[gimp-web/testing: 5/7] Add tipue_search plugin

[Pat David <patdavid src gnome org>]

commit b34736b1f503952813e3d1d7e42e4d4fb0f15b39
Author: Pat David <patdavid gmail com>
Date:   Fri May 27 16:45:43 2016 -0500

    Add tipue_search plugin
    
    Add the tipue_search plugin that aggregates site content
    into a single .js file at the document root.
    
    Pat David had to modify the output to prepend the JSON
    output with a javascript var (var tipuesearch = ) so that
    we can use the "Static" method of the tipue search (fastest).

 plugins/tipue_search/__init__.py     |    1 +
 plugins/tipue_search/tipue_search.py |  116 ++++++++++++++++++++++++++++++++++
 2 files changed, 117 insertions(+), 0 deletions(-)
---
diff --git a/plugins/tipue_search/__init__.py b/plugins/tipue_search/__init__.py
new file mode 100644
index 0000000..ebd6c06
--- /dev/null
+++ b/plugins/tipue_search/__init__.py
@@ -0,0 +1 @@
+from .tipue_search import *
diff --git a/plugins/tipue_search/tipue_search.py b/plugins/tipue_search/tipue_search.py
new file mode 100644
index 0000000..54471c3
--- /dev/null
+++ b/plugins/tipue_search/tipue_search.py
@@ -0,0 +1,116 @@
+# -*- coding: utf-8 -*-
+"""
+Tipue Search
+============
+
+A Pelican plugin to serialize generated HTML to JSON
+that can be used by jQuery plugin - Tipue Search.
+
+Copyright (c) Talha Mansoor
+"""
+
+from __future__ import unicode_literals
+
+import os.path
+import json
+from bs4 import BeautifulSoup
+from codecs import open
+try:
+    from urlparse import urljoin
+except ImportError:
+    from urllib.parse import urljoin
+
+from pelican import signals
+
+
+class Tipue_Search_JSON_Generator(object):
+
+    def __init__(self, context, settings, path, theme, output_path, *null):
+
+        self.output_path = output_path
+        self.context = context
+        self.siteurl = settings.get('SITEURL')
+        self.tpages = settings.get('TEMPLATE_PAGES')
+        self.output_path = output_path
+        self.json_nodes = []
+
+
+    def create_json_node(self, page):
+
+        if getattr(page, 'status', 'published') != 'published':
+            return
+
+        soup_title = BeautifulSoup(page.title.replace('&nbsp;', ' '), 'html.parser')
+        page_title = soup_title.get_text(' ', strip=True).replace('“', '"').replace('”', '"').replace('’', 
"'").replace('^', '&#94;')
+
+        soup_text = BeautifulSoup(page.content, 'html.parser')
+        page_text = soup_text.get_text(' ', strip=True).replace('“', '"').replace('”', '"').replace('’', 
"'").replace('¶', ' ').replace('^', '&#94;')
+        page_text = ' '.join(page_text.split())
+
+        if getattr(page, 'category', 'None') == 'None':
+            page_category = ''
+        else:
+            page_category = page.category.name
+
+        page_url = self.siteurl + '/' + page.url
+
+        node = {'title': page_title,
+                'text': page_text,
+                'tags': page_category,
+                'url': page_url}
+
+        self.json_nodes.append(node)
+
+
+    def create_tpage_node(self, srclink):
+
+        srcfile = open(os.path.join(self.output_path, self.tpages[srclink]), encoding='utf-8')
+        soup = BeautifulSoup(srcfile, 'html.parser')
+        page_text = soup.get_text()
+
+        # What happens if there is not a title.
+        if soup.title is not None:
+            page_title = soup.title.string
+        else:
+            page_title = ''
+
+        # Should set default category?
+        page_category = ''
+
+        page_url = urljoin(self.siteurl, self.tpages[srclink])
+
+        node = {'title': page_title,
+                'text': page_text,
+                'tags': page_category,
+                'url': page_url}
+
+        self.json_nodes.append(node)
+
+
+    def generate_output(self, writer):
+        path = os.path.join(self.output_path, 'tipuesearch_content.json')
+
+        pages = self.context['pages'] + self.context['articles']
+
+        for article in self.context['articles']:
+            pages += article.translations
+
+        for srclink in self.tpages:
+            self.create_tpage_node(srclink)
+
+        for page in pages:
+            self.create_json_node(page)
+        root_node = {'pages': self.json_nodes}
+
+        with open(path, 'w', encoding='utf-8') as fd:
+            # Added below line so you can use tipuesearch static (vs. json)
+            fd.write("var tipuesearch = ")
+            json.dump(root_node, fd, separators=(',', ':'), ensure_ascii=False)
+
+
+def get_generators(generators):
+    return Tipue_Search_JSON_Generator
+
+
+def register():
+    signals.get_generators.connect(get_generators)