[PATCHES] lucene plugin

--- lucene.py	2008-03-01 19:00:12.000000000 +0100
+++ mylucene.py	2008-03-01 20:26:06.000000000 +0100
@@ -30,13 +30,14 @@
     this path
  4) Download lucene from http://jakarta.apache.org/lucene and unzip/untar it.
     In config.py, set py['lucene_home'] to the lucene directory (which contains
-    lucene-1.3-final.jar and lucene-demos-1.3-final.jar)
+    lucene-core-XX.jar and lucene-demos-XX.jar)
  5) In config.py set py['lucene_index'] to the name of the Lucene index file
     (pybosxom must be able to read and write this file)
- 6) In config.py set py['JAVA_HOME'] to point at the 'java' command
- 7) Set up a cron job to run py['lucene_bin']/index.sh periodically to
+ 6) In config.py set py['lucene_highlight'] to True or False
+ 7) In config.py set py['JAVA_HOME'] to point at the 'java' command
+ 8) Set up a cron job to run py['lucene_bin']/index.sh periodically to
     reindex your blog
- 8) Somewhere in your web page you need a lucene search form:
+ 9) Somewhere in your web page you need a lucene search form:
 
    <form id="searchform" method="get" action="/blog">
     <table>
@@ -49,19 +50,17 @@
 
     The action of the form should be the top level URI of your blog
 
- 9) You should add $searchHeader somewhere in the header of your webpage; this
+10) You should add $searchHeader somewhere in the header of your webpage; this
    is where statements like, "Your search returned X results for Y" are
    placed. This statement is enclosed in a div tag with a class of 
    "searchtext" so that you can define it as you like in your stylesheet.
 
-10) Uses of Apple's Java on Mac OS X should be sure to use Lucene 1.3
-
 """
 
 __author__ = "Ted Leung - [email protected]"
 __version__ = "$Id$"
 
-import glob, os, urllib
+import glob, os, urllib, re
 from Pyblosxom.entries import fileentry
 
 def verify_installation(request):
@@ -70,15 +69,24 @@
     retval = 1
 
     if not config.has_key('lucene_home') or not os.path.isdir(config['lucene_home']):
-        print 'The "lucene_home" property in the config file must refer to a directory that contains the lucene libraries.'
+        print 'The "lucene_home" property in the config file must refer to ' \
+            'a directory that contains the lucene libraries.'
         retval = 0
 
     if not config.has_key('JAVA_HOME') or not os.path.isfile(config['JAVA_HOME']):
-        print 'The "JAVA_HOME" property in the config file must refer to the "java" command on your system.'
+        print 'The "JAVA_HOME" property in the config file must refer to ' \
+            'the "java" command on your system.'
         retval = 0
 
     if not config.has_key('lucene_bin') or not os.path.isdir(config['lucene_bin']):
-        print 'The "lucene_bin" property in the config file must refer to a directory that contains the lucene plugin scripts index.sh and search.sh.'
+        print 'The "lucene_bin" property in the config file must refer to a ' \
+            'directory that contains the lucene plugin scripts index.sh and ' \
+            'search.sh.'
+        retval = 0
+
+    if not config.has_key('lucene_highlight'):
+        print 'The "lucene_highlight" property in the config file must be '\
+            'set to either True or False'
         retval = 0
 
     return retval
@@ -107,15 +115,17 @@
     @type term: a string
     """
     urllib.quote(term)
-    JAVA_HOME=config['JAVA_HOME']
-    classpath = ':'.join(glob.glob(config['lucene_home']+'/*.jar')+[config['lucene_bin']])
+    JAVA_HOME = config['JAVA_HOME']
+    classpath = ':'.join(glob.glob(config['lucene_home'] + '/*.jar') \
+        + [config['lucene_bin']])
     index = config['lucene_index']
-    cmd =JAVA_HOME+' -cp '+classpath+' LuceneSearch '+index+' '+term
-    pipe = os.popen(cmd,'r')
+    cmd = JAVA_HOME + ' -cp ' + classpath + ' LuceneSearch ' + index + ' ' \
+        + term
+    pipe = os.popen(cmd, "r")
     results = pipe.readlines()
     status = pipe.close()
     results = [ os.path.join(config['datadir'], x[2:-1]) for x in results ]
-    entries = [ makeEntry(x, request) for x in results]
+    entries = [ makeEntry(x, request) for x in results ]
     entries = [ ( x._mtime, x ) for x in entries ]
     entries.sort()
     entries.reverse()
@@ -134,20 +144,20 @@
         return None
                                                                                 
     data = request.getData()
-
     resultnumber = len(data['luceneResults'])
 
     if resultnumber < 1:
-	data['searchHeader'] = "<div class=\"searchtext\">Your search returned no results. Maybe I just never talk about <b>" + form["search"].value + "</b>. Try again.</div><div>"
+	    data['searchHeader'] = "<div class=\"searchtext\">Your search " \
+            "returned no results. Maybe I just never talk about <b>" \
+            + form["search"].value + "</b></div>"
     else:
-	data['searchHeader'] = "<div class=\"searchtext\">Your search returned " + str(resultnumber) + " results for <b>" + form["search"].value + "</b>. They are below:</div>"
+    	data['searchHeader'] = "<div class=\"searchtext\">Your search " \
+            "returned " + str(resultnumber) + " results for <b>" \
+            + form["search"].value + "</b>:</div>"
 
 def cb_filelist(args):
     """
     Lucene search handling
-    
-    @param request: the Pyblosxom request
-    @type request: a Pyblosxom request object
     """
     # do nothing if the form is not a lucene form
     request = args["request"]
@@ -160,3 +170,33 @@
     
     data['luceneResults'] = search(request, config, form["search"].value)
     return data['luceneResults']
+
+def cb_postformat(args):
+    """
+    Highlight matches if lucene_highlight is set to True in config.py
+    """
+    # do nothing if the form is not a lucene form
+    request = args["request"]
+    form = request.getHttp()["form"]
+    if not form.has_key("search"):
+        return None
+
+    # do nothing if highlighting is not enabled
+    config = request.getConfiguration()
+    if not config["lucene_highlight"]:
+        return None
+
+    title = args["entry_data"]["title"]
+    body = args["entry_data"]["body"]
+
+    term = form["search"].value
+    term = re.sub('\*', '\w*', term)
+    term = re.sub('\?', '\w', term)
+    pattern = re.compile('\\b' + term + '\\b', re.IGNORECASE)
+    args["entry_data"]["title"] = pattern.sub(lambda match: \
+        "<span style='color:black;background-color:#ffff66'>" \
+        + match.group() + "</span>", title)
+    args["entry_data"]["body"] = pattern.sub(lambda match: \
+        "<span style='color:black;background-color:#ffff66'>" \
+        + match.group() + "</span>", body)
+    return args["entry_data"]