"""Several HTML builders."""
from __future__ import annotations
import contextlib
import html
import os
import os.path
import posixpath
import re
import shutil
import sys
import warnings
from pathlib import Path
from types import NoneType
from typing import TYPE_CHECKING
from urllib.parse import quote
import docutils.readers.doctree
from docutils import nodes
from docutils.core import Publisher
from docutils.frontend import OptionParser
from docutils.io import DocTreeInput, StringOutput
from sphinx import __display_version__, package_dir
from sphinx import version_info as sphinx_version
from sphinx._cli.util.colour import bold
from sphinx.builders import Builder
from sphinx.builders.html._assets import (
_CascadingStyleSheet,
_file_checksum,
_JavaScript,
)
from sphinx.builders.html._build_info import BuildInfo
from sphinx.config import ENUM
from sphinx.deprecation import _deprecation_warning
from sphinx.domains import Index, IndexEntry
from sphinx.environment.adapters.asset import ImageAdapter
from sphinx.environment.adapters.indexentries import IndexEntries
from sphinx.environment.adapters.toctree import document_toc, global_toctree_for_doc
from sphinx.errors import ConfigError, ThemeError
from sphinx.highlighting import PygmentsBridge
from sphinx.locale import _, __
from sphinx.search import js_index
from sphinx.theming import HTMLThemeFactory
from sphinx.util import logging
from sphinx.util._pathlib import _StrPath
from sphinx.util._timestamps import _format_rfc3339_microseconds
from sphinx.util._uri import is_url
from sphinx.util.display import progress_message, status_iterator
from sphinx.util.docutils import new_document
from sphinx.util.fileutil import copy_asset
from sphinx.util.i18n import format_date
from sphinx.util.inventory import InventoryFile
from sphinx.util.matching import DOTFILES, Matcher, patmatch
from sphinx.util.osutil import (
SEP,
_last_modified_time,
_relative_path,
copyfile,
ensuredir,
relative_uri,
)
from sphinx.writers.html import HTMLWriter
from sphinx.writers.html5 import HTML5Translator
if TYPE_CHECKING:
from collections.abc import Iterator, Set
from typing import Any, TypeAlias
from docutils.nodes import Node
from docutils.readers import Reader
from sphinx.application import Sphinx
from sphinx.config import Config
from sphinx.environment import BuildEnvironment
from sphinx.util.typing import ExtensionMetadata
#: the filename for the inventory of objects
INVENTORY_FILENAME = 'objects.inv'
logger = logging.getLogger(__name__)
return_codes_re = re.compile('[\r\n]+')
DOMAIN_INDEX_TYPE: TypeAlias = tuple[
# Index name (e.g. py-modindex)
str,
# Index class
type[Index],
# list of (heading string, list of index entries) pairs.
list[tuple[str, list[IndexEntry]]],
# whether sub-entries should start collapsed
bool,
]
def convert_locale_to_language_tag(locale: str | None) -> str | None:
"""Convert a locale string to a language tag (ex. en_US -> en-US).
refs: BCP 47 (:rfc:`5646`)
"""
if locale:
return locale.replace('_', '-')
else:
return None
class StandaloneHTMLBuilder(Builder):
"""Builds standalone HTML docs."""
name = 'html'
format = 'html'
epilog = __('The HTML pages are in %(outdir)s.')
default_translator_class = HTML5Translator
copysource = True
allow_parallel = True
out_suffix = '.html'
link_suffix = '.html' # defaults to matching out_suffix
indexer_format: Any = js_index
indexer_dumps_unicode = True
# create links to original images from images [True/False]
html_scaled_image_link = True
supported_image_types = ['image/svg+xml', 'image/png', 'image/gif', 'image/jpeg']
supported_remote_images = True
supported_data_uri_images = True
searchindex_filename = 'searchindex.js'
add_permalinks = True
allow_sharp_as_current_path = True
embedded = False # for things like HTML help or Qt help: suppresses sidebar
search = True # for things like HTML help and Apple help: suppress search
use_index = False
download_support = True # enable download role
imgpath: str = ''
domain_indices: list[DOMAIN_INDEX_TYPE] = []
def __init__(self, app: Sphinx, env: BuildEnvironment) -> None:
super().__init__(app, env)
# Static and asset directories
self._static_dir = Path(self.outdir / '_static')
self._sources_dir = Path(self.outdir / '_sources')
self._downloads_dir = Path(self.outdir / '_downloads')
self._images_dir = Path(self.outdir / '_images')
# CSS files
self._css_files: list[_CascadingStyleSheet] = []
# JS files
self._js_files: list[_JavaScript] = []
# Cached Publisher for writing doctrees to HTML
reader: Reader[DocTreeInput] = docutils.readers.doctree.Reader(
parser_name='restructuredtext'
)
pub = Publisher(
reader=reader,
parser=reader.parser,
writer=HTMLWriter(self),
source_class=DocTreeInput,
destination=StringOutput(encoding='unicode'),
)
pub.get_settings(output_encoding='unicode', traceback=True)
self._publisher = pub
def init(self) -> None:
self.build_info = self.create_build_info()
# basename of images directory
self.imagedir = '_images'
# section numbers for headings in the currently visited document
self.secnumbers: dict[str, tuple[int, ...]] = {}
# currently written docname
self.current_docname: str = ''
self.init_templates()
self.init_highlighter()
self.init_css_files()
self.init_js_files()
html_file_suffix = self.get_builder_config('file_suffix', 'html')
if html_file_suffix is not None:
self.out_suffix = html_file_suffix
html_link_suffix = self.get_builder_config('link_suffix', 'html')
if html_link_suffix is not None:
self.link_suffix = html_link_suffix
else:
self.link_suffix = self.out_suffix
self.use_index = self.get_builder_config('use_index', 'html')
def create_build_info(self) -> BuildInfo:
return BuildInfo(self.config, self.tags, frozenset({'html'}))
def _get_translations_js(self) -> Path | None:
for dir_ in self.config.locale_dirs:
js_file = Path(dir_, self.config.language, 'LC_MESSAGES', 'sphinx.js')
if js_file.is_file():
return js_file
js_file = package_dir.joinpath(
'locale', self.config.language, 'LC_MESSAGES', 'sphinx.js'
)
if js_file.is_file():
return js_file
js_file = Path(
sys.prefix, 'share', 'sphinx', 'locale', self.config.language, 'sphinx.js'
)
if js_file.is_file():
return js_file
return None
def _get_style_filenames(self) -> Iterator[str]:
if isinstance(self.config.html_style, str):
yield self.config.html_style
elif self.config.html_style is not None:
yield from self.config.html_style
elif self.theme:
yield from self.theme.stylesheets
else:
yield 'default.css'
def get_theme_config(self) -> tuple[str, dict[str, str | int | bool]]:
return self.config.html_theme, self.config.html_theme_options
def init_templates(self) -> None:
theme_factory = HTMLThemeFactory(self.app)
theme_name, theme_options = self.get_theme_config()
self.theme = theme_factory.create(theme_name)
self.theme_options = theme_options
self.create_template_bridge()
self.templates.init(self, self.theme)
def init_highlighter(self) -> None:
# determine Pygments style and create the highlighter
if self.config.pygments_style is not None:
style = self.config.pygments_style
elif self.theme:
# From the ``pygments_style`` theme setting
style = self.theme.pygments_style_default or 'none'
else:
style = 'sphinx'
self.highlighter = PygmentsBridge('html', style)
if self.theme:
# From the ``pygments_dark_style`` theme setting
dark_style = self.theme.pygments_style_dark
else:
dark_style = None
self.dark_highlighter: PygmentsBridge | None
if dark_style is not None:
self.dark_highlighter = PygmentsBridge('html', dark_style)
self.app.add_css_file(
'pygments_dark.css',
media='(prefers-color-scheme: dark)',
id='pygments_dark_css',
)
else:
self.dark_highlighter = None
@property
def css_files(self) -> list[_CascadingStyleSheet]:
_deprecation_warning(
__name__, f'{self.__class__.__name__}.css_files', remove=(9, 0)
)
return self._css_files
def init_css_files(self) -> None:
self._css_files = []
self.add_css_file('pygments.css', priority=200)
for filename in self._get_style_filenames():
self.add_css_file(filename, priority=200)
for filename, attrs in self.env._registry.css_files:
self.add_css_file(filename, **attrs)
for filename, attrs in self.get_builder_config('css_files', 'html'):
# User's CSSs are loaded after extensions'
attrs.setdefault('priority', 800)
self.add_css_file(filename, **attrs)
def add_css_file(self, filename: str, **kwargs: Any) -> None:
if '://' not in filename:
filename = posixpath.join('_static', filename)
if (asset := _CascadingStyleSheet(filename, **kwargs)) not in self._css_files:
self._css_files.append(asset)
@property
def script_files(self) -> list[_JavaScript]:
canonical_name = f'{self.__class__.__name__}.script_files'
_deprecation_warning(__name__, canonical_name, remove=(9, 0))
return self._js_files
def init_js_files(self) -> None:
self._js_files = []
self.add_js_file('documentation_options.js', priority=200)
self.add_js_file('doctools.js', priority=200)
self.add_js_file('sphinx_highlight.js', priority=200)
for filename, attrs in self.env._registry.js_files:
self.add_js_file(filename or '', **attrs)
for filename, attrs in self.get_builder_config('js_files', 'html'):
attrs.setdefault('priority', 800) # User's JSs are loaded after extensions'
self.add_js_file(filename or '', **attrs)
if self._get_translations_js():
self.add_js_file('translations.js')
def add_js_file(self, filename: str, **kwargs: Any) -> None:
if filename and '://' not in filename:
filename = posixpath.join('_static', filename)
if (asset := _JavaScript(filename, **kwargs)) not in self._js_files:
self._js_files.append(asset)
@property
def math_renderer_name(self) -> str | None:
name = self.get_builder_config('math_renderer', 'html')
if name is not None:
# use given name
return name
else:
# not given: choose a math_renderer from registered ones as possible
renderers = list(self.env._registry.html_inline_math_renderers)
if len(renderers) == 1:
# only default math_renderer (mathjax) is registered
return renderers[0]
elif len(renderers) == 2:
# default and another math_renderer are registered; prior the another
renderers.remove('mathjax')
return renderers[0]
else:
# many math_renderers are registered. can't choose automatically!
return None
def get_outdated_docs(self) -> Iterator[str]:
build_info_path = self.outdir / '.buildinfo'
try:
build_info = BuildInfo.load(build_info_path)
except ValueError as exc:
logger.warning(__('Failed to read build info file: %r'), exc)
except OSError:
# ignore errors on reading
pass
else:
if self.build_info != build_info:
# log the mismatch and backup the old build info
build_info_backup = build_info_path.with_name('.buildinfo.bak')
try:
shutil.move(build_info_path, build_info_backup)
self.build_info.dump(build_info_path)
except OSError:
pass # ignore errors
else:
# only log on success
msg = __(
'build_info mismatch, copying .buildinfo to .buildinfo.bak'
)
logger.info(bold(__('building [html]: ')) + msg) # NoQA: G003
yield from self.env.found_docs
return
if self.templates:
template_mtime = int(self.templates.newest_template_mtime() * 10**6)
try:
old_mtime = _last_modified_time(build_info_path)
except Exception:
pass
else:
# Let users know they have a newer template
if template_mtime > old_mtime:
logger.info(
bold('building [html]: ') # NoQA: G003
+ __(
'template %s has been changed since the previous build, '
'all docs will be rebuilt'
),
self.templates.newest_template_name(),
)
else:
template_mtime = 0
for docname in self.env.found_docs:
if docname not in self.env.all_docs:
logger.debug('[build target] did not in env: %r', docname)
yield docname
continue
target_name = self.get_output_path(docname)
try:
target_mtime = _last_modified_time(target_name)
except OSError:
target_mtime = 0
try:
doc_mtime = _last_modified_time(self.env.doc2path(docname))
srcmtime = max(doc_mtime, template_mtime)
if srcmtime > target_mtime:
logger.debug(
'[build target] target_name %r(%s), template(%s), docname %r(%s)',
target_name,
_format_rfc3339_microseconds(target_mtime),
_format_rfc3339_microseconds(template_mtime),
docname,
_format_rfc3339_microseconds(doc_mtime),
)
yield docname
except OSError:
# source doesn't exist anymore
pass
def get_asset_paths(self) -> list[str]:
return self.config.html_extra_path + self.config.html_static_path
def render_partial(self, node: Node | None) -> dict[str, str]:
"""Utility: Render a lone doctree node."""
if node is None:
return {'fragment': ''}
doc = new_document('')
doc.append(node)
self._publisher.set_source(doc)
self._publisher.publish()
return self._publisher.writer.parts
def prepare_writing(self, docnames: Set[str]) -> None:
# create the search indexer
self.indexer = None
if self.search:
from sphinx.search import IndexBuilder
lang = self.config.html_search_language or self.config.language
self.indexer = IndexBuilder(
self.env,
lang,
self.config.html_search_options,
self.config.html_search_scorer,
)
self.load_indexer(docnames)
self.docwriter = HTMLWriter(self)
with warnings.catch_warnings():
warnings.filterwarnings('ignore', category=DeprecationWarning)
# DeprecationWarning: The frontend.OptionParser class will be replaced
# by a subclass of argparse.ArgumentParser in Docutils 0.21 or later.
self.docsettings: Any = OptionParser(
defaults=self.env.settings,
components=(self.docwriter,),
read_config_files=True,
).get_default_values()
self.docsettings.compact_lists = bool(self.config.html_compact_lists)
# determine the additional indices to include
self.domain_indices = []
# html_domain_indices can be False/True or a list of index names
if indices_config := self.config.html_domain_indices:
if not isinstance(indices_config, bool):
check_names = True
indices_config = frozenset(indices_config)
else:
check_names = False
for domain in self.env.domains.sorted():
for index_cls in domain.indices:
index_name = f'{domain.name}-{index_cls.name}'
if check_names and index_name not in indices_config:
continue
content, collapse = index_cls(domain).generate()
if content:
self.domain_indices.append((
index_name,
index_cls,
content,
collapse,
))
# format the "last updated on" string, only once is enough since it
# typically doesn't include the time of day
last_updated: str | None
if (lu_fmt := self.config.html_last_updated_fmt) is not None:
last_updated = format_date(
lu_fmt or _('%b %d, %Y'),
language=self.config.language,
local_time=not self.config.html_last_updated_use_utc,
)
else:
last_updated = None
# If the logo or favicon are urls, keep them as-is, otherwise
# strip the relative path as the files will be copied into _static.
logo = self.config.html_logo or ''
favicon = self.config.html_favicon or ''
if not is_url(logo):
logo = os.path.basename(logo)
if not is_url(favicon):
favicon = os.path.basename(favicon)
self.relations = self.env.collect_relations()
rellinks: list[tuple[str, str, str, str]] = []
if self.use_index:
rellinks.append(('genindex', _('General Index'), 'I', _('index')))
for index_name, index_cls, _content, _collapse in self.domain_indices:
# if it has a short name
if index_cls.shortname:
rellinks.append((
index_name,
index_cls.localname,
'',
index_cls.shortname,
))
# add assets registered after ``Builder.init()``.
for css_filename, attrs in self.env._registry.css_files:
self.add_css_file(css_filename, **attrs)
for js_filename, attrs in self.env._registry.js_files:
self.add_js_file(js_filename or '', **attrs)
# back up _css_files and _js_files to allow adding CSS/JS files to a specific page.
self._orig_css_files = list(dict.fromkeys(self._css_files))
self._orig_js_files = list(dict.fromkeys(self._js_files))
styles = list(self._get_style_filenames())
self.globalcontext = {
'embedded': self.embedded,
'project': self.config.project,
'release': return_codes_re.sub('', self.config.release),
'version': self.config.version,
'last_updated': last_updated,
'copyright': self.config.copyright,
'master_doc': self.config.root_doc,
'root_doc': self.config.root_doc,
'use_opensearch': self.config.html_use_opensearch,
'docstitle': self.config.html_title,
'shorttitle': self.config.html_short_title,
'show_copyright': self.config.html_show_copyright,
'show_search_summary': self.config.html_show_search_summary,
'show_sphinx': self.config.html_show_sphinx,
'has_source': self.config.html_copy_source,
'show_source': self.config.html_show_sourcelink,
'sourcelink_suffix': self.config.html_sourcelink_suffix,
'file_suffix': self.out_suffix,
'link_suffix': self.link_suffix,
'script_files': self._js_files,
'language': convert_locale_to_language_tag(self.config.language),
'css_files': self._css_files,
'sphinx_version': __display_version__,
'sphinx_version_tuple': sphinx_version,
'docutils_version_info': docutils.__version_info__[:5],
'styles': styles,
'rellinks': rellinks,
'builder': self.name,
'parents': [],
'logo_url': logo,
'logo_alt': _('Logo of %s') % self.config.project,
'favicon_url': favicon,
'html5_doctype': True,
}
if self.theme:
self.globalcontext |= {
f'theme_{key}': val
for key, val in self.theme.get_options(self.theme_options).items()
}
self.globalcontext |= self.config.html_context
if self.copysource:
# Create _sources
ensuredir(self._sources_dir)
def get_doc_context(self, docname: str, body: str, metatags: str) -> dict[str, Any]:
"""Collect items for the template context of a page."""
# find out relations
prev = next = None
parents = []
rellinks = self.globalcontext['rellinks'][:]
related = self.relations.get(docname)
titles = self.env.titles
if related and related[2]:
try:
next = {
'link': self.get_relative_uri(docname, related[2]),
'title': self.render_partial(titles[related[2]])['title'],
}
rellinks.append((related[2], next['title'], 'N', _('next')))
except KeyError:
next = None
if related and related[1]:
try:
prev = {
'link': self.get_relative_uri(docname, related[1]),
'title': self.render_partial(titles[related[1]])['title'],
}
rellinks.append((related[1], prev['title'], 'P', _('previous')))
except KeyError:
# the relation is (somehow) not in the TOC tree, handle
# that gracefully
prev = None
while related and related[0]:
with contextlib.suppress(KeyError):
parents.append({
'link': self.get_relative_uri(docname, related[0]),
'title': self.render_partial(titles[related[0]])['title'],
})
related = self.relations.get(related[0])
if parents:
# remove link to the master file; we have a generic
# "back to index" link already
parents.pop()
parents.reverse()
# title rendered as HTML
title_node = self.env.longtitles.get(docname)
title = self.render_partial(title_node)['title'] if title_node else ''
# Suffix for the document
source_suffix = str(self.env.doc2path(docname, False))[len(docname) :]
# the name for the copied source
if self.config.html_copy_source:
sourcename = docname + source_suffix
if source_suffix != self.config.html_sourcelink_suffix:
sourcename += self.config.html_sourcelink_suffix
else:
sourcename = ''
# metadata for the document
meta = self.env.metadata.get(docname)
# local TOC and global TOC tree
self_toc = document_toc(self.env, docname, self.tags)
toc = self.render_partial(self_toc)['fragment']
return {
'parents': parents,
'prev': prev,
'next': next,
'title': title,
'meta': meta,
'body': body,
'metatags': metatags,
'rellinks': rellinks,
'sourcename': sourcename,
'toc': toc,
# only display a TOC if there's more than one item to show
'display_toc': (self.env.toc_num_entries[docname] > 1),
'page_source_suffix': source_suffix,
}
def copy_assets(self) -> None:
self.finish_tasks.add_task(self.copy_download_files)
self.finish_tasks.add_task(self.copy_static_files)
self.finish_tasks.add_task(self.copy_extra_files)
self.finish_tasks.join()
def write_doc(self, docname: str, doctree: nodes.document) -> None:
destination = StringOutput(encoding='utf-8')
doctree.settings = self.docsettings
self.secnumbers = self.env.toc_secnumbers.get(docname, {})
self.fignumbers = self.env.toc_fignumbers.get(docname, {})
self.imgpath = relative_uri(self.get_target_uri(docname), '_images')
self.dlpath = relative_uri(self.get_target_uri(docname), '_downloads')
self.current_docname = docname
self.docwriter.write(doctree, destination)
self.docwriter.assemble_parts()
body = self.docwriter.parts['fragment']
metatags = self.docwriter.clean_meta
ctx = self.get_doc_context(docname, body, metatags)
ctx['has_maths_elements'] = self.docwriter._has_maths_elements
self.handle_page(docname, ctx, event_arg=doctree)
def write_doc_serialized(self, docname: str, doctree: nodes.document) -> None:
self.imgpath = relative_uri(self.get_target_uri(docname), self.imagedir)
self.post_process_images(doctree)
title_node = self.env.longtitles.get(docname)
title = self.render_partial(title_node)['title'] if title_node else ''
self.index_page(docname, doctree, title)
def finish(self) -> None:
self.finish_tasks.add_task(self.gen_indices)
self.finish_tasks.add_task(self.gen_pages_from_extensions)
self.finish_tasks.add_task(self.gen_additional_pages)
self.finish_tasks.add_task(self.copy_image_files)
self.finish_tasks.add_task(self.write_buildinfo)
# dump the search index
self.handle_finish()
@progress_message(__('generating indices'))
def gen_indices(self) -> None:
# the global general index
if self.use_index:
self.write_genindex()
# the global domain-specific indices
self.write_domain_indices()
def gen_pages_from_extensions(self) -> None:
# pages from extensions
for pagelist in self.events.emit('html-collect-pages'):
for pagename, context, template in pagelist:
self.handle_page(pagename, context, template)
@progress_message(__('writing additional pages'))
def gen_additional_pages(self) -> None:
# additional pages from conf.py
for pagename, template in self.config.html_additional_pages.items():
logger.info('%s ', pagename, nonl=True)
self.handle_page(pagename, {}, template)
# the search page
if self.search:
logger.info('search ', nonl=True)
self.handle_page('search', {}, 'search.html')
# the opensearch xml file
if self.config.html_use_opensearch and self.search:
logger.info('opensearch ', nonl=True)
self.handle_page(
'opensearch',
{},
'opensearch.xml',
outfilename=self._static_dir / 'opensearch.xml',
)
def write_genindex(self) -> None:
# the total count of lines for each index letter, used to distribute
# the entries into two columns
genindex = IndexEntries(self.env).create_index(self)
indexcounts = [
sum(1 + len(subitems) for _, (_, subitems, _) in entries)
for _k, entries in genindex
]
genindexcontext = {
'genindexentries': genindex,
'genindexcounts': indexcounts,
'split_index': self.config.html_split_index,
}
logger.info('genindex ', nonl=True)
if self.config.html_split_index:
self.handle_page('genindex', genindexcontext, 'genindex-split.html')
self.handle_page('genindex-all', genindexcontext, 'genindex.html')
for (key, entries), count in zip(genindex, indexcounts, strict=True):
ctx = {
'key': key,
'entries': entries,
'count': count,
'genindexentries': genindex,
}
self.handle_page('genindex-' + key, ctx, 'genindex-single.html')
else:
self.handle_page('genindex', genindexcontext, 'genindex.html')
def write_domain_indices(self) -> None:
for index_name, index_cls, content, collapse in self.domain_indices:
index_context = {
'indextitle': index_cls.localname,
'content': content,
'collapse_index': collapse,
}
logger.info('%s ', index_name, nonl=True)
self.handle_page(index_name, index_context, 'domainindex.html')
def copy_image_files(self) -> None:
if self.images:
stringify_func = ImageAdapter(self.env).get_original_image_uri
ensuredir(self._images_dir)
for src in status_iterator(
self.images,
__('copying images... '),
'brown',
len(self.images),
self.app.verbosity,
stringify_func=stringify_func,
):
dest = self.images[src]
try:
copyfile(
self.srcdir / src,
self._images_dir / dest,
force=True,
)
except Exception as err:
logger.warning(
__("cannot copy image file '%s': %s"), self.srcdir / src, err
)
def copy_download_files(self) -> None:
def to_relpath(f: str) -> str:
return _relative_path(Path(f), self.srcdir).as_posix()
# copy downloadable files
if self.env.dlfiles:
ensuredir(self._downloads_dir)
for src in status_iterator(
self.env.dlfiles,
__('copying downloadable files... '),
'brown',
len(self.env.dlfiles),
self.app.verbosity,
stringify_func=to_relpath,
):
try:
dest = self._downloads_dir / self.env.dlfiles[src][1]
ensuredir(dest.parent)
copyfile(self.srcdir / src, dest, force=True)
except OSError as err:
logger.warning(
__('cannot copy downloadable file %r: %s'),
self.srcdir / src,
err,
)
def create_pygments_style_file(self) -> None:
"""Create a style file for pygments."""
pyg_path = self._static_dir / 'pygments.css'
with open(pyg_path, 'w', encoding='utf-8') as f:
f.write(self.highlighter.get_stylesheet())
if self.dark_highlighter:
dark_path = self._static_dir / 'pygments_dark.css'
with open(dark_path, 'w', encoding='utf-8') as f:
f.write(self.dark_highlighter.get_stylesheet())
def copy_translation_js(self) -> None:
"""Copy a JavaScript file for translations."""
if js_file := self._get_translations_js():
copyfile(
js_file,
self._static_dir / 'translations.js',
force=True,
)
def copy_stemmer_js(self) -> None:
"""Copy a JavaScript file for stemmer."""
if self.indexer is not None:
if hasattr(self.indexer, 'get_js_stemmer_rawcodes'):
for jsfile in self.indexer.get_js_stemmer_rawcodes():
js_path = Path(jsfile)
copyfile(
js_path,
self._static_dir / js_path.name,
force=True,
)
else:
if js_stemmer_rawcode := self.indexer.get_js_stemmer_rawcode():
copyfile(
js_stemmer_rawcode,
self._static_dir / '_stemmer.js',
force=True,
)
def copy_theme_static_files(self, context: dict[str, Any]) -> None:
def onerror(filename: str, error: Exception) -> None:
msg = __("Failed to copy a file in the theme's 'static' directory: %s: %r")
logger.warning(msg, filename, error)
if self.theme:
for entry in reversed(self.theme.get_theme_dirs()):
copy_asset(
Path(entry, 'static'),
self._static_dir,
excluded=DOTFILES,
context=context,
renderer=self.templates,
onerror=onerror,
force=True,
)
def copy_html_static_files(self, context: dict[str, Any]) -> None:
def onerror(filename: str, error: Exception) -> None:
logger.warning(
__('Failed to copy a file in html_static_file: %s: %r'), filename, error
)
excluded = Matcher([*self.config.exclude_patterns, '**/.*'])
for entry in self.config.html_static_path:
copy_asset(
self.confdir / entry,
self._static_dir,
excluded=excluded,
context=context,
renderer=self.templates,
onerror=onerror,
force=True,
)
def copy_html_logo(self) -> None:
if self.config.html_logo and not is_url(self.config.html_logo):
source_path = self.confdir / self.config.html_logo
copyfile(
source_path,
self._static_dir / source_path.name,
force=True,
)
def copy_html_favicon(self) -> None:
if self.config.html_favicon and not is_url(self.config.html_favicon):
source_path = self.confdir / self.config.html_favicon
copyfile(
source_path,
self._static_dir / source_path.name,
force=True,
)
def copy_static_files(self) -> None:
try:
with progress_message(__('copying static files'), nonl=False):
# Ensure that the static directory exists
self._static_dir.mkdir(parents=True, exist_ok=True)
# prepare context for templates
context = self.globalcontext.copy()
if self.indexer is not None:
context.update(self.indexer.context_for_searchtool())
self.create_pygments_style_file()
self.copy_translation_js()
self.copy_stemmer_js()
self.copy_theme_static_files(context)
self.copy_html_static_files(context)
self.copy_html_logo()
self.copy_html_favicon()
except OSError as err:
logger.warning(__('cannot copy static file %r'), err)
def copy_extra_files(self) -> None:
"""Copy html_extra_path files."""
try:
with progress_message(__('copying extra files'), nonl=False):
excluded = Matcher(self.config.exclude_patterns)
for extra_path in self.config.html_extra_path:
copy_asset(
self.confdir / extra_path,
self.outdir,
excluded=excluded,
force=True,
)
except OSError as err:
logger.warning(__('cannot copy extra file %r'), err)
def write_buildinfo(self) -> None:
try:
self.build_info.dump(self.outdir / '.buildinfo')
except OSError as exc:
logger.warning(__('Failed to write build info file: %r'), exc)
def cleanup(self) -> None:
# clean up theme stuff
if self.theme:
self.theme._cleanup()
def post_process_images(self, doctree: Node) -> None:
"""Pick the best candidate for an image and link down-scaled images to
their high resolution version.
"""
super().post_process_images(doctree)
if self.config.html_scaled_image_link and self.html_scaled_image_link:
for node in doctree.findall(nodes.image):
if not any((key in node) for key in ('scale', 'width', 'height')):
# resizing options are not given. scaled image link is available
# only for resized images.
continue
if isinstance(node.parent, nodes.reference):
# A image having hyperlink target
continue
if 'no-scaled-link' in node['classes']:
# scaled image link is disabled for this node
continue
uri = node['uri']
reference = nodes.reference('', '', internal=True)
if uri in self.images:
reference['refuri'] = posixpath.join(self.imgpath, self.images[uri])
else:
reference['refuri'] = uri
node.replace_self(reference)
reference.append(node)
def load_indexer(self, docnames: Set[str]) -> None:
assert self.indexer is not None
keep = set(self.env.all_docs).difference(docnames)
try:
search_index_path = self.outdir / self.searchindex_filename
if self.indexer_dumps_unicode:
with open(search_index_path, encoding='utf-8') as ft:
self.indexer.load(ft, self.indexer_format)
else:
with open(search_index_path, 'rb') as fb:
self.indexer.load(fb, self.indexer_format)
except (OSError, ValueError):
if keep:
logger.warning(
__(
"search index couldn't be loaded, but not all "
'documents will be built: the index will be '
'incomplete.'
)
)
# delete all entries for files that will be rebuilt
self.indexer.prune(keep)
def index_page(self, pagename: str, doctree: nodes.document, title: str) -> None:
# only index pages with title
if self.indexer is not None and title:
filename = self.env.doc2path(pagename, base=False)
metadata = self.env.metadata.get(pagename, {})
if 'no-search' in metadata or 'nosearch' in metadata:
self.indexer.feed(pagename, filename, '', new_document(''))
else:
self.indexer.feed(pagename, filename, title, doctree)
def _get_local_toctree(
self, docname: str, collapse: bool = True, **kwargs: Any
) -> str:
if 'includehidden' not in kwargs:
kwargs['includehidden'] = False
if kwargs.get('maxdepth') == '': # NoQA: PLC1901
kwargs.pop('maxdepth')
toctree = global_toctree_for_doc(
self.env, docname, self, collapse=collapse, **kwargs
)
return self.render_partial(toctree)['fragment']
def get_output_path(self, page_name: str, /) -> Path:
return Path(self.outdir, page_name + self.out_suffix)
def get_outfilename(self, pagename: str) -> _StrPath:
return _StrPath(self.get_output_path(pagename))
def add_sidebars(self, pagename: str, ctx: dict[str, Any]) -> None:
def has_wildcard(pattern: str) -> bool:
return any(char in pattern for char in '*?[')
matched = None
# default sidebars settings for selected theme
sidebars = list(self.theme.sidebar_templates)
# user sidebar settings
html_sidebars = self.get_builder_config('sidebars', 'html')
msg = __('page %s matches two patterns in html_sidebars: %r and %r')
for pattern, pat_sidebars in html_sidebars.items():
if patmatch(pagename, pattern):
if matched and has_wildcard(pattern):
# warn if both patterns contain wildcards
if has_wildcard(matched):
logger.warning(msg, pagename, matched, pattern)
# else the already matched pattern is more specific
# than the present one, because it contains no wildcard
continue
matched = pattern
sidebars = pat_sidebars
ctx['sidebars'] = list(sidebars)
# --------- these are overwritten by the serialization builder
def get_target_uri(self, docname: str, typ: str | None = None) -> str:
return quote(docname) + self.link_suffix
def handle_page(
self,
pagename: str,
addctx: dict[str, Any],
templatename: str = 'page.html',
*,
outfilename: Path | None = None,
event_arg: Any = None,
) -> None:
ctx = self.globalcontext.copy()
# current_page_name is backwards compatibility
ctx['pagename'] = ctx['current_page_name'] = pagename
ctx['encoding'] = self.config.html_output_encoding
default_baseuri = self.get_target_uri(pagename)
# in the singlehtml builder, default_baseuri still contains an #anchor
# part, which relative_uri doesn't really like...
default_baseuri = default_baseuri.rsplit('#', 1)[0]
if self.config.html_baseurl:
ctx['pageurl'] = posixpath.join(
self.config.html_baseurl, self.get_target_uri(pagename)
)
else:
ctx['pageurl'] = None
def pathto(
otheruri: str,
resource: bool = False,
baseuri: str = default_baseuri,
) -> str:
if resource and '://' in otheruri:
# allow non-local resources given by scheme
return otheruri
elif not resource:
otheruri = self.get_target_uri(otheruri)
uri = relative_uri(baseuri, otheruri) or '#'
if uri == '#' and not self.allow_sharp_as_current_path:
uri = baseuri
return uri
ctx['pathto'] = pathto
def hasdoc(name: str) -> bool:
if name in self.env.all_docs:
return True
if name == 'search' and self.search:
return True
return name == 'genindex' and self.get_builder_config('use_index', 'html')
ctx['hasdoc'] = hasdoc
ctx['toctree'] = lambda **kwargs: self._get_local_toctree(pagename, **kwargs)
self.add_sidebars(pagename, ctx)
ctx.update(addctx)
# 'blah.html' should have content_root = './' not ''.
ctx['content_root'] = (f'..{SEP}' * default_baseuri.count(SEP)) or f'.{SEP}'
outdir = self.app.outdir
def css_tag(css: _CascadingStyleSheet) -> str:
attrs = [
f'{key}="{html.escape(value, quote=True)}"'
for key, value in css.attributes.items()
if value is not None
]
uri = pathto(os.fspath(css.filename), resource=True)
# the EPUB format does not allow the use of query components
# the Windows help compiler requires that css links
# don't have a query component
if self.name not in {'epub', 'htmlhelp'}:
if checksum := _file_checksum(outdir, css.filename):
uri += f'?v={checksum}'
return f''
ctx['css_tag'] = css_tag
def js_tag(js: _JavaScript | str) -> str:
if not isinstance(js, _JavaScript):
# str value (old styled)
return f''
body = js.attributes.get('body', '')
attrs = [
f'{key}="{html.escape(value, quote=True)}"'
for key, value in js.attributes.items()
if key != 'body' and value is not None
]
if not js.filename:
if attrs:
return f''
return f''
js_filename_str = os.fspath(js.filename)
uri = pathto(js_filename_str, resource=True)
if 'MathJax.js?' in js_filename_str:
# MathJax v2 reads a ``?config=...`` query parameter,
# special case this and just skip adding the checksum.
# https://docs.mathjax.org/en/v2.7-latest/configuration.html#considerations-for-using-combined-configuration-files
# https://github.com/sphinx-doc/sphinx/issues/11658
pass
# the EPUB format does not allow the use of query components
elif self.name != 'epub':
if checksum := _file_checksum(outdir, js.filename):
uri += f'?v={checksum}'
if attrs:
return f''
return f''
ctx['js_tag'] = js_tag
# revert _css_files and _js_files
self._css_files[:] = self._orig_css_files
self._js_files[:] = self._orig_js_files
self.update_page_context(pagename, templatename, ctx, event_arg)
if new_template := self.events.emit_firstresult(
'html-page-context', pagename, templatename, ctx, event_arg
):
templatename = new_template
# sort JS/CSS before rendering HTML
try: # NoQA: SIM105
# Convert script_files to list to support non-list script_files
# See: https://github.com/sphinx-doc/sphinx/issues/8889
ctx['script_files'] = sorted(
ctx['script_files'], key=lambda js: js.priority
)
except AttributeError:
# Skip sorting if users modifies script_files directly (maybe via `html_context`).
# See: https://github.com/sphinx-doc/sphinx/issues/8885
#
# Note: priority sorting feature will not work in this case.
pass
with contextlib.suppress(AttributeError):
ctx['css_files'] = sorted(ctx['css_files'], key=lambda css: css.priority)
try:
output = self.templates.render(templatename, ctx)
except UnicodeError:
logger.warning(
__(
'a Unicode error occurred when rendering the page %s. '
'Please make sure all config values that contain '
'non-ASCII content are Unicode strings.'
),
pagename,
)
return
except Exception as exc:
msg = __('An error happened in rendering the page %s.\nReason: %r') % (
pagename,
exc,
)
raise ThemeError(msg) from exc
if outfilename:
output_path = Path(outfilename)
else:
output_path = self.get_output_path(pagename)
# The output path is in general different from self.outdir
ensuredir(output_path.parent)
try:
output_path.write_text(
output, encoding=ctx['encoding'], errors='xmlcharrefreplace'
)
except OSError as err:
logger.warning(__('error writing file %s: %s'), output_path, err)
if self.copysource and ctx.get('sourcename'):
# copy the source file for the "show source" link
source_file_path = self._sources_dir / ctx['sourcename']
source_file_path.parent.mkdir(parents=True, exist_ok=True)
copyfile(self.env.doc2path(pagename), source_file_path, force=True)
def update_page_context(
self, pagename: str, templatename: str, ctx: dict[str, Any], event_arg: Any
) -> None:
pass
def handle_finish(self) -> None:
self.finish_tasks.add_task(self.dump_search_index)
self.finish_tasks.add_task(self.dump_inventory)
@progress_message(__('dumping object inventory'))
def dump_inventory(self) -> None:
InventoryFile.dump(self.outdir / INVENTORY_FILENAME, self.env, self)
def dump_search_index(self) -> None:
if self.indexer is None:
return
with progress_message(__('dumping search index in %s') % self.indexer.label()):
self.indexer.prune(self.env.all_docs)
search_index_path = self.outdir / self.searchindex_filename
search_index_tmp = self.outdir / f'{self.searchindex_filename}.tmp'
# first write to a temporary file, so that if dumping fails,
# the existing index won't be overwritten
if self.indexer_dumps_unicode:
with open(search_index_tmp, 'w', encoding='utf-8') as ft:
self.indexer.dump(ft, self.indexer_format)
else:
with open(search_index_tmp, 'wb') as fb:
self.indexer.dump(fb, self.indexer_format)
Path(search_index_tmp).replace(search_index_path)
def convert_html_css_files(app: Sphinx, config: Config) -> None:
"""Convert string styled html_css_files to tuple styled one."""
html_css_files: list[tuple[str, dict[str, str]]] = []
for entry in config.html_css_files:
if isinstance(entry, str):
html_css_files.append((entry, {}))
else:
try:
filename, attrs = entry
html_css_files.append((filename, attrs))
except Exception:
logger.warning(__('invalid css_file: %r, ignored'), entry)
continue
config.html_css_files = html_css_files
def convert_html_js_files(app: Sphinx, config: Config) -> None:
"""Convert string styled html_js_files to tuple styled one."""
html_js_files: list[tuple[str, dict[str, str]]] = []
for entry in config.html_js_files:
if isinstance(entry, str):
html_js_files.append((entry, {}))
else:
try:
filename, attrs = entry
html_js_files.append((filename, attrs))
except Exception:
logger.warning(__('invalid js_file: %r, ignored'), entry)
continue
config.html_js_files = html_js_files
def setup_resource_paths(
app: Sphinx,
pagename: str,
templatename: str,
context: dict[str, Any],
doctree: Node,
) -> None:
"""Set up relative resource paths."""
pathto = context['pathto']
# favicon_url
favicon_url = context.get('favicon_url')
if favicon_url and not is_url(favicon_url):
context['favicon_url'] = pathto('_static/' + favicon_url, resource=True)
# logo_url
logo_url = context.get('logo_url')
if logo_url and not is_url(logo_url):
context['logo_url'] = pathto('_static/' + logo_url, resource=True)
def validate_math_renderer(app: Sphinx) -> None:
if app.builder.format != 'html':
return
name = app.builder.math_renderer_name # type: ignore[attr-defined]
if name is None:
msg = __(
'Many math_renderers are registered. But no math_renderer is selected.'
)
raise ConfigError(msg)
if name not in app.registry.html_inline_math_renderers:
raise ConfigError(__('Unknown math_renderer %r is given.') % name)
def validate_html_extra_path(app: Sphinx, config: Config) -> None:
"""Check html_extra_paths setting."""
html_extra_path = []
for entry in config.html_extra_path:
extra_path = (app.confdir / entry).resolve()
if extra_path.exists():
if (
app.outdir.drive == extra_path.drive
and extra_path.is_relative_to(app.outdir)
): # fmt: skip
logger.warning(
__('html_extra_path entry %r is placed inside outdir'), entry
)
else:
html_extra_path.append(entry)
else:
logger.warning(__('html_extra_path entry %r does not exist'), entry)
config.html_extra_path = html_extra_path
def validate_html_static_path(app: Sphinx, config: Config) -> None:
"""Check html_static_paths setting."""
html_static_path = []
for entry in config.html_static_path:
static_path = (app.confdir / entry).resolve()
if static_path.exists():
if (
app.outdir.drive == static_path.drive
and static_path.is_relative_to(app.outdir)
): # fmt: skip
logger.warning(
__('html_static_path entry %r is placed inside outdir'), entry
)
else:
html_static_path.append(entry)
else:
logger.warning(__('html_static_path entry %r does not exist'), entry)
config.html_static_path = html_static_path
def validate_html_logo(app: Sphinx, config: Config) -> None:
"""Check html_logo setting."""
if (
config.html_logo
and not (app.confdir / config.html_logo).is_file()
and not is_url(config.html_logo)
):
logger.warning(__('logo file %r does not exist'), config.html_logo)
config.html_logo = None
def validate_html_favicon(app: Sphinx, config: Config) -> None:
"""Check html_favicon setting."""
if (
config.html_favicon
and not (app.confdir / config.html_favicon).is_file()
and not is_url(config.html_favicon)
):
logger.warning(__('favicon file %r does not exist'), config.html_favicon)
config.html_favicon = None
def error_on_html_sidebars_string_values(app: Sphinx, config: Config) -> None:
"""Support removed in Sphinx 2."""
errors = {}
for pattern, pat_sidebars in config.html_sidebars.items():
if isinstance(pat_sidebars, str):
errors[pattern] = [pat_sidebars]
if not errors:
return
msg = __(
"Values in 'html_sidebars' must be a list of strings. "
'At least one pattern has a string value: %s. '
'Change to `html_sidebars = %r`.'
)
bad_patterns = ', '.join(map(repr, errors))
fixed = config.html_sidebars | errors
raise ConfigError(msg % (bad_patterns, fixed))
def error_on_html_4(_app: Sphinx, config: Config) -> None:
"""Error on HTML 4."""
if config.html4_writer:
msg = __(
'HTML 4 is no longer supported by Sphinx. '
'("html4_writer=True" detected in configuration options)',
)
raise ConfigError(msg)
def setup(app: Sphinx) -> ExtensionMetadata:
# builders
app.add_builder(StandaloneHTMLBuilder)
# config values
app.add_config_value('html_theme', 'alabaster', 'html', types=frozenset({str}))
app.add_config_value('html_theme_path', [], 'html', types=frozenset({list, tuple}))
app.add_config_value('html_theme_options', {}, 'html', types=frozenset({dict}))
app.add_config_value(
'html_title',
lambda c: _('%s %s documentation') % (c.project, c.release),
'html',
types=frozenset({str}),
)
app.add_config_value(
'html_short_title', lambda self: self.html_title, 'html', types=frozenset({str})
)
app.add_config_value(
'html_style', None, 'html', types=frozenset({list, str, tuple})
)
app.add_config_value('html_logo', None, 'html', types=frozenset({str}))
app.add_config_value('html_favicon', None, 'html', types=frozenset({str}))
app.add_config_value('html_css_files', [], 'html', types=frozenset({list, tuple}))
app.add_config_value('html_js_files', [], 'html', types=frozenset({list, tuple}))
app.add_config_value('html_static_path', [], 'html', types=frozenset({list, tuple}))
app.add_config_value('html_extra_path', [], 'html', types=frozenset({list, tuple}))
app.add_config_value('html_last_updated_fmt', None, 'html', types=frozenset({str}))
app.add_config_value(
'html_last_updated_use_utc', False, 'html', types=frozenset({bool})
)
app.add_config_value('html_sidebars', {}, 'html', types=frozenset({dict}))
app.add_config_value('html_additional_pages', {}, 'html', types=frozenset({dict}))
app.add_config_value(
'html_domain_indices',
True,
'html',
types=frozenset({frozenset, list, set, tuple}),
)
app.add_config_value('html_permalinks', True, 'html', types=frozenset({bool}))
app.add_config_value('html_permalinks_icon', 'ΒΆ', 'html', types=frozenset({str}))
app.add_config_value('html_use_index', True, 'html', types=frozenset({bool}))
app.add_config_value('html_split_index', False, 'html', types=frozenset({bool}))
app.add_config_value('html_copy_source', True, 'html', types=frozenset({bool}))
app.add_config_value('html_show_sourcelink', True, 'html', types=frozenset({bool}))
app.add_config_value(
'html_sourcelink_suffix', '.txt', 'html', types=frozenset({str})
)
app.add_config_value('html_use_opensearch', '', 'html', types=frozenset({str}))
app.add_config_value('html_file_suffix', None, 'html', types=frozenset({str}))
app.add_config_value('html_link_suffix', None, 'html', types=frozenset({str}))
app.add_config_value('html_show_copyright', True, 'html', types=frozenset({bool}))
app.add_config_value(
'html_show_search_summary', True, 'html', types=frozenset({bool})
)
app.add_config_value('html_show_sphinx', True, 'html', types=frozenset({bool}))
app.add_config_value('html_context', {}, 'html', types=frozenset({dict}))
app.add_config_value(
'html_output_encoding', 'utf-8', 'html', types=frozenset({str})
)
app.add_config_value('html_compact_lists', True, 'html', types=frozenset({bool}))
app.add_config_value('html_secnumber_suffix', '. ', 'html', types=frozenset({str}))
app.add_config_value('html_search_language', None, 'html', types=frozenset({str}))
app.add_config_value('html_search_options', {}, 'html', types=frozenset({dict}))
app.add_config_value('html_search_scorer', '', '', types=frozenset({str}))
app.add_config_value(
'html_scaled_image_link', True, 'html', types=frozenset({bool})
)
app.add_config_value('html_baseurl', '', 'html', types=frozenset({str}))
# removal is indefinitely on hold
# See: https://github.com/sphinx-doc/sphinx/issues/10265
app.add_config_value(
'html_codeblock_linenos_style', 'inline', 'html', types=ENUM('table', 'inline')
)
app.add_config_value(
'html_math_renderer', None, 'env', types=frozenset({str, NoneType})
)
app.add_config_value('html4_writer', False, 'html', types=frozenset({bool}))
# events
app.add_event('html-collect-pages')
app.add_event('html-page-context')
# event handlers
app.connect('config-inited', convert_html_css_files, priority=800)
app.connect('config-inited', convert_html_js_files, priority=800)
app.connect('config-inited', validate_html_extra_path, priority=800)
app.connect('config-inited', validate_html_static_path, priority=800)
app.connect('config-inited', validate_html_logo, priority=800)
app.connect('config-inited', validate_html_favicon, priority=800)
app.connect('config-inited', error_on_html_sidebars_string_values, priority=800)
app.connect('config-inited', error_on_html_4, priority=800)
app.connect('builder-inited', validate_math_renderer)
app.connect('html-page-context', setup_resource_paths)
# load default math renderer
app.setup_extension('sphinx.ext.mathjax')
return {
'version': 'builtin',
'parallel_read_safe': True,
'parallel_write_safe': True,
}
# deprecated name -> (object to return, canonical path or empty string, removal version)
_DEPRECATED_OBJECTS: dict[str, tuple[Any, str, tuple[int, int]]] = {
'Stylesheet': (
_CascadingStyleSheet,
'sphinx.builders.html._assets._CascadingStyleSheet',
(9, 0),
),
'JavaScript': (_JavaScript, 'sphinx.builders.html._assets._JavaScript', (9, 0)),
}
def __getattr__(name: str) -> Any:
if name not in _DEPRECATED_OBJECTS:
msg = f'module {__name__!r} has no attribute {name!r}'
raise AttributeError(msg)
from sphinx.deprecation import _deprecation_warning
deprecated_object, canonical_name, remove = _DEPRECATED_OBJECTS[name]
_deprecation_warning(__name__, name, canonical_name, remove=remove)
return deprecated_object