def test_equality(self):
tag = Tag('test', settings={})
cat = Category('test', settings={})
author = Author('test', settings={})
# same name, but different class
self.assertNotEqual(tag, cat)
self.assertNotEqual(tag, author)
# should be equal vs text representing the same name
self.assertEqual(tag, u'test')
# should not be equal vs binary
self.assertNotEqual(tag, b'test')
# Tags describing the same should be equal
tag_equal = Tag('Test', settings={})
self.assertEqual(tag, tag_equal)
# Author describing the same should be equal
author_equal = Author('Test', settings={})
self.assertEqual(author, author_equal)
cat_ascii = Category('指導書', settings={})
self.assertEqual(cat_ascii, u'zhi dao shu')
def test_tag_cloud_random(self):
tag_cloud.generate_tag_cloud(self.generator)
expected = [
(Tag('plugins', self._settings), 1),
(Tag('fun', self._settings), 4),
(Tag('python', self._settings), 4),
(Tag('pelican', self._settings), 1)
]
six.assertCountEqual(self, self.generator.tag_cloud, expected)
def test_tag_cloud_size_rev(self):
self.generator.settings['TAG_CLOUD_STEPS'] = 10
self.generator.settings['TAG_CLOUD_SORTING'] = 'size-rev'
tag_cloud.generate_tag_cloud(self.generator)
expected = [(Tag('python', self._settings), 10),
(Tag('fun', self._settings), 3),
(Tag('plugins', self._settings), 2),
(Tag('pelican', self._settings), 1)]
self.assertEqual(self.generator.tag_cloud, expected)
def test_tag_cloud_badge(self):
self.generator.settings['TAG_CLOUD_STEPS'] = 10
self.generator.settings['TAG_CLOUD_BADGE'] = True
tag_cloud.generate_tag_cloud(self.generator)
expected = [(Tag('pelican', self._settings), 1, 5),
(Tag('plugins', self._settings), 2, 4),
(Tag('fun', self._settings), 3, 3),
(Tag('python', self._settings), 10, 1)]
six.assertCountEqual(self, self.generator.tag_cloud, expected)
def test_tag_cloud_random(self):
self.generator.settings['TAG_CLOUD_STEPS'] = 10
self.generator.settings['TAG_CLOUD_BADGE'] = False
tag_cloud.generate_tag_cloud(self.generator)
expected = [(Tag('pelican', self._settings), 1),
(Tag('plugins', self._settings), 2),
(Tag('fun', self._settings), 3),
(Tag('python', self._settings), 10)]
six.assertCountEqual(self, self.generator.tag_cloud, expected)
def test_tag_cloud_size(self):
self.generator.settings['TAG_CLOUD_SORTING'] = 'size'
tag_cloud.generate_tag_cloud(self.generator)
expected = [
(Tag('pelican', self._settings), 1),
(Tag('plugins', self._settings), 1),
(Tag('fun', self._settings), 4),
(Tag('python', self._settings), 4)
]
self.assertEqual(self.generator.tag_cloud, expected)
def test_tag_cloud_size(self):
self.generator.settings["TAG_CLOUD_STEPS"] = 10
self.generator.settings["TAG_CLOUD_SORTING"] = "size"
tag_cloud.generate_tag_cloud(self.generator)
expected = [
(Tag("pelican", self._settings), 1),
(Tag("plugins", self._settings), 2),
(Tag("fun", self._settings), 3),
(Tag("python", self._settings), 10),
]
self.assertEqual(self.generator.tag_cloud, expected)
def test_tag_cloud_badge(self):
self.generator.settings["TAG_CLOUD_STEPS"] = 10
self.generator.settings["TAG_CLOUD_BADGE"] = True
tag_cloud.generate_tag_cloud(self.generator)
expected = [
(Tag("pelican", self._settings), 1, 5),
(Tag("plugins", self._settings), 2, 4),
(Tag("fun", self._settings), 3, 3),
(Tag("python", self._settings), 10, 1),
]
six.assertCountEqual(self, self.generator.tag_cloud, expected)
def replacer(m):
what = m.group('what')
value = m.group('value')
origin = m.group('path')
# we support only filename for now. the plan is to support
# categories, tags, etc. in the future, but let's keep things
# simple for now.
# XXX Put this in a different location.
if what == 'filename':
if value.startswith('/'):
value = value[1:]
else:
# relative to the source path of this content
value = self.get_relative_source_path(
os.path.join(self.relative_dir, value)
)
if value in self._context['filenames']:
origin = '/'.join((siteurl,
self._context['filenames'][value].url))
origin = origin.replace('\\', '/') # Fow windows paths.
else:
logger.warning("Unable to find {fn}, skipping url"
" replacement".format(fn=value))
elif what == 'category':
origin = Category(value, self.settings).url
elif what == 'tag':
origin = Tag(value, self.settings).url
return ''.join((m.group('markup'), m.group('quote'), origin,
m.group('quote')))
def replacer(m):
what = m.group('what')
value = urlparse(m.group('value'))
path = value.path
origin = m.group('path')
# XXX Put this in a different location.
if what == 'filename':
if path.startswith('/'):
path = path[1:]
else:
# relative to the source path of this content
path = self.get_relative_source_path(
os.path.join(self.relative_dir, path))
if path in self._context['filenames']:
origin = '/'.join(
(siteurl, self._context['filenames'][path].url))
origin = origin.replace('\\', '/') # for Windows paths.
else:
logger.warning("Unable to find {fn}, skipping url"
" replacement".format(fn=path))
elif what == 'category':
origin = Category(path, self.settings).url
elif what == 'tag':
origin = Tag(path, self.settings).url
# keep all other parts, such as query, fragment, etc.
parts = list(value)
parts[2] = origin
origin = urlunparse(parts)
return ''.join((m.group('markup'), m.group('quote'), origin,
m.group('quote')))
def __init__(self, tag_names=None, date=()):
if tag_names:
self.tags = [Tag(name=name, settings={}) for name in tag_names]
if date:
self.date = datetime.datetime(*date)
else:
self.date = datetime.datetime(1970, 1, 1)
def replacer(m):
what = m.group('what')
value = urlparse(m.group('value'))
path = value.path
origin = m.group('path')
# XXX Put this in a different location.
if what in {'filename', 'attach'}:
if path.startswith('/'):
path = path[1:]
else:
# relative to the source path of this content
path = self.get_relative_source_path(
os.path.join(self.relative_dir, path))
if path not in self._context['filenames']:
unquoted_path = path.replace('%20', ' ')
if unquoted_path in self._context['filenames']:
path = unquoted_path
linked_content = self._context['filenames'].get(path)
if linked_content:
if what == 'attach':
if isinstance(linked_content, Static):
linked_content.attach_to(self)
else:
logger.warning(
"%s used {attach} link syntax on a "
"non-static file. Use {filename} instead.",
self.get_relative_source_path())
origin = '/'.join((siteurl, linked_content.url))
origin = origin.replace('\\', '/') # for Windows paths.
else:
logger.warning(
"Unable to find `%s`, skipping url replacement.",
value.geturl(),
extra={
'limit_msg': ("Other resources were not found "
"and their urls not replaced")
})
elif what == 'category':
origin = '/'.join((siteurl, Category(path, self.settings).url))
elif what == 'tag':
origin = '/'.join((siteurl, Tag(path, self.settings).url))
else:
logger.warning(
"Replacement Indicator '%s' not recognized, "
"skipping replacement", what)
# keep all other parts, such as query, fragment, etc.
parts = list(value)
parts[2] = origin
origin = urlunparse(parts)
return ''.join((m.group('markup'), m.group('quote'), origin,
m.group('quote')))
def test_slugify_with_substitutions_and_dots(self):
tag = Tag('Tag Dot', settings={'TAG_REGEX_SUBSTITUTIONS': [
('Tag Dot', 'tag.dot'),
]})
cat = Category('Category Dot',
settings={'CATEGORY_REGEX_SUBSTITUTIONS': [
('Category Dot', 'cat.dot'),
]})
self.assertEqual(tag.slug, 'tag.dot')
self.assertEqual(cat.slug, 'cat.dot')
def replacer(m):
what = m.group('what')
value = urlparse(m.group('value'))
path = value.path
origin = m.group('path')
# XXX Put this in a different location.
if what == 'filename':
if path.startswith('/'):
path = path[1:]
else:
# relative to the source path of this content
path = self.get_relative_source_path(
os.path.join(self.relative_dir, path))
if path not in self._context['filenames']:
unquoted_path = path.replace('%20', ' ')
if unquoted_path in self._context['filenames']:
path = unquoted_path
if path in self._context['filenames']:
origin = '/'.join(
(siteurl, self._context['filenames'][path].url))
origin = origin.replace('\\', '/') # for Windows paths.
else:
logger.warning(
"Unable to find `%s`, skipping url replacement.",
value.geturl(),
extra={
'limit_msg': ("Other resources were not found "
"and their urls not replaced")
})
elif what == 'category':
origin = Category(path, self.settings).url
elif what == 'tag':
origin = Tag(path, self.settings).url
# keep all other parts, such as query, fragment, etc.
parts = list(value)
parts[2] = origin
origin = urlunparse(parts)
return ''.join((m.group('markup'), m.group('quote'), origin,
m.group('quote')))
def _link_replacer(self, siteurl, m):
what = m.group('what')
value = urlparse(m.group('value'))
path = value.path
origin = m.group('path')
# urllib.parse.urljoin() produces `a.html` for urljoin("..", "a.html")
# so if RELATIVE_URLS are enabled, we fall back to os.path.join() to
# properly get `../a.html`. However, os.path.join() produces
# `baz/http://foo/bar.html` for join("baz", "http://foo/bar.html")
# instead of correct "http://foo/bar.html", so one has to pick a side
# as there is no silver bullet.
if self.settings['RELATIVE_URLS']:
joiner = os.path.join
else:
joiner = urljoin
# However, it's not *that* simple: urljoin("blog", "index.html")
# produces just `index.html` instead of `blog/index.html` (unlike
# os.path.join()), so in order to get a correct answer one needs to
# append a trailing slash to siteurl in that case. This also makes
# the new behavior fully compatible with Pelican 3.7.1.
if not siteurl.endswith('/'):
siteurl += '/'
# XXX Put this in a different location.
if what in {'filename', 'static', 'attach'}:
def _get_linked_content(key, url):
nonlocal value
def _find_path(path):
if path.startswith('/'):
path = path[1:]
else:
# relative to the source path of this content
path = self.get_relative_source_path(
os.path.join(self.relative_dir, path))
return self._context[key].get(path, None)
# try path
result = _find_path(url.path)
if result is not None:
return result
# try unquoted path
result = _find_path(unquote(url.path))
if result is not None:
return result
# try html unescaped url
unescaped_url = urlparse(unescape(url.geturl()))
result = _find_path(unescaped_url.path)
if result is not None:
value = unescaped_url
return result
# check if a static file is linked with {filename}
if what == 'filename' and key == 'generated_content':
linked_content = _get_linked_content(
'static_content', value)
if linked_content:
logger.warning(
'{filename} used for linking to static'
' content %s in %s. Use {static} instead',
value.path, self.get_relative_source_path())
return linked_content
return None
if what == 'filename':
key = 'generated_content'
else:
key = 'static_content'
linked_content = _get_linked_content(key, value)
if linked_content:
if what == 'attach':
linked_content.attach_to(self)
origin = joiner(siteurl, linked_content.url)
origin = origin.replace('\\', '/') # for Windows paths.
else:
logger.warning(
"Unable to find '%s', skipping url replacement.",
value.geturl(),
extra={
'limit_msg': ("Other resources were not found "
"and their urls not replaced")
})
elif what == 'category':
origin = joiner(siteurl, Category(path, self.settings).url)
elif what == 'tag':
origin = joiner(siteurl, Tag(path, self.settings).url)
elif what == 'index':
origin = joiner(siteurl, self.settings['INDEX_SAVE_AS'])
elif what == 'author':
origin = joiner(siteurl, Author(path, self.settings).url)
else:
logger.warning(
"Replacement Indicator '%s' not recognized, "
"skipping replacement", what)
# keep all other parts, such as query, fragment, etc.
parts = list(value)
parts[2] = origin
origin = urlunparse(parts)
return ''.join(
(m.group('markup'), m.group('quote'), origin, m.group('quote')))
def parse_jpeg(self, *, source_path: str) -> Tuple[str, dict]:
JpegReader.logger.info(source_path)
img = Image.open(source_path)
image_data = Exiv2Parser.get_values(source_path)
title = image_data.get(Exiv.DESCRIPTION.value, 'Untitled')
author = image_data.get(Exiv.ARTIST.value, 'Unknown')
date_string = image_data.get(Exiv.DATETIME.value, '')
date = datetime.strptime(date_string, "%Y:%m:%d %H:%M:%S")
slug = URLWrapper(image_data.get(Exiv.HEADLINE.value, title),
self.settings).slug
description_long = image_data.get(Exiv.COMMENT.value, '')
summary = image_data.get(Exiv.CAPTION.value, description_long[:140])
tags = [
Tag(tag, self.settings)
for tag in image_data.get(Exiv.KEYWORDS.value, list())
]
content_root = self.settings[PelicanConfig.PATH.value]
path_output = self.settings[PelicanConfig.OUTPUT_PATH.value]
relative_source = dirname(source_path[len(content_root):]).lstrip(sep)
if self.settings[PelicanConfig.USE_FOLDER_AS_CATEGORY.value]:
category = relative_source.split(sep)[-1]
else:
category = image_data.get(Exiv.CATEGORY.value, None)
type_of_content = relative_source.split(sep)[
0] # either 'blog' or 'pages' as far as I know.
url_site = self.settings[PelicanConfig.SITE_URL.value]
if type_of_content.lower() == PelicanClass.PAGES.value:
url_document = self.settings[PelicanConfig.PAGE_URL.value]
document_save_as = self.settings[PelicanConfig.PAGE_SAVE_AS.value]
else: # Assume PelicanClass.BLOG
url_document = self.settings[PelicanConfig.ARTICLE_URL.value]
document_save_as = self.settings[
PelicanConfig.ARTICLE_SAVE_AS.value]
page_url_complete = join(url_site, url_document)
author_wrapper = Author(author, self.settings)
# Move image in place:
metadata = {
PelicanMetadata.TITLE.value: title,
PelicanMetadata.AUTHORS.value: [author_wrapper],
PelicanMetadata.DATE.value: date,
PelicanMetadata.SLUG.value: slug,
PelicanMetadata.TAGS.value: tags,
PelicanMetadata.CUSTOM_ALL.value: image_data
}
if category is not None:
metadata[PelicanMetadata.CATEGORY.value] = Category(
category, self.settings)
thumb_name = '{0}_thumb.jpg'.format(slug)
original_name = '{0}.jpg'.format(slug)
path_output_html = join(path_output,
document_save_as).format(**metadata)
path_output_dir = dirname(path_output_html)
path_output_original = join(path_output_dir, original_name)
path_output_thumb = join(path_output_dir, thumb_name)
# Here we generate the summary info incase this is used for articles we get nice thumbnails and summary
metadata[PelicanMetadata.SUMMARY.value] = summary
metadata[PelicanMetadata.FEATURED_IMAGE.value] = join(
url_site, path_output_thumb[len(path_output):])
if Exiv.OBJECT_NAME.value in image_data:
metadata[PelicanMetadata.TEMPLATE.value] = image_data[
Exiv.OBJECT_NAME.value]
# Write the size/HTML out before we reduce the image to a thumb
content = "<img src='{src}' alt='{alt}' style='width: {width}px; height: auto; max-width: 100%;'></img><p>{body}</p>" \
.format(src=original_name, alt=title, width=img.width, height=img.height, body=description_long)
# Ensure the directory levels exist
if not isdir(path_output_dir):
makedirs(path_output_dir)
img.save(path_output_original)
img.thumbnail(self.thumb_size)
img.save(path_output_thumb)
# Debug info if we need it
JpegReader.logger.debug(content)
JpegReader.logger.debug(str(metadata))
JpegReader.logger.debug(path_output_html)
return content, metadata
def _link_replacer(self, siteurl, m):
what = m.group('what')
value = urlparse(m.group('value'))
path = value.path
origin = m.group('path')
# urllib.parse.urljoin() produces `a.html` for urljoin("..", "a.html")
# so if RELATIVE_URLS are enabled, we fall back to os.path.join() to
# properly get `../a.html`. However, os.path.join() produces
# `baz/http://foo/bar.html` for join("baz", "http://foo/bar.html")
# instead of correct "http://foo/bar.html", so one has to pick a side
# as there is no silver bullet.
if self.settings['RELATIVE_URLS']:
joiner = os.path.join
else:
joiner = urljoin
# However, it's not *that* simple: urljoin("blog", "index.html")
# produces just `index.html` instead of `blog/index.html` (unlike
# os.path.join()), so in order to get a correct answer one needs to
# append a trailing slash to siteurl in that case. This also makes
# the new behavior fully compatible with Pelican 3.7.1.
if not siteurl.endswith('/'):
siteurl += '/'
# XXX Put this in a different location.
if what in {'filename', 'attach'}:
if path.startswith('/'):
path = path[1:]
else:
# relative to the source path of this content
path = self.get_relative_source_path(
os.path.join(self.relative_dir, path))
if path not in self._context['filenames']:
unquoted_path = path.replace('%20', ' ')
if unquoted_path in self._context['filenames']:
path = unquoted_path
linked_content = self._context['filenames'].get(path)
if linked_content:
if what == 'attach':
if isinstance(linked_content, Static):
linked_content.attach_to(self)
else:
logger.warning(
"%s used {attach} link syntax on a "
"non-static file. Use {filename} instead.",
self.get_relative_source_path())
origin = joiner(siteurl, linked_content.url)
origin = origin.replace('\\', '/') # for Windows paths.
else:
logger.warning(
"Unable to find '%s', skipping url replacement.",
value.geturl(),
extra={
'limit_msg': ("Other resources were not found "
"and their urls not replaced")
})
elif what == 'category':
origin = joiner(siteurl, Category(path, self.settings).url)
elif what == 'tag':
origin = joiner(siteurl, Tag(path, self.settings).url)
elif what == 'index':
origin = joiner(siteurl, self.settings['INDEX_SAVE_AS'])
elif what == 'author':
origin = joiner(siteurl, Author(path, self.settings).url)
else:
logger.warning(
"Replacement Indicator '%s' not recognized, "
"skipping replacement", what)
# keep all other parts, such as query, fragment, etc.
parts = list(value)
parts[2] = origin
origin = urlunparse(parts)
return ''.join(
(m.group('markup'), m.group('quote'), origin, m.group('quote')))
