def test_readfile_path_metadata_implicit_dates(self):
test_file = 'article_with_metadata_implicit_dates.html'
page = self.read_file(path=test_file, DEFAULT_DATE='fs')
expected = {
'date':
SafeDatetime.fromtimestamp(os.stat(_path(test_file)).st_mtime),
'modified':
SafeDatetime.fromtimestamp(os.stat(_path(test_file)).st_mtime)
}
self.assertDictHasSubset(page.metadata, expected)
def path_metadata(full_path, source_path, settings=None):
metadata = {}
if settings:
if settings.get("DEFAULT_DATE", None) == "fs":
metadata["date"] = SafeDatetime.fromtimestamp(os.stat(full_path).st_ctime)
metadata.update(settings.get("EXTRA_PATH_METADATA", {}).get(source_path, {}))
return metadata
def path_metadata(full_path, source_path, settings=None):
metadata = {}
if settings:
if settings.get('DEFAULT_DATE', None) == 'fs':
metadata['date'] = SafeDatetime.fromtimestamp(
os.stat(full_path).st_mtime)
metadata.update(
settings.get('EXTRA_PATH_METADATA', {}).get(source_path, {}))
return metadata
def path_metadata(full_path, source_path, settings=None):
metadata = {}
if settings:
if settings.get('DEFAULT_DATE', None) == 'fs':
metadata['date'] = SafeDatetime.fromtimestamp(
os.stat(full_path).st_mtime)
metadata.update(settings.get('EXTRA_PATH_METADATA', {}).get(
source_path, {}))
return metadata
def path_metadata(full_path, source_path, settings=None):
metadata = {}
if settings:
if settings.get('DEFAULT_DATE', None) == 'fs':
metadata['date'] = SafeDatetime.fromtimestamp(
os.stat(full_path).st_mtime)
# Apply EXTRA_PATH_METADATA for the source path and the paths of any
# parent directories. Sorting EPM first ensures that the most specific
# path wins conflicts.
epm = settings.get('EXTRA_PATH_METADATA', {})
for path, meta in sorted(epm.items()):
# Enforce a trailing slash when checking for parent directories.
# This prevents false positives when one file or directory's name
# is a prefix of another's.
dirpath = os.path.join(path, '')
if source_path == path or source_path.startswith(dirpath):
metadata.update(meta)
return metadata
def tumblr2fields(api_key, blogname):
""" Imports Tumblr posts (API v2)"""
try:
# py3k import
import json
except ImportError:
# py2 import
import simplejson as json
try:
# py3k import
import urllib.request as urllib_request
except ImportError:
# py2 import
import urllib2 as urllib_request
def get_tumblr_posts(api_key, blogname, offset=0):
url = ("http://api.tumblr.com/v2/blog/%s.tumblr.com/"
"posts?api_key=%s&offset=%d&filter=raw") % (blogname, api_key,
offset)
request = urllib_request.Request(url)
handle = urllib_request.urlopen(request)
posts = json.loads(handle.read().decode('utf-8'))
return posts.get('response').get('posts')
offset = 0
posts = get_tumblr_posts(api_key, blogname, offset)
settings = read_settings()
subs = settings['SLUG_REGEX_SUBSTITUTIONS']
while len(posts) > 0:
for post in posts:
title = \
post.get('title') or \
post.get('source_title') or \
post.get('type').capitalize()
slug = post.get('slug') or slugify(title, regex_subs=subs)
tags = post.get('tags')
timestamp = post.get('timestamp')
date = SafeDatetime.fromtimestamp(
int(timestamp)).strftime("%Y-%m-%d %H:%M:%S")
slug = SafeDatetime.fromtimestamp(
int(timestamp)).strftime("%Y-%m-%d-") + slug
format = post.get('format')
content = post.get('body')
type = post.get('type')
if type == 'photo':
if format == 'markdown':
fmtstr = ''
else:
fmtstr = '<img alt="%s" src="%s" />'
content = ''
for photo in post.get('photos'):
content += '\n'.join(
fmtstr % (photo.get('caption'),
photo.get('original_size').get('url')))
content += '\n\n' + post.get('caption')
elif type == 'quote':
if format == 'markdown':
fmtstr = '\n\n— %s'
else:
fmtstr = '<p>— %s</p>'
content = post.get('text') + fmtstr % post.get('source')
elif type == 'link':
if format == 'markdown':
fmtstr = '[via](%s)\n\n'
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
content = fmtstr % post.get('url') + post.get('description')
elif type == 'audio':
if format == 'markdown':
fmtstr = '[via](%s)\n\n'
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
content = fmtstr % post.get('source_url') + \
post.get('caption') + \
post.get('player')
elif type == 'video':
if format == 'markdown':
fmtstr = '[via](%s)\n\n'
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
source = fmtstr % post.get('source_url')
caption = post.get('caption')
players = '\n'.join(
player.get('embed_code') for player in post.get('player'))
content = source + caption + players
elif type == 'answer':
title = post.get('question')
content = ('<p>'
'<a href="%s" rel="external nofollow">%s</a>'
': %s'
'</p>\n'
' %s' %
(post.get('asking_name'), post.get('asking_url'),
post.get('question'), post.get('answer')))
content = content.rstrip() + '\n'
kind = 'article'
status = 'published' # TODO: Find a way for draft posts
yield (title, content, slug, date, post.get('blog_name'), [type],
tags, status, kind, format)
offset += len(posts)
posts = get_tumblr_posts(api_key, blogname, offset)
def parse(self):
""" Imports Tumblr posts (API v2)"""
offset = 0
posts = self._get_tumblr_posts(offset)
settings = read_settings()
subs = settings["SLUG_REGEX_SUBSTITUTIONS"]
while len(posts) > 0:
for post in posts:
title = (
post.get("title")
or post.get("source_title")
or post.get("type").capitalize()
)
slug = post.get("slug") or slugify(title, regex_subs=subs)
tags = post.get("tags")
timestamp = post.get("timestamp")
date = SafeDatetime.fromtimestamp(int(timestamp)).strftime(
"%Y-%m-%d %H:%M:%S"
)
slug = (
SafeDatetime.fromtimestamp(int(timestamp)).strftime(
"%Y-%m-%d-"
)
+ slug
)
format = post.get("format")
content = post.get("body")
type = post.get("type")
if type == "photo":
if format == "markdown":
fmtstr = ""
else:
fmtstr = '<img alt="%s" src="%s" />'
content = ""
for photo in post.get("photos"):
content += "\n".join(
fmtstr
% (
photo.get("caption"),
photo.get("original_size").get("url"),
)
)
content += "\n\n" + post.get("caption")
elif type == "quote":
if format == "markdown":
fmtstr = "\n\n— %s"
else:
fmtstr = "<p>— %s</p>"
content = post.get("text") + fmtstr % post.get("source")
elif type == "link":
if format == "markdown":
fmtstr = "[via](%s)\n\n"
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
content = fmtstr % post.get("url") + post.get(
"description"
)
elif type == "audio":
if format == "markdown":
fmtstr = "[via](%s)\n\n"
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
content = (
fmtstr % post.get("source_url")
+ post.get("caption")
+ post.get("player")
)
elif type == "video":
if format == "markdown":
fmtstr = "[via](%s)\n\n"
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
source = fmtstr % post.get("source_url")
caption = post.get("caption")
players = "\n".join(
player.get("embed_code")
for player in post.get("player")
)
content = source + caption + players
elif type == "answer":
title = post.get("question")
content = (
"<p>"
'<a href="%s" rel="external nofollow">%s</a>'
": %s"
"</p>\n"
" %s"
% (
post.get("asking_name"),
post.get("asking_url"),
post.get("question"),
post.get("answer"),
)
)
content = content.rstrip() + "\n"
kind = "article"
status = "published" # TODO: Find a way for draft posts
yield (
title,
content,
slug,
date,
post.get("blog_name"),
[type],
tags,
status,
kind,
format,
)
offset += len(posts)
posts = self._get_tumblr_posts(offset)
def tumblr2fields(api_key, blogname):
""" Imports Tumblr posts (API v2)"""
try:
# py3k import
import json
except ImportError:
# py2 import
import simplejson as json
try:
# py3k import
import urllib.request as urllib_request
except ImportError:
# py2 import
import urllib2 as urllib_request
def get_tumblr_posts(api_key, blogname, offset=0):
url = ("http://api.tumblr.com/v2/blog/%s.tumblr.com/"
"posts?api_key=%s&offset=%d&filter=raw") % (
blogname, api_key, offset)
request = urllib_request.Request(url)
handle = urllib_request.urlopen(request)
posts = json.loads(handle.read().decode('utf-8'))
return posts.get('response').get('posts')
offset = 0
posts = get_tumblr_posts(api_key, blogname, offset)
settings = read_settings()
subs = settings['SLUG_REGEX_SUBSTITUTIONS']
while len(posts) > 0:
for post in posts:
title = \
post.get('title') or \
post.get('source_title') or \
post.get('type').capitalize()
slug = post.get('slug') or slugify(title, regex_subs=subs)
tags = post.get('tags')
timestamp = post.get('timestamp')
date = SafeDatetime.fromtimestamp(int(timestamp)).strftime(
"%Y-%m-%d %H:%M:%S")
slug = SafeDatetime.fromtimestamp(int(timestamp)).strftime(
"%Y-%m-%d-") + slug
format = post.get('format')
content = post.get('body')
type = post.get('type')
if type == 'photo':
if format == 'markdown':
fmtstr = ''
else:
fmtstr = '<img alt="%s" src="%s" />'
content = ''
for photo in post.get('photos'):
content += '\n'.join(
fmtstr % (photo.get('caption'),
photo.get('original_size').get('url')))
content += '\n\n' + post.get('caption')
elif type == 'quote':
if format == 'markdown':
fmtstr = '\n\n— %s'
else:
fmtstr = '<p>— %s</p>'
content = post.get('text') + fmtstr % post.get('source')
elif type == 'link':
if format == 'markdown':
fmtstr = '[via](%s)\n\n'
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
content = fmtstr % post.get('url') + post.get('description')
elif type == 'audio':
if format == 'markdown':
fmtstr = '[via](%s)\n\n'
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
content = fmtstr % post.get('source_url') + \
post.get('caption') + \
post.get('player')
elif type == 'video':
if format == 'markdown':
fmtstr = '[via](%s)\n\n'
else:
fmtstr = '<p><a href="%s">via</a></p>\n'
source = fmtstr % post.get('source_url')
caption = post.get('caption')
players = '\n'.join(player.get('embed_code')
for player in post.get('player'))
content = source + caption + players
elif type == 'answer':
title = post.get('question')
content = ('<p>'
'<a href="%s" rel="external nofollow">%s</a>'
': %s'
'</p>\n'
' %s' % (post.get('asking_name'),
post.get('asking_url'),
post.get('question'),
post.get('answer')))
content = content.rstrip() + '\n'
kind = 'article'
status = 'published' # TODO: Find a way for draft posts
yield (title, content, slug, date, post.get('blog_name'), [type],
tags, status, kind, format)
offset += len(posts)
posts = get_tumblr_posts(api_key, blogname, offset)
def to_date(s):
d = SafeDatetime.fromtimestamp(s)
#d = d.replace(hour=0, minute=0, second=0, microsecond=0)
return d
