def manual_join(template,
files_contents,
stylesheet,
remove=None,
extra_css=None,
remove_selectors=None,
hook_before_toc=None,
references=None,
resolve_references=True,
hook_before_final_pass=None,
require_toc_placeholder=False,
permalink_prefix=None,
crossrefs_aug=None,
aug0=None):
"""
files_contents: a list of tuples that can be cast to DocToJoin:
where the string is a unique one to be used for job naming.
extra_css: if not None, a string of more CSS to be added
Remove_selectors: list of selectors to remove (e.g. ".draft").
hook_before_toc if not None is called with hook_before_toc(soup=soup)
just before generating the toc
"""
result = AugmentedResult()
if references is None:
references = {}
check_isinstance(files_contents, list)
if crossrefs_aug is None:
crossrefs = Tag(name='no-cross-refs')
else:
crossrefs = bs(crossrefs_aug.get_result())
result.merge(crossrefs_aug)
if aug0 is not None:
result.merge(aug0)
@contextmanager
def timeit(_):
yield
with timeit('manual_join'):
files_contents = [DocToJoin(*_) for _ in files_contents]
# cannot use bs because entire document
with timeit('parsing template'):
template0 = template
template = replace_macros(template)
template_soup = BeautifulSoup(template,
'lxml',
from_encoding='utf-8')
d = template_soup
if d.html is None:
s = "Invalid template"
raise_desc(ValueError, s, template0=template0)
with timeit('adding head'):
assert d.html is not None
assert '<html' in str(d)
head = d.find('head')
if head is None:
msg = 'Could not find <head> in template:'
logger.error(msg)
logger.error(str(d))
raise Exception(msg)
assert head is not None
for x in get_manual_css_frag().contents:
head.append(x.__copy__())
with timeit('adding stylesheet'):
if stylesheet is not None:
link = Tag(name='link')
link['rel'] = 'stylesheet'
link['type'] = 'text/css'
from mcdp_report.html import get_css_filename
link['href'] = get_css_filename('compiled/%s' % stylesheet)
head.append(link)
with timeit('making basename2soup'):
basename2soup = OrderedDict()
for doc_to_join in files_contents:
if doc_to_join.docname in basename2soup:
msg = 'Repeated docname %r' % doc_to_join.docname
raise ValueError(msg)
from .latex.latex_preprocess import assert_not_inside
if isinstance(doc_to_join.contents, AugmentedResult):
result.merge(doc_to_join.contents)
contents = doc_to_join.contents.get_result()
else:
contents = doc_to_join.contents
assert_not_inside(contents, '<fragment')
assert_not_inside(contents, 'DOCTYPE')
frag = bs(contents)
basename2soup[doc_to_join.docname] = frag
# with timeit('fix_duplicate_ids'):
# XXX
# fix_duplicated_ids(basename2soup)
with timeit('copy contents'):
body = d.find('body')
add_comments = False
for docname, content in basename2soup.items():
if add_comments:
body.append(NavigableString('\n\n'))
body.append(
Comment('Beginning of document dump of %r' % docname))
body.append(NavigableString('\n\n'))
try_faster = True
if try_faster:
for e in list(content.children):
body.append(e.extract())
else:
copy_contents_into(content, body)
if add_comments:
body.append(NavigableString('\n\n'))
body.append(Comment('End of document dump of %r' %
docname))
body.append(NavigableString('\n\n'))
with timeit('extract_bibtex_blocks'):
extract_bibtex_blocks(d)
with timeit('ID_PUT_BIB_HERE'):
ID_PUT_BIB_HERE = MCDPManualConstants.ID_PUT_BIB_HERE
bibhere = d.find('div', id=ID_PUT_BIB_HERE)
if bibhere is None:
msg = ('Could not find #%s in document. '
'Adding one at end of document.') % ID_PUT_BIB_HERE
result.note_warning(msg)
bibhere = Tag(name='div')
bibhere.attrs['id'] = ID_PUT_BIB_HERE
d.find('body').append(bibhere)
do_bib(d, bibhere)
with timeit('hook_before_final_pass'):
if hook_before_final_pass is not None:
hook_before_final_pass(soup=d)
with timeit('document_final_pass_before_toc'):
location = LocationUnknown()
document_final_pass_before_toc(d, remove, remove_selectors, result,
location)
with timeit('hook_before_toc'):
if hook_before_toc is not None:
hook_before_toc(soup=d)
with timeit('generate_and_add_toc'):
try:
generate_and_add_toc(d, raise_error=True, res=result)
except NoTocPlaceholder as e:
if require_toc_placeholder:
msg = 'Could not find toc placeholder: %s' % e
# logger.error(msg)
if aug0 is not None:
result.note_error(msg)
else:
raise Exception(msg)
with timeit('document_final_pass_after_toc'):
document_final_pass_after_toc(
soup=d,
crossrefs=crossrefs,
resolve_references=resolve_references,
res=result)
if extra_css is not None:
logger.info('adding extra CSS')
add_extra_css(d, extra_css)
with timeit('document_only_once'):
document_only_once(d)
location = LocationUnknown()
substitute_github_refs(d, defaults={}, res=result, location=location)
with timeit('another A pass'):
for a in d.select('a[href]'):
href = a.attrs['href']
if href in references:
r = references[href]
a.attrs['href'] = r.url
if not a.children: # empty
a.append(r.title)
# do not use to_html_stripping_fragment - this is a complete doc
# mark_in_html(result, soup=d)
add_github_links_if_edit_url(soup=d, permalink_prefix=permalink_prefix)
with timeit('converting to string'):
res = unicode(d)
with timeit('encoding'):
res = res.encode('utf8')
logger.info('done - %.1f MB' % (len(res) / (1024 * 1024.0)))
result.set_result(res)
return result
def get_cross_refs(src_dirs, permalink_prefix, extra_crossrefs, ignore=[]):
res = AugmentedResult()
files = look_for_files(src_dirs, "crossref.html")
id2file = {}
soup = Tag(name='div')
def add_from_soup(s, f, ignore_alread_present, ignore_if_conflict):
for img in list(s.find_all('img')):
img.extract()
for e in s.select('[base_url]'):
e['external_crossref_file'] = f
# Remove the ones with the same base_url
for e in list(s.select('[base_url]')):
if e.attrs['base_url'] == permalink_prefix:
e.extract()
for e in s.select('[id]'):
id_ = e.attrs['id']
if id_ == 'container': continue # XXX:
if id_ in id2file:
if not ignore_alread_present:
msg = 'Found two elements with same ID "%s":' % id_
msg += '\n %s' % id2file[id_]
msg += '\n %s' % f
res.note_error(msg)
else:
id2file[id_] = f
e2 = e.__copy__()
if ignore_if_conflict:
e2.attrs['ignore_if_conflict'] = '1'
soup.append(e2)
soup.append('\n')
ignore = [os.path.realpath(_) for _ in ignore]
for _f in files:
if os.path.realpath(_f) in ignore:
msg = 'Ignoring file %r' % _f
logger.info(msg)
continue
logger.info('cross ref file %s' % _f)
data = open(_f).read()
if permalink_prefix in data:
msg = 'skipping own file'
logger.debug(msg)
continue
s = bs(data)
add_from_soup(s,
_f,
ignore_alread_present=False,
ignore_if_conflict=False)
if extra_crossrefs is not None:
logger.info('Reading external refs\n%s' % extra_crossrefs)
try:
r = requests.get(extra_crossrefs)
except Exception as ex:
msg = 'Could not read external cross reference links'
msg += '\n %s' % extra_crossrefs
msg += '\n\n' + indent(str(ex), ' > ')
res.note_error(msg)
else:
logger.debug('%s %s' % (r.status_code, extra_crossrefs))
if r.status_code == 404:
msg = 'Could not read external cross refs: %s' % r.status_code
msg += '\n url: ' + extra_crossrefs
msg += '\n This is normal if you have not pushed this branch yet.'
res.note_warning(msg)
# logger.error(msg)
s = bs(r.text)
add_from_soup(s,
extra_crossrefs,
ignore_alread_present=True,
ignore_if_conflict=True)
# print soup
res.set_result(str(soup))
return res