def generateAtomFeed(dom, out_dir):
rss_node = __getFirstNodeByCIName(dom, "__rss")
# don't continue, if there is no rss node
if rss_node is None:
return None
# iterate thru feed records
first = True
entries = ""
update_times = []
for node in rss_node.find("node"):
# skip first iteration (main node containing information about feed)
if first:
first = False
continue
# convert node from rich_text to html
html_node = d.parseString(convertToHtml(dom, node.params["unique_id"]))
if len(html_node.find("a")) > 0:
first_link = html_node.find("a")[0]
else:
raise ValueError(
"Item '" +
node.params["name"] +
"' doesn't have date and/or URL!"
)
updated = first_link.getContent()
# get url from first link, or set it to default
url = first_link.params["href"] if "href" in first_link.params else ""
url = "./" + url[5:] if url.startswith("./../") and len(url) > 5 else url
# remove first link (and it's content) from html code
if first_link is not None:
first_link.replaceWith(d.HTMLElement(""))
# preprocess content
content = html_node.getContent().replace("<p></p>", "").strip()
for key, val in HTML_ENTITIES.iteritems():
content = content.replace(val, key)
entries += Template(ATOM_ENTRY_TEMPLATE).substitute(
title=node.params["name"],
url=url,
uid=hashlib.md5(
node.params["name"] +
str(url) +
str(updated)
).hexdigest(),
updated=updated,
content=content
)
update_times.append(updated)
# remove node from DOM
node.replaceWith(d.HTMLElement(""))
# extract Atom template from .ctd
atom_template = rss_node.find("codebox")
if len(atom_template) <= 0:
raise ValueError("There is no codebox with Atom template!")
atom_template = atom_template[0].getContent()
atom_template = __removeHTMLEntities(atom_template)
atom_feed = Template(atom_template).substitute(
updated=update_times[0],
entries=entries
)
# get feed's filename - it is specified in atom template
filename = d.parseString(atom_feed).find("link")
if len(filename) <= 0:
raise ValueError("There has to be link in your Atom template!")
filename = filename[0]
if not "href" in filename.params:
raise ValueError(
"Link in your Atom template has to have 'href' parameter!"
)
filename = filename.params["href"].split("/")[-1]
if "." not in filename:
filename = "atom.xml"
writeln(
"You didn't specified filename of your feed, so I choosed " +
"'%s'" % (filename)
)
fh = open(out_dir + "/" + filename, "wt")
fh.write(atom_feed)
fh.close()
# get rid of RSS node
rss_node.replaceWith(d.HTMLElement(""))
def _transformLink(tag, dom, node_id, out_dir, root_path):
"""
Transform <rich_text link="webs http://kitakitsune.org">odkaz</rich_text>
to <a href="http://kitakitsune.org">odkaz</a>.
Also some basic link handling, ala local links and links to other nodes.
"""
if "link" in tag.params:
el = d.HTMLElement("<a>")
el.childs = tag.childs
# cherrytree puts string "webs "/"node " before every link for some
# reason
link = tag.params["link"]
link = link[5:]
if tag.params["link"].startswith("webs"):
# absolute path to local files
if link.startswith("http:///"):
link = link[7:]
# relative path to local files
if link.startswith("http://.."):
link = link[7:]
# relative path to local files in current directory
if link.startswith("http://./"):
link = link[7:]
elif tag.params["link"].startswith("file "):
link = base64.b64decode(tag.params["link"].split()[1])
# support for local images - I did tried to make it work as node,
# but that failed miserably, because there is limit to picture
# dimensions and other shitty crap
file_type = link.split(".")
pic_types = ["png", "gif", "jpg", "jpeg"]
if len(file_type) >= 1 and file_type[-1].lower() in pic_types:
directory = out_dir + "/pictures"
if not os.path.exists(directory):
os.makedirs(directory)
local_name = "%s/%s_%s" % (
directory,
hashlib.md5(link).hexdigest(),
os.path.basename(link)
)
shutil.copyfile(link, local_name)
elif tag.params["link"].startswith("node "):
# internal links contains only node id
link_id = link.strip()
# get nodename
linked_nodename = dom.find("node", {"unique_id": str(link_id)})
if not linked_nodename:
writeln("Broken link to node ID '" + link_id + "'", sys.stderr)
link = "[broken link to internal node]"
else:
# get (this) node depth
depth = len(getNodePath(dom, node_id).split("/")) - 1
link = "./" + (depth * "../") + getNodePath(dom, link_id)
el.params["href"] = link.strip()
el.endtag = d.HTMLElement("</a>")
tag.replaceWith(el)