summary plugin added

6 years ago · baffbbbab9
4 changed files with 258 additions and 0 deletions
--- a/plugins/summary/Readme.rst
+++ b/plugins/summary/Readme.rst
@ -0,0 +1,56 @@
+Summary
+-------
+
+This plugin allows easy, variable length summaries directly embedded into the
+body of your articles. It introduces two new settings: ``SUMMARY_BEGIN_MARKER``
+and ``SUMMARY_END_MARKER``: strings which can be placed directly into an article
+to mark the beginning and end of a summary. When found, the standard
+``SUMMARY_MAX_LENGTH`` setting will be ignored. The markers themselves will also
+be removed from your articles before they are published. The default values
+are ``<!-- PELICAN_BEGIN_SUMMARY -->`` and ``<!-- PELICAN_END_SUMMARY -->``.
+For example::
+
+    Title: My super title
+    Date: 2010-12-03 10:20
+    Tags: thats, awesome
+    Category: yeah
+    Slug: my-super-post
+    Author: Alexis Metaireau
+
+    This is the content of my super blog post.
+    <!-- PELICAN_END_SUMMARY -->
+    and this content occurs after the summary.
+
+Here, the summary is taken to be the first line of the post. Because no
+beginning marker was found, it starts at the top of the body. It is possible
+to leave out the end marker instead, in which case the summary will start at the
+beginning marker and continue to the end of the body.
+
+If no beginning or end marker is found, and if ``SUMMARY_USE_FIRST_PARAGRAPH``
+is enabled in the settings, the summary will be the first paragraph of the post.
+
+The plugin also sets a ``has_summary`` attribute on every article. It is True
+for articles with an explicitly-defined summary, and False otherwise.  (It is
+also False for an article truncated by ``SUMMARY_MAX_LENGTH``.)  Your templates
+can use this e.g. to add a link to the full text at the end of the summary.
+
+reST example
+~~~~~~~~~~~~
+
+Inserting the markers into a reStructuredText document makes use of the
+comment directive, because raw HTML is automatically escaped. The reST equivalent of the above Markdown example looks like this::
+
+    My super title
+    ##############
+
+    :date: 2010-12-03 10:20
+    :tags: thats, awesome
+    :category: yeah
+    :slug: my-super-post
+    :author: Alexis Metaireau
+
+    This is the content of my super blog post.
+
+    .. PELICAN_END_SUMMARY
+
+    and this content occurs after the summary.
--- a/plugins/summary/init.py
+++ b/plugins/summary/init.py
@ -0,0 +1 @@
+from .summary import *
--- a/plugins/summary/summary.py
+++ b/plugins/summary/summary.py
@ -0,0 +1,105 @@
+"""
+Summary
+-------
+
+This plugin allows easy, variable length summaries directly embedded into the
+body of your articles.
+"""
+
+from __future__ import unicode_literals
+from pelican import signals
+from pelican.generators import ArticlesGenerator, StaticGenerator, PagesGenerator
+import re
+
+def initialized(pelican):
+    from pelican.settings import DEFAULT_CONFIG
+    DEFAULT_CONFIG.setdefault('SUMMARY_BEGIN_MARKER',
+                              '<!-- PELICAN_BEGIN_SUMMARY -->')
+    DEFAULT_CONFIG.setdefault('SUMMARY_END_MARKER',
+                              '<!-- PELICAN_END_SUMMARY -->')
+    DEFAULT_CONFIG.setdefault('SUMMARY_USE_FIRST_PARAGRAPH', False)
+    if pelican:
+        pelican.settings.setdefault('SUMMARY_BEGIN_MARKER',
+                                    '<!-- PELICAN_BEGIN_SUMMARY -->')
+        pelican.settings.setdefault('SUMMARY_END_MARKER',
+                                    '<!-- PELICAN_END_SUMMARY -->')
+        pelican.settings.setdefault('SUMMARY_USE_FIRST_PARAGRAPH', False)
+
+def extract_summary(instance):
+    # if summary is already specified, use it
+    # if there is no content, there's nothing to do
+    if hasattr(instance, '_summary'):
+        instance.has_summary = True
+        return
+
+    if not instance._content:
+        instance.has_summary = False
+        return
+
+    begin_marker = instance.settings['SUMMARY_BEGIN_MARKER']
+    end_marker   = instance.settings['SUMMARY_END_MARKER']
+    use_first_paragraph = instance.settings['SUMMARY_USE_FIRST_PARAGRAPH']
+    remove_markers = True
+
+    content = instance._content
+    begin_summary = -1
+    end_summary = -1
+    if begin_marker:
+        begin_summary = content.find(begin_marker)
+    if end_marker:
+        end_summary = content.find(end_marker)
+
+    if begin_summary == -1 and end_summary == -1 and use_first_paragraph:
+        begin_marker, end_marker = '<p>', '</p>'
+        remove_markers = False
+        begin_summary = content.find(begin_marker)
+        end_summary = content.find(end_marker)
+
+    if begin_summary == -1 and end_summary == -1:
+        instance.has_summary = False
+        return
+
+    # skip over the begin marker, if present
+    if begin_summary == -1:
+        begin_summary = 0
+    else:
+        begin_summary = begin_summary + len(begin_marker)
+
+    if end_summary == -1:
+        end_summary = None
+
+    summary = content[begin_summary:end_summary]
+
+    if remove_markers:
+        # remove the markers from the content
+        if begin_summary:
+            content = content.replace(begin_marker, '', 1)
+        if end_summary:
+            content = content.replace(end_marker, '', 1)
+
+    summary = re.sub(r"<div.*>", "", summary)
+    summary = re.sub(r"</div>", "", summary)
+
+    instance._content = content
+    instance._summary = summary
+    instance.has_summary = True
+
+
+def run_plugin(generators):
+    for generator in generators:
+        if isinstance(generator, ArticlesGenerator):
+            for article in generator.articles:
+                extract_summary(article)
+        elif isinstance(generator, PagesGenerator):
+            for page in generator.pages:
+                extract_summary(page)
+
+
+def register():
+    signals.initialized.connect(initialized)
+    try:
+        signals.all_generators_finalized.connect(run_plugin)
+    except AttributeError:
+        # NOTE: This results in #314 so shouldn't really be relied on
+        # https://github.com/getpelican/pelican-plugins/issues/314
+        signals.content_object_init.connect(extract_summary)
--- a/plugins/summary/test_summary.py
+++ b/plugins/summary/test_summary.py
@ -0,0 +1,96 @@
+# -*- coding: utf-8 -*-
+
+import unittest
+
+from jinja2.utils import generate_lorem_ipsum
+
+# generate one paragraph, enclosed with <p>
+TEST_CONTENT = str(generate_lorem_ipsum(n=1))
+TEST_SUMMARY = generate_lorem_ipsum(n=1, html=False)
+
+
+from pelican.contents import Page
+import pelican.settings
+
+import summary
+
+class TestSummary(unittest.TestCase):
+    def setUp(self):
+        super(TestSummary, self).setUp()
+        pelican.settings.DEFAULT_CONFIG['SUMMARY_MAX_LENGTH'] = None
+        pelican.settings.DEFAULT_CONFIG['SUMMARY_USE_FIRST_PARAGRAPH'] = False
+
+        summary.register()
+        summary.initialized(None)
+        self.page_kwargs = {
+            'content': TEST_CONTENT,
+            'context': {
+                'localsiteurl': '',
+            },
+            'metadata': {
+                'summary': TEST_SUMMARY,
+                'title': 'foo bar',
+                'author': 'Blogger',
+            },
+        }
+
+    def _copy_page_kwargs(self):
+        # make a deep copy of page_kwargs
+        page_kwargs = dict([(key, self.page_kwargs[key]) for key in
+                            self.page_kwargs])
+        for key in page_kwargs:
+            if not isinstance(page_kwargs[key], dict):
+                break
+            page_kwargs[key] = dict([(subkey, page_kwargs[key][subkey])
+                                     for subkey in page_kwargs[key]])
+
+        return page_kwargs
+
+    def test_end_summary(self):
+        page_kwargs = self._copy_page_kwargs()
+        del page_kwargs['metadata']['summary']
+        page_kwargs['content'] = (
+            TEST_SUMMARY + '<!-- PELICAN_END_SUMMARY -->' + TEST_CONTENT)
+        page = Page(**page_kwargs)
+        summary.extract_summary(page)
+        # test both the summary and the marker removal
+        self.assertEqual(page.summary, TEST_SUMMARY)
+        self.assertEqual(page.content, TEST_SUMMARY + TEST_CONTENT)
+
+    def test_begin_summary(self):
+        page_kwargs = self._copy_page_kwargs()
+        del page_kwargs['metadata']['summary']
+        page_kwargs['content'] = (
+            'FOOBAR<!-- PELICAN_BEGIN_SUMMARY -->' + TEST_CONTENT)
+        page = Page(**page_kwargs)
+        summary.extract_summary(page)
+        # test both the summary and the marker removal
+        self.assertEqual(page.summary, TEST_CONTENT)
+        self.assertEqual(page.content, 'FOOBAR' + TEST_CONTENT)
+
+    def test_begin_end_summary(self):
+        page_kwargs = self._copy_page_kwargs()
+        del page_kwargs['metadata']['summary']
+        page_kwargs['content'] = (
+                'FOOBAR<!-- PELICAN_BEGIN_SUMMARY -->' + TEST_SUMMARY +
+                '<!-- PELICAN_END_SUMMARY -->' + TEST_CONTENT)
+        page = Page(**page_kwargs)
+        summary.extract_summary(page)
+        # test both the summary and the marker removal
+        self.assertEqual(page.summary, TEST_SUMMARY)
+        self.assertEqual(page.content, 'FOOBAR' + TEST_SUMMARY + TEST_CONTENT)
+
+    def test_use_first_paragraph(self):
+        page_kwargs = self._copy_page_kwargs()
+        del page_kwargs['metadata']['summary']
+        pelican.settings.DEFAULT_CONFIG['SUMMARY_USE_FIRST_PARAGRAPH'] = True
+        page_kwargs['content'] = '<p>' + TEST_SUMMARY + '</p>' + TEST_CONTENT
+        page = Page(**page_kwargs)
+        summary.extract_summary(page)
+        # test both the summary and the marker removal
+        self.assertEqual(page.summary, TEST_SUMMARY)
+        self.assertEqual(page.content, '<p>' + TEST_SUMMARY + '</p>' + TEST_CONTENT)
+
+
+if __name__ == '__main__':
+    unittest.main()