def needles(self):
"""Fill out path (and path.trigrams)."""
yield "path", self.path
extension = splitext(self.path)[1]
if extension:
yield "ext", extension[1:] # skip the period
if is_image(self.path):
bytestring = self.contents.encode("utf-8") if self.contains_text() else self.contents
yield "raw_data", b64encode(bytestring)
# binary, but not an image
elif not self.contains_text():
yield "is_binary", True
def needles(self):
"""Fill out path (and path.trigrams)."""
if self.is_link():
# realpath will keep following symlinks until it gets to the 'real' thing.
yield 'link', relpath(realpath(self.absolute_path()), self.tree.source_folder)
yield 'path', self.path
extension = splitext(self.path)[1]
if extension:
yield 'ext', extension[1:] # skip the period
if is_image(self.path):
bytestring = (self.contents.encode('utf-8') if self.contains_text()
else self.contents)
yield 'raw_data', b64encode(bytestring)
# binary, but not an image
elif not self.contains_text():
yield 'is_binary', True
def needles(self):
"""Fill out path (and path.trigrams)."""
if self.is_link():
# realpath will keep following symlinks until it gets to the 'real' thing.
yield 'link', relpath(realpath(self.absolute_path()), self.tree.source_folder)
yield 'path', self.path
extension = splitext(self.path)[1]
if extension:
yield 'ext', extension[1:] # skip the period
if is_image(self.path):
bytestring = (self.contents.encode('utf-8') if self.contains_text()
else self.contents)
yield 'raw_data', b64encode(bytestring)
# binary, but not an image
elif not self.contains_text():
yield 'is_binary', True
def _browse_file(tree, path, line_docs, file_doc, config, date=None, contents=None):
"""Return a rendered page displaying a source file.
:arg string tree: name of tree on which file is found
:arg string path: relative path from tree root of file
:arg list line_docs: LINE documents as defined in the mapping of core.py,
where the `content` field is dereferenced
:arg file_doc: the FILE document as defined in core.py
:arg config: TreeConfig object of this tree
:arg date: a formatted string representing the generated date, default to now
:arg string contents: the contents of the source file, defaults to joining
the `content` field of all line_docs
"""
def sidebar_links(sections):
"""Return data structure to build nav sidebar from. ::
[('Section Name', [{'icon': ..., 'title': ..., 'href': ...}])]
"""
# Sort by order, resolving ties by section name:
return sorted(sections, key=lambda section: (section['order'],
section['heading']))
if not date:
# Then assume that the file is generated now. Remark: we can't use this
# as the default param because that is only evaluated once, so the same
# time would always be used.
date = datetime.utcnow().strftime("%a, %d %b %Y %H:%M:%S +0000")
common = _build_common_file_template(tree, path, date, config)
links = file_doc.get('links', [])
if is_image(path):
return render_template(
'image_file.html',
**common)
else: # We don't allow browsing binary files, so this must be a text file.
# We concretize the lines into a list because we iterate over it multiple times
lines = [doc['content'] for doc in line_docs]
if not contents:
# If contents are not provided, we can reconstruct them by
# stitching the lines together.
contents = ''.join(lines)
offsets = cumulative_sum(imap(len, lines))
tree_config = config.trees[tree]
# Construct skimmer objects for all enabled plugins that define a
# file_to_skim class.
skimmers = [plugin.file_to_skim(path,
contents,
plugin.name,
tree_config,
file_doc,
line_docs)
for plugin in tree_config.enabled_plugins
if plugin.file_to_skim]
skim_links, refses, regionses, annotationses = skim_file(skimmers, len(line_docs))
index_refs = (Ref.es_to_triple(ref, tree_config) for ref in
chain.from_iterable(doc.get('refs', [])
for doc in line_docs))
index_regions = (Region.es_to_triple(region) for region in
chain.from_iterable(doc.get('regions', [])
for doc in line_docs))
tags = finished_tags(lines,
chain(chain.from_iterable(refses), index_refs),
chain(chain.from_iterable(regionses), index_regions))
return render_template(
'text_file.html',
**merge(common, {
# Someday, it would be great to stream this and not concretize
# the whole thing in RAM. The template will have to quit
# looping through the whole thing 3 times.
'lines': [(html_line(doc['content'], tags_in_line, offset),
doc.get('annotations', []) + skim_annotations)
for doc, tags_in_line, offset, skim_annotations
in izip(line_docs, tags_per_line(tags), offsets, annotationses)],
'is_text': True,
'sections': sidebar_links(links + skim_links)}))
