Move generate_notes_list to external file

1 files changed, 89 insertions, 0 deletions

diff --git a/generators/generators.py b/generators/generators.py
index 1247a1c5..1792fa57 100644
--- a/generators/generators.py
+++ b/generators/generators.py
@@ -5,6 +5,9 @@ import sys
 import tempfile
 import re
 
+import bs4
+import dateutil
+
 from config.config import *
 
 
@@ -195,3 +198,89 @@ def generate_blog_list(feed):
                     with open("build/blog/index.html", 'w', encoding='utf-8') as indexmd:
                         for line in lines:
                             indexmd.write(re.sub(r'{% generate blog_list here %}', div_blog_list, line))
+
+
+def generate_notes_list():
+    """"Generate notes list """
+
+    sys.stderr.write("generating notes list\n")
+
+    html_fileList = []
+    for root, dirs, files in os.walk(BUILDDIR):
+        for name in files:
+            if re.search(r'notes',root):
+                if name.endswith(".html"):
+                    try:
+                        html_fileList.append(os.path.join(root.split('notes/')[1], name))
+                    except IndexError:
+                        html_fileList.append(name)
+
+    div_notes_list = u'<div class="blog-index" id="toc">\n</table>\n'
+    year = 10000  # will be larger than the latest year for quite a while
+    # recall that entries are in reverse chronological order
+    table_opened = False
+    for name in list(reversed(sorted(os.listdir(os.path.join(BUILDDIR, "notes"))))):
+        if re.match(r"^[0-9]{4}-[0-9]{2}-[0-9]{2}.*\.html", name):
+            htmlpath = os.path.join(BUILDDIR, "notes", name)
+            #tentry = AtomEntry()
+            #item = RssItem()
+            try:
+                with open(htmlpath, encoding="utf-8") as htmlfile:
+                    soup = bs4.BeautifulSoup(htmlfile.read(), "lxml")
+                    # generate atom entry
+                    #entry.author = copy.deepcopy(feed.author)  # assume it's always the same author
+                    #entry_url = urllib.parse.urljoin(BLOG_HOME, "blog/%s" % name)
+                    #entry.id_text = entry_url
+                    #entry.id = ET.Element("id")
+                    #entry.id.text = entry_url
+                    relpath = "/notes/%s" % name
+
+                    #entry.link = ET.Element("link", href=entry_url)
+                    title_text = soup.title.text
+
+                    #entry.title = ET.Element("title", type="html")
+                    #entry.title.text = entry.title_text
+                    post_date = soup.find("meta", attrs={"name": "date"})["content"]
+                    updated_datetime = dateutil.parser.parse(post_date)
+
+                    date = updated_datetime
+                    if date.year < year:
+                        # close the previous table if there is one
+                        if table_opened:
+                            div_notes_list += u'</table>\n'
+                        # write a new <h2 class="blog-index-year-title"> tag with the smaller year
+                        year = date.year
+                        div_notes_list += u'\n<h2 class="blog-index-year-title" id="{0}"><span class="left-h2">.:</span><span class="title-h2">{0}</span><span class="right-h2">:.</span></h2>\n\n'.format(year)
+                        div_notes_list += u'<table class="blog-index-yearly-index">\n'
+                        table_opened = True
+
+                    # write a new table row entry in Markdown, in the format:
+                    #
+                    #   <tr>
+                    #     <td class="blog-index-post-date"><time class="date" datetime="2015-05-05T00:06:04-0700">May 5</time></td>
+                    #     <td class="blog-index-post-title">[Blah blah](/blog/2015-05-04-blah-blah.html)</td>
+                    #   </tr>
+                    monthday = date.strftime("%b %d")
+                    div_notes_list += (u'<tr><td class="blog-index-post-date"><time class="date" datetime="%s">%s</time></td>'
+                                  '<td class="blog-index-post-title"><a href="%s">%s</a></td></tr>\n' %
+                                  (date.isoformat(), monthday, relpath, title_text))
+
+            except Exception:
+                sys.stderr.write("error: failed to generate feed entry from %s\n" % name)
+                with open(htmlpath, encoding="utf-8") as htmlfile:
+                    sys.stderr.write("dumping HTML:%s\n\n" % htmlfile.read())
+                raise
+
+    if table_opened:
+        div_notes_list += u'</table>\n'
+    div_notes_list += u'</div>'
+
+    fd, tmppath = tempfile.mkstemp()
+    os.close(fd)
+    with open(tmppath, 'w', encoding='utf-8') as tmpfile:
+        if os.path.exists("build/notes/index.html"):
+            with open("build/notes/index.html", 'r', encoding='utf-8') as indexmd:
+                lines = indexmd.readlines()
+                with open("build/notes/index.html", 'w', encoding='utf-8') as indexmd:
+                    for line in lines:
+                        indexmd.write(re.sub(r'{% generate notes_list here %}', div_notes_list, line))