def apply(doc):
output_dir = config.output_dir()
if not os.path.exists(output_dir):
os.makedirs(output_dir)
log = planet.logger
planet_filters = config.filters('Planet')
# Go-go-gadget-template
for template_file in config.template_files():
output_file = shell.run(template_file, doc)
# run any template specific filters
if config.filters(template_file) != planet_filters:
output = open(output_file).read()
for filter in config.filters(template_file):
if filter in planet_filters: continue
if filter.find('>') > 0:
# tee'd output
filter, dest = filter.split('>', 1)
tee = shell.run(filter.strip(), output, mode="filter")
if tee:
output_dir = planet.config.output_dir()
dest_file = os.path.join(output_dir, dest.strip())
dest_file = open(dest_file, 'w')
dest_file.write(tee)
dest_file.close()
else:
# pipe'd output
output = shell.run(filter, output, mode="filter")
if not output:
os.unlink(output_file)
break
else:
handle = open(output_file, 'w')
handle.write(output)
handle.close()
# Process bill of materials
for copy_file in config.bill_of_materials():
dest = os.path.join(output_dir, copy_file)
for template_dir in config.template_directories():
source = os.path.join(template_dir, copy_file)
if os.path.exists(source): break
else:
log.error('Unable to locate %s', copy_file)
log.info("Template search path:")
for template_dir in config.template_directories():
log.info(" %s", os.path.realpath(template_dir))
continue
mtime = os.stat(source).st_mtime
if not os.path.exists(dest) or os.stat(dest).st_mtime < mtime:
dest_dir = os.path.split(dest)[0]
if not os.path.exists(dest_dir): os.makedirs(dest_dir)
log.info("Copying %s to %s", source, dest)
if os.path.exists(dest): os.chmod(dest, 0644)
shutil.copyfile(source, dest)
shutil.copystat(source, dest)
def apply(doc):
output_dir = config.output_dir()
if not os.path.exists(output_dir): os.makedirs(output_dir)
log = planet.getLogger(config.log_level(),config.log_format())
# Go-go-gadget-template
for template_file in config.template_files():
shell.run(template_file, doc)
# Process bill of materials
for copy_file in config.bill_of_materials():
dest = os.path.join(output_dir, copy_file)
for template_dir in config.template_directories():
source = os.path.join(template_dir, copy_file)
if os.path.exists(source): break
else:
log.error('Unable to locate %s', copy_file)
continue
mtime = os.stat(source).st_mtime
if not os.path.exists(dest) or os.stat(dest).st_mtime < mtime:
dest_dir = os.path.split(dest)[0]
if not os.path.exists(dest_dir): os.makedirs(dest_dir)
log.info("Copying %s to %s", source, dest)
shutil.copyfile(source, dest)
shutil.copystat(source, dest)
def apply(doc):
output_dir = config.output_dir()
if not os.path.exists(output_dir): os.makedirs(output_dir)
log = planet.logger
planet_filters = config.filters('Planet')
# Go-go-gadget-template
for template_file in config.template_files():
output_file = shell.run(template_file, doc)
# run any template specific filters
if config.filters(template_file) != planet_filters:
output = open(output_file).read()
for filter in config.filters(template_file):
if filter in planet_filters: continue
if filter.find('>')>0:
# tee'd output
filter,dest = filter.split('>',1)
tee = shell.run(filter.strip(), output, mode="filter")
if tee:
output_dir = planet.config.output_dir()
dest_file = os.path.join(output_dir, dest.strip())
dest_file = open(dest_file,'w')
dest_file.write(tee)
dest_file.close()
else:
# pipe'd output
output = shell.run(filter, output, mode="filter")
if not output:
os.unlink(output_file)
break
else:
handle = open(output_file,'w')
handle.write(output)
handle.close()
# Process bill of materials
for copy_file in config.bill_of_materials():
dest = os.path.join(output_dir, copy_file)
for template_dir in config.template_directories():
source = os.path.join(template_dir, copy_file)
if os.path.exists(source): break
else:
log.error('Unable to locate %s', copy_file)
log.info("Template search path:")
for template_dir in config.template_directories():
log.info(" %s", os.path.realpath(template_dir))
continue
mtime = os.stat(source).st_mtime
if not os.path.exists(dest) or os.stat(dest).st_mtime < mtime:
dest_dir = os.path.split(dest)[0]
if not os.path.exists(dest_dir): os.makedirs(dest_dir)
log.info("Copying %s to %s", source, dest)
if os.path.exists(dest): os.chmod(dest, 0644)
shutil.copyfile(source, dest)
shutil.copystat(source, dest)
def load(config_files):
""" initialize and load a configuration"""
global parser
parser = ConfigParser()
parser.read(config_files)
import config, planet
from planet import opml, foaf, csv_config
log = planet.logger
if not log:
log = planet.getLogger(config.log_level(),config.log_format())
# Theme support
theme = config.output_theme()
if theme:
for path in ("", os.path.join(sys.path[0],'themes')):
theme_dir = os.path.join(path,theme)
theme_file = os.path.join(theme_dir,'config.ini')
if os.path.exists(theme_file):
# initial search list for theme directories
dirs = config.template_directories()
if theme_dir not in dirs:
dirs.append(theme_dir)
if not hasattr(config_files, 'append'):
config_files = [config_files]
for config_file in config_files:
if os.path.dirname(config_file) not in dirs:
dirs.append(os.path.dirname(config_file))
# read in the theme
parser = ConfigParser()
parser.read(theme_file)
bom = config.bill_of_materials()
# complete search list for theme directories
dirs += [os.path.join(theme_dir,dir) for dir in
config.template_directories() if dir not in dirs]
# merge configurations, allowing current one to override theme
template_files = config.template_files()
parser.set('Planet','template_files','')
parser.read(config_files)
for file in config.bill_of_materials():
if not file in bom: bom.append(file)
parser.set('Planet', 'bill_of_materials', ' '.join(bom))
parser.set('Planet', 'template_directories', ' '.join(dirs))
parser.set('Planet', 'template_files',
' '.join(template_files + config.template_files()))
break
else:
log.error('Unable to find theme %s', theme)
# Filter support
dirs = config.filter_directories()
filter_dir = os.path.join(sys.path[0],'filters')
if filter_dir not in dirs and os.path.exists(filter_dir):
parser.set('Planet', 'filter_directories', ' '.join(dirs+[filter_dir]))
# Reading list support
reading_lists = config.reading_lists()
if reading_lists:
if not os.path.exists(config.cache_lists_directory()):
os.makedirs(config.cache_lists_directory())
def data2config(data, cached_config):
if content_type(list).find('opml')>=0:
opml.opml2config(data, cached_config)
elif content_type(list).find('foaf')>=0:
foaf.foaf2config(data, cached_config)
elif content_type(list).find('csv')>=0:
csv_config.csv2config(data, cached_config)
elif content_type(list).find('config')>=0:
cached_config.readfp(data)
else:
from planet import shell
import StringIO
cached_config.readfp(StringIO.StringIO(shell.run(
content_type(list), data.getvalue(), mode="filter")))
if cached_config.sections() in [[], [list]]:
raise Exception
for list in reading_lists:
downloadReadingList(list, parser, data2config)
def load(config_file):
""" initialize and load a configuration"""
global parser
parser = ConfigParser()
parser.read(config_file)
import config, planet
from planet import opml, foaf, csv_config
log = planet.logger
if not log:
log = planet.getLogger(config.log_level(),config.log_format())
# Theme support
theme = config.output_theme()
if theme:
for path in ("", os.path.join(sys.path[0],'themes')):
theme_dir = os.path.join(path,theme)
theme_file = os.path.join(theme_dir,'config.ini')
if os.path.exists(theme_file):
# initial search list for theme directories
dirs = config.template_directories()
if theme_dir not in dirs:
dirs.append(theme_dir)
if os.path.dirname(config_file) not in dirs:
dirs.append(os.path.dirname(config_file))
# read in the theme
parser = ConfigParser()
parser.read(theme_file)
bom = config.bill_of_materials()
# complete search list for theme directories
dirs += [os.path.join(theme_dir,dir) for dir in
config.template_directories() if dir not in dirs]
# merge configurations, allowing current one to override theme
template_files = config.template_files()
parser.set('Planet','template_files','')
parser.read(config_file)
for file in config.bill_of_materials():
if not file in bom: bom.append(file)
parser.set('Planet', 'bill_of_materials', ' '.join(bom))
parser.set('Planet', 'template_directories', ' '.join(dirs))
parser.set('Planet', 'template_files',
' '.join(template_files + config.template_files()))
break
else:
log.error('Unable to find theme %s', theme)
# Filter support
dirs = config.filter_directories()
filter_dir = os.path.join(sys.path[0],'filters')
if filter_dir not in dirs and os.path.exists(filter_dir):
parser.set('Planet', 'filter_directories', ' '.join(dirs+[filter_dir]))
# Reading list support
reading_lists = config.reading_lists()
if reading_lists:
if not os.path.exists(config.cache_lists_directory()):
os.makedirs(config.cache_lists_directory())
def data2config(data, cached_config):
if content_type(list).find('opml')>=0:
opml.opml2config(data, cached_config)
elif content_type(list).find('foaf')>=0:
foaf.foaf2config(data, cached_config)
elif content_type(list).find('csv')>=0:
csv_config.csv2config(data, cached_config)
elif content_type(list).find('config')>=0:
cached_config.readfp(data)
else:
from planet import shell
import StringIO
cached_config.readfp(StringIO.StringIO(shell.run(
content_type(list), data.getvalue(), mode="filter")))
if cached_config.sections() in [[], [list]]:
raise Exception
for list in reading_lists:
downloadReadingList(list, parser, data2config)
def splice():
""" Splice together a planet from a cache of entries """
import planet
log = planet.logger
log.info("Loading cached data")
cache = config.cache_directory()
dir = [(os.stat(file).st_mtime, file) for file in glob.glob(cache + "/*")
if not os.path.isdir(file)]
dir.sort()
dir.reverse()
max_items = max([
config.items_per_page(templ)
for templ in config.template_files() or ['Planet']
])
doc = minidom.parseString('<feed xmlns="http://www.w3.org/2005/Atom"/>')
feed = doc.documentElement
# insert feed information
createTextElement(feed, 'title', config.name())
date(feed, 'updated', time.gmtime())
gen = createTextElement(feed, 'generator', config.generator())
gen.setAttribute('uri', config.generator_uri())
author = doc.createElement('author')
createTextElement(author, 'name', config.owner_name())
createTextElement(author, 'email', config.owner_email())
feed.appendChild(author)
if config.feed():
createTextElement(feed, 'id', config.feed())
link = doc.createElement('link')
link.setAttribute('rel', 'self')
link.setAttribute('href', config.feed())
if config.feedtype():
link.setAttribute('type', "application/%s+xml" % config.feedtype())
feed.appendChild(link)
if config.link():
link = doc.createElement('link')
link.setAttribute('rel', 'alternate')
link.setAttribute('href', config.link())
feed.appendChild(link)
# insert subscription information
sub_ids = []
feed.setAttribute('xmlns:planet', planet.xmlns)
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data = feedparser.parse(filename(sources, sub))
if data.feed.has_key('id'): sub_ids.append(data.feed.id)
if not data.feed: continue
xdoc = minidom.parseString('''<planet:source xmlns:planet="%s"
xmlns="http://www.w3.org/2005/Atom"/>\n''' % planet.xmlns)
reconstitute.source(xdoc.documentElement, data.feed, None, None)
feed.appendChild(xdoc.documentElement)
index = idindex.open()
# insert entry information
items = 0
count = {}
new_feed_items = config.new_feed_items()
for mtime, file in dir:
if index != None:
base = os.path.basename(file)
if index.has_key(base) and index[base] not in sub_ids: continue
try:
entry = minidom.parse(file)
# verify that this entry is currently subscribed to and that the
# number of entries contributed by this feed does not exceed
# config.new_feed_items
entry.normalize()
sources = entry.getElementsByTagName('source')
if sources:
ids = sources[0].getElementsByTagName('id')
if ids:
id = ids[0].childNodes[0].nodeValue
count[id] = count.get(id, 0) + 1
if new_feed_items and count[id] > new_feed_items: continue
if id not in sub_ids:
ids = sources[0].getElementsByTagName('planet:id')
if not ids: continue
id = ids[0].childNodes[0].nodeValue
if id not in sub_ids: continue
# add entry to feed
feed.appendChild(entry.documentElement)
items = items + 1
if items >= max_items: break
except:
log.error("Error parsing %s", file)
if index: index.close()
return doc
def splice():
""" Splice together a planet from a cache of entries """
import planet
log = planet.logger
log.info("Loading cached data")
cache = config.cache_directory()
dir=[(os.stat(file).st_mtime,file) for file in glob.glob(cache+"/*")
if not os.path.isdir(file)]
dir.sort()
dir.reverse()
max_items=max([config.items_per_page(templ)
for templ in config.template_files() or ['Planet']])
doc = minidom.parseString('<feed xmlns="http://www.w3.org/2005/Atom"/>')
feed = doc.documentElement
# insert feed information
createTextElement(feed, 'title', config.name())
date(feed, 'updated', time.gmtime())
gen = createTextElement(feed, 'generator', config.generator())
gen.setAttribute('uri', config.generator_uri())
author = doc.createElement('author')
createTextElement(author, 'name', config.owner_name())
createTextElement(author, 'email', config.owner_email())
feed.appendChild(author)
if config.feed():
createTextElement(feed, 'id', config.feed())
link = doc.createElement('link')
link.setAttribute('rel', 'self')
link.setAttribute('href', config.feed())
if config.feedtype():
link.setAttribute('type', "application/%s+xml" % config.feedtype())
feed.appendChild(link)
if config.pubsubhubbub_hub():
hub = doc.createElement('link')
hub.setAttribute('rel', 'hub')
hub.setAttribute('href', config.pubsubhubbub_hub())
feed.appendChild(hub)
if config.link():
link = doc.createElement('link')
link.setAttribute('rel', 'alternate')
link.setAttribute('href', config.link())
feed.appendChild(link)
# insert subscription information
sub_ids = []
feed.setAttribute('xmlns:planet',planet.xmlns)
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data=feedparser.parse(filename(sources,sub))
if data.feed.has_key('id'): sub_ids.append(data.feed.id)
if not data.feed: continue
# warn on missing links
if not data.feed.has_key('planet_message'):
if not data.feed.has_key('links'): data.feed['links'] = []
for link in data.feed.links:
if link.rel == 'self': break
else:
log.debug('missing self link for ' + sub)
for link in data.feed.links:
if link.rel == 'alternate' and 'html' in link.type: break
else:
log.debug('missing html link for ' + sub)
xdoc=minidom.parseString('''<planet:source xmlns:planet="%s"
xmlns="http://www.w3.org/2005/Atom"/>\n''' % planet.xmlns)
reconstitute.source(xdoc.documentElement, data.feed, None, None)
feed.appendChild(xdoc.documentElement)
index = idindex.open()
# insert entry information
items = 0
count = {}
atomNS='http://www.w3.org/2005/Atom'
new_feed_items = config.new_feed_items()
for mtime,file in dir:
if index != None:
base = os.path.basename(file)
if index.has_key(base) and index[base] not in sub_ids: continue
try:
entry=minidom.parse(file)
# verify that this entry is currently subscribed to and that the
# number of entries contributed by this feed does not exceed
# config.new_feed_items
entry.normalize()
sources = entry.getElementsByTagNameNS(atomNS, 'source')
if sources:
ids = sources[0].getElementsByTagName('id')
if ids:
id = ids[0].childNodes[0].nodeValue
count[id] = count.get(id,0) + 1
if new_feed_items and count[id] > new_feed_items: continue
if id not in sub_ids:
ids = sources[0].getElementsByTagName('planet:id')
if not ids: continue
id = ids[0].childNodes[0].nodeValue
if id not in sub_ids:
log.warn('Skipping: ' + id)
if id not in sub_ids: continue
# add entry to feed
feed.appendChild(entry.documentElement)
items = items + 1
if items >= max_items: break
except:
log.error("Error parsing %s", file)
if index: index.close()
return doc
def splice():
""" Splice together a planet from a cache of entries """
import planet
log = planet.logger
log.info("Loading cached data")
cache = config.cache_directory()
dir = [(os.stat(file).st_mtime, file) for file in glob.glob(cache + "/*")
if not os.path.isdir(file)]
dir.sort()
dir.reverse()
max_items = max([
config.items_per_page(templ)
for templ in config.template_files() or ['Planet']
])
doc = minidom.parseString('<feed xmlns="http://www.w3.org/2005/Atom"/>')
feed = doc.documentElement
# insert feed information
createTextElement(feed, 'title', config.name())
date(feed, 'updated', time.gmtime())
gen = createTextElement(feed, 'generator', config.generator())
gen.setAttribute('uri', config.generator_uri())
author = doc.createElement('author')
createTextElement(author, 'name', config.owner_name())
createTextElement(author, 'email', config.owner_email())
feed.appendChild(author)
if config.feed():
createTextElement(feed, 'id', config.feed())
link = doc.createElement('link')
link.setAttribute('rel', 'self')
link.setAttribute('href', config.feed())
if config.feedtype():
link.setAttribute('type', "application/%s+xml" % config.feedtype())
feed.appendChild(link)
if config.pubsubhubbub_hub():
hub = doc.createElement('link')
hub.setAttribute('rel', 'hub')
hub.setAttribute('href', config.pubsubhubbub_hub())
feed.appendChild(hub)
if config.link():
link = doc.createElement('link')
link.setAttribute('rel', 'alternate')
link.setAttribute('href', config.link())
feed.appendChild(link)
# insert subscription information
sub_ids = []
feed.setAttribute('xmlns:planet', planet.xmlns)
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data = feedparser.parse(filename(sources, sub))
if data.feed.has_key('id'):
sub_ids.append(data.feed.id)
if not data.feed:
continue
# warn on missing links
if not data.feed.has_key('planet_message'):
if not data.feed.has_key('links'):
data.feed['links'] = []
for link in data.feed.links:
if link.rel == 'self':
break
else:
log.debug('missing self link for ' + sub)
for link in data.feed.links:
if link.rel == 'alternate' and 'html' in link.type:
break
else:
log.debug('missing html link for ' + sub)
xdoc = minidom.parseString('''<planet:source xmlns:planet="%s"
xmlns="http://www.w3.org/2005/Atom"/>\n''' % planet.xmlns)
reconstitute.source(xdoc.documentElement, data.feed, None, None)
feed.appendChild(xdoc.documentElement)
index = idindex.open()
# insert entry information
items = 0
count = {}
atomNS = 'http://www.w3.org/2005/Atom'
new_feed_items = config.new_feed_items()
posted_urls = set()
if config.post_to_twitter():
if os.path.exists(posted_urls_file):
try:
with open(posted_urls_file, 'rb') as f:
posted_urls = pickle.load(f)
except Exception as ex:
log.error("Error reading posted_urls %s", ex)
# print(posted_urls)
for mtime, file in dir:
if index != None:
base = os.path.basename(file)
if index.has_key(base) and index[base] not in sub_ids:
continue
try:
entry = minidom.parse(file)
# verify that this entry is currently subscribed to and that the
# number of entries contributed by this feed does not exceed
# config.new_feed_items
entry.normalize()
sources = entry.getElementsByTagNameNS(atomNS, 'source')
if sources:
ids = sources[0].getElementsByTagName('id')
if ids:
id = ids[0].childNodes[0].nodeValue
count[id] = count.get(id, 0) + 1
if new_feed_items and count[id] > new_feed_items:
continue
if id not in sub_ids:
ids = sources[0].getElementsByTagName('planet:id')
if not ids:
continue
id = ids[0].childNodes[0].nodeValue
if id not in sub_ids:
log.warn('Skipping: ' + id)
if id not in sub_ids:
continue
# Twitter integration
if config.post_to_twitter():
url = None
twitter = None
title = "Untitled post..."
links = entry.getElementsByTagName('link')
if links:
for link in links:
if link.hasAttribute('rel') and link.hasAttribute(
'type') and link.hasAttribute('href'):
if (link.getAttribute('rel') == 'alternate' and
link.getAttribute('type') == 'text/html'):
url = link.getAttribute('href')
break
titles = entry.getElementsByTagName('title')
if titles:
title = unicode(
titles[0].firstChild.nodeValue.encode('utf-8'),
'utf-8').strip()
handles = entry.getElementsByTagName('planet:twitter')
if (handles):
twitter = unicode(
handles[0].firstChild.nodeValue.encode('utf-8'),
"utf-8")
if url is not None and url not in posted_urls:
# log.debug("Going to post URL to Twitter: twitter='{}' title='{}', url='{}'".format(twitter, title, url))
txt_append = u''
if twitter:
txt_append = u" (by @" + twitter.encode(
'utf-8').strip() + u")"
max_title_len = 280 - 20 - len(txt_append)
if (len(title) > max_title_len):
title = title[:max_title_len]
txt = title + txt_append + u"\n" + url
log.debug(u"Text to post '{}'".format(txt))
try:
posted_urls.add(url)
config.twitter_api.update_status(txt)
except Exception as ex:
log.error(u"Error posting to Twitter: %s", ex)
# add entry to feed
feed.appendChild(entry.documentElement)
items = items + 1
if items >= max_items:
break
except Exception as ex:
log.error("Error parsing %s: %s", file, ex)
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type,
exc_value,
exc_traceback,
limit=2,
file=sys.stdout)
if config.post_to_twitter():
with open(posted_urls_file, 'wb') as f:
pickle.dump(posted_urls, f, protocol=pickle.HIGHEST_PROTOCOL)
if index:
index.close()
return doc
def splice():
""" Splice together a planet from a cache of entries """
import planet
log = planet.getLogger(config.log_level(),config.log_format())
log.info("Loading cached data")
cache = config.cache_directory()
dir=[(os.stat(file).st_mtime,file) for file in glob.glob(cache+"/*")
if not os.path.isdir(file)]
dir.sort()
dir.reverse()
max_items=max([config.items_per_page(templ)
for templ in config.template_files() or ['Planet']])
doc = minidom.parseString('<feed xmlns="http://www.w3.org/2005/Atom"/>')
feed = doc.documentElement
# insert feed information
createTextElement(feed, 'title', config.name())
date(feed, 'updated', time.gmtime())
gen = createTextElement(feed, 'generator', config.generator())
gen.setAttribute('uri', config.generator_uri())
author = doc.createElement('author')
createTextElement(author, 'name', config.owner_name())
createTextElement(author, 'email', config.owner_email())
feed.appendChild(author)
if config.feed():
createTextElement(feed, 'id', config.feed())
link = doc.createElement('link')
link.setAttribute('rel', 'self')
link.setAttribute('href', config.feed())
if config.feedtype():
link.setAttribute('type', "application/%s+xml" % config.feedtype())
feed.appendChild(link)
if config.link():
link = doc.createElement('link')
link.setAttribute('rel', 'alternate')
link.setAttribute('href', config.link())
feed.appendChild(link)
# insert subscription information
sub_ids = []
feed.setAttribute('xmlns:planet',planet.xmlns)
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data=feedparser.parse(filename(sources,sub))
if data.feed.has_key('id'): sub_ids.append(data.feed.id)
if not data.feed: continue
xdoc=minidom.parseString('''<planet:source xmlns:planet="%s"
xmlns="http://www.w3.org/2005/Atom"/>\n''' % planet.xmlns)
reconstitute.source(xdoc.documentElement, data.feed, None, None)
feed.appendChild(xdoc.documentElement)
index = idindex.open()
# insert entry information
items = 0
for mtime,file in dir:
if index != None:
base = os.path.basename(file)
if index.has_key(base) and index[base] not in sub_ids: continue
try:
entry=minidom.parse(file)
# verify that this entry is currently subscribed to
entry.normalize()
sources = entry.getElementsByTagName('source')
if sources:
ids = sources[0].getElementsByTagName('id')
if ids and ids[0].childNodes[0].nodeValue not in sub_ids:
ids = sources[0].getElementsByTagName('planet:id')
if not ids: continue
if ids[0].childNodes[0].nodeValue not in sub_ids: continue
# add entry to feed
feed.appendChild(entry.documentElement)
items = items + 1
if items >= max_items: break
except:
log.error("Error parsing %s", file)
if index: index.close()
return doc
