java.exe) do set java_executable=%%~$PATH:i
if "%java_executable%"=="" goto no_java_error
@echo on
%java_executable% -Xmx256M -classpath %lib_path%thumbelina.jar;%lib_path%htmllexer.jar org.htmlparser.lexerapplications.thumbelina.Thumbelina %1 %2
@echo off
goto end
:no_extensions_error
echo Unable to use CMD extensions
goto end
:no_htmllexer_jar_error
echo Unable to find htmllexer.jar
goto end
:no_thumbelina_jar_error
echo Unable to find thumbelina.jar
goto end
:no_java_error
echo Unable to find java.exe
goto end
:end