def _build_html_head(self):
'''Add title, metadata, and styles'''
self.elements.head.append(E.meta(charset='UTF-8'))
self._build_html_title()
for key, value in self._default_meta_values:
self.meta[key] = value
self._build_html_head_meta()
self._build_html_head_links()
self._build_html_stylesheets()
def gen_body_redirect(code, location):
from lxml.html.builder import E
from lxml.html import tostring
return tostring(E.HTML(
E.HEAD(
E.meta(**{
"http-equiv": "content-type",
"content": "text/html;charset=utf-8",
}),
E.TITLE(code),
),
E.BODY(
E.H1(code),
E.P("The document has moved"),
E.A("here", HREF=location),
".",
)
))
def gen_body_redirect(code, location):
from lxml.html.builder import E
from lxml.html import tostring
return tostring(
E.HTML(
E.HEAD(
E.meta(
**{
"http-equiv": "content-type",
"content": "text/html;charset=utf-8",
}),
E.TITLE(code),
),
E.BODY(
E.H1(code),
E.P("The document has moved"),
E.A("here", HREF=location),
".",
)))
def _build_html_head_meta(self):
'''Builds the meta elements.'''
head = self.elements.head
for key, value in self.meta.items():
head.append(E.meta(name=key, content=value))
def main():
# TODO: combine command-line and option file.
# TODO: option to generate a default configuration file
parser = argparse.ArgumentParser() # TODO: doc
parser.add_argument("-s", "--standalone", action="store_true") # TODO: doc
args = parser.parse_args()
standalone = args.standalone
conf = json.load((DATA / "artdoc.js").open())
if Path("artdoc.js").exists():
user_conf = json.load(Path("artdoc.js").open())
conf.update(user_conf)
info("Document:")
doc_patterns = conf["doc"]
if isinstance(doc_patterns, basestring):
doc_patterns = [doc_patterns]
docs = []
for pattern in doc_patterns:
matches = list(WORKDIR.glob(pattern))
#subinfo("matching {!r}:".format(pattern))
for match in matches:
subinfo(str(match))
docs.extend(matches)
if not docs:
sys.exit("error: no document found")
# info("HTML template:")
# template_file = HTML / "index.html"
# subinfo(str(template_file))
# template = template_file.open().read().encode("utf-8")
info("Bibliography:")
bib_patterns = conf["bib"]
if isinstance(bib_patterns, basestring):
bib_patterns = [bib_patterns]
bibs = []
for pattern in bib_patterns:
matches = list(WORKDIR.glob(pattern))
#subinfo("matching {!r}:".format(pattern))
for match in matches:
subinfo(str(match))
bibs.extend(matches)
if not bibs:
print()
info("JS:")
cmd = coffee["-c", str(JS / "main.coffee")]
subinfo(cmd)
cmd()
info("CSS:")
cmd = stylus[str(CSS / "style.styl")]
subinfo(str(cmd))
cmd()
# TODO: copy only what is required.
shutil.copytree(str(DATA), str(ARTDOC))
for doc in docs:
pass
info("PANDOC: generate JSON file")
args = ["-t", "json", "--smart"]
for bib in bibs:
args.extend(["--bibliography", str(bib)])
args.append(str(doc))
cmd = pandoc[args]
subinfo(cmd, "> json")
json_str = cmd()
info("Convert raw TeX to raw HTML")
cmd = local[str(BIN / "rawHTML.hs")]
subinfo(cmd, "< json > json")
json_str = (cmd << json_str)()
# info("Flag/Box Proofs")
# cmd = local[str(BIN / "proof.hs")]
# subinfo(cmd, "< json > json")
# try:
# json_str = (cmd << json_str)()
# except Exception as error:
# print(repr(error))
# info("Wrap Section-Like Sequence of Blocks")
# cmd = local[str(BIN / "div.hs")]
# subinfo(cmd, "< json > json")
# try:
# json_str = (cmd << json_str)()
# except Exception as error:
# print(repr(error))
info("Wrap Section-Like Sequence of Blocks")
cmd = local[str(BIN / "section.hs")]
subinfo(cmd, "< json > json")
try:
json_str = (cmd << json_str)()
except Exception as error:
print(repr(error))
info("Flag Tombstones (end of proofs)")
cmd = local[str(BIN / "tombstone.hs")]
subinfo(cmd, "< json > json")
try:
json_str = (cmd << json_str)()
except Exception as error:
print(repr(error))
info("Convert Images to SVG Images")
cmd = local[str(BIN / "svg.hs")]
subinfo(cmd, "< json > json")
json_str = (cmd << json_str)()
info("Generate HTML body from markdown")
args = ["--email-obfuscation", "none",
"-f", "json",
"--mathjax",
"-t", "html5", "--section-divs"]
cmd = pandoc[args]
subinfo(cmd, "< json > body")
pandoc_body_str = (cmd << json_str)()
pandoc_html = lxml.html.document_fromstring(pandoc_body_str)
pandoc_body = pandoc_html.cssselect("body")[0]
info("Generate standalone HTML doc")
html = HTML.html(HTML.head, HTML.body)
body = html.cssselect("body")[0]
head = html.cssselect("head")[0]
head.append(HTML.meta(charset="utf-8"))
body.attrib.update(pandoc_body.attrib)
body.extend(pandoc_body[:])
# ----------------------------------------------------------------------
info("Add JQuery")
head.extend(jquery(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Velocity")
head.extend(velocity(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Clipboard.js")
head.extend(clipboard(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Highlight.js")
head.extend(highlight(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Google Fonts support")
head.extend(google_fonts(["Alegreya", "Alegreya SC"], standalone=standalone))
# ----------------------------------------------------------------------
info("Add Mathjax support")
head.extend(mathjax(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Font Awesome support")
head.extend(font_awesome(standalone=standalone))
# ----------------------------------------------------------------------
info("Add artdoc css & js files")
head.extend(artdoc())
# ----------------------------------------------------------------------
info("Setting language to english (required for hyphens)")
html.set("lang", "en")
# ----------------------------------------------------------------------
info("Ensure ids uniqueness")
id_count = {}
for elt in html.iter():
_id = elt.get("id")
if _id is not None:
count = id_count.get(_id, 0)
if count > 0:
elt.set("id", _id + "-" + str(count))
id_count[_id] = count + 1
# ----------------------------------------------------------------------
info("Turning headers into self-links")
sections = html.cssselect("section")
for section in sections:
id_ = section.get("id")
heading = None
if len(section):
first = section[0]
if first.tag in "h1 h2 h3 h4 h5 h6".split():
heading = first
if id_ and heading is not None:
contents = [heading.text or ""] + heading[:]
heading.text, heading[:] = None, []
href = {"href": "#" + id_}
link = HTML.a(href, *contents)
heading.insert(0, link)
# ----------------------------------------------------------------------
# TODO: deal with metadata & insert a document header with:
# - title,
# - date (format: Month Day, Year), autoformat, autogen ?
# - author(s) (with mail & affiliation when available ?).
# Assume custom metadata or parse the author field ?
# Representation of multiple authors ? MMm eLIFEsciences use
# popup for author info. Ex: http://elifesciences.org/content/4/e06356 !
# here, use hints from http://dtd.nlm.nih.gov/book/tag-library/:
#
# - name (don't be more precise)
# - affiliation (concatenate)
# - address ???
# - email --> Font Awesome Icon
# - url / uri ?
# - form of ID ? (like HAL ? or ZBlatt ?)
# TODO: look at the rendering of
# http://kieranhealy.org/blog/archives/2014/01/23/plain-text/:
# - small grey date on top, bold title, bold author name,
# italics affiliation, repeat.
metadata = get_metadata(str(doc))
items = []
date = parse_html(metadata.get("date"))
if date is not None:
items.append(HTML.p({"class": "date"}, *date))
# def textify(item):
# if isinstance(item, basestring):
# return item
# elif hasattr(item, "text"):
# return item.text
# else:
# return "".join([textify(it) or "" for it in item])
title = parse_html(metadata.get("title"))
title_id = None
if title is not None:
#title_id = textify(title).lower().replace(" ", "-")
items.append(
HTML.h1(
{"class": "title"},
HTML.a(
{"href": "#"},
*title
)
)
)
head.insert(0, HTML.title(*title))
authors = metadata.get("author") or []
for author in authors:
if isinstance(author, basestring):
name = parse_html(author)
email = None
affiliation = None
else:
name = parse_html(author.get("name"))
email = parse_html(author.get("email"))
affiliation = parse_html(author.get("affiliation"))
if name is not None:
if email is not None:
name = [HTML.a({"href": "mailto:" + email[0]}, *name)]
name = HTML.p({"class": "author"}, *name)
items.append(name)
if affiliation is not None:
affiliation = HTML.p({"class": "affiliation"}, *affiliation)
items.append(affiliation)
header_attr = {"class": "main"}
# if title_id is not None:
# header_attr["id"] = title_id
header = HTML.header(header_attr, *items)
# print("HEADER", lxml.html.tostring(header))
body.insert(0, header)
# print("BODY", lxml.html.tostring(body))
# print("HTML", lxml.html.tostring(html))
# ----------------------------------------------------------------------
info("Generate the standalone HTML file")
html_str = lxml.html.tostring(html, encoding="utf-8", doctype="<!DOCTYPE html>")
doc.with_suffix(".html").open("wb").write(html_str)
sys.exit(0)
def main():
# TODO: combine command-line and option file.
# TODO: option to generate a default configuration file
parser = argparse.ArgumentParser() # TODO: doc
parser.add_argument("-s", "--standalone", action="store_true") # TODO: doc
args = parser.parse_args()
standalone = args.standalone
conf = json.load((DATA / "artdoc.js").open())
if Path("artdoc.js").exists():
user_conf = json.load(Path("artdoc.js").open())
conf.update(user_conf)
info("Document:")
doc_patterns = conf["doc"]
if isinstance(doc_patterns, basestring):
doc_patterns = [doc_patterns]
docs = []
for pattern in doc_patterns:
matches = list(WORKDIR.glob(pattern))
#subinfo("matching {!r}:".format(pattern))
for match in matches:
subinfo(str(match))
docs.extend(matches)
if not docs:
sys.exit("error: no document found")
# info("HTML template:")
# template_file = HTML / "index.html"
# subinfo(str(template_file))
# template = template_file.open().read().encode("utf-8")
info("Bibliography:")
bib_patterns = conf["bib"]
if isinstance(bib_patterns, basestring):
bib_patterns = [bib_patterns]
bibs = []
for pattern in bib_patterns:
matches = list(WORKDIR.glob(pattern))
#subinfo("matching {!r}:".format(pattern))
for match in matches:
subinfo(str(match))
bibs.extend(matches)
if not bibs:
print()
info("JS:")
cmd = coffee["-c", str(JS / "main.coffee")]
subinfo(cmd)
cmd()
info("CSS:")
cmd = stylus[str(CSS / "style.styl")]
subinfo(str(cmd))
cmd()
# TODO: copy only what is required.
shutil.copytree(str(DATA), str(ARTDOC))
for doc in docs:
pass
info("PANDOC: generate JSON file")
args = ["-t", "json", "--smart"]
for bib in bibs:
args.extend(["--bibliography", str(bib)])
args.append(str(doc))
cmd = pandoc[args]
subinfo(cmd, "> json")
json_str = cmd()
info("Convert raw TeX to raw HTML")
cmd = local[str(BIN / "rawHTML.hs")]
subinfo(cmd, "< json > json")
json_str = (cmd << json_str)()
# info("Flag/Box Proofs")
# cmd = local[str(BIN / "proof.hs")]
# subinfo(cmd, "< json > json")
# try:
# json_str = (cmd << json_str)()
# except Exception as error:
# print(repr(error))
# info("Wrap Section-Like Sequence of Blocks")
# cmd = local[str(BIN / "div.hs")]
# subinfo(cmd, "< json > json")
# try:
# json_str = (cmd << json_str)()
# except Exception as error:
# print(repr(error))
info("Wrap Section-Like Sequence of Blocks")
cmd = local[str(BIN / "section.hs")]
subinfo(cmd, "< json > json")
try:
json_str = (cmd << json_str)()
except Exception as error:
print(repr(error))
info("Flag Tombstones (end of proofs)")
cmd = local[str(BIN / "tombstone.hs")]
subinfo(cmd, "< json > json")
try:
json_str = (cmd << json_str)()
except Exception as error:
print(repr(error))
info("Convert Images to SVG Images")
cmd = local[str(BIN / "svg.hs")]
subinfo(cmd, "< json > json")
json_str = (cmd << json_str)()
info("Generate HTML body from markdown")
args = [
"--email-obfuscation", "none", "-f", "json", "--mathjax", "-t",
"html5", "--section-divs"
]
cmd = pandoc[args]
subinfo(cmd, "< json > body")
pandoc_body_str = (cmd << json_str)()
pandoc_html = lxml.html.document_fromstring(pandoc_body_str)
pandoc_body = pandoc_html.cssselect("body")[0]
info("Generate standalone HTML doc")
html = HTML.html(HTML.head, HTML.body)
body = html.cssselect("body")[0]
head = html.cssselect("head")[0]
head.append(HTML.meta(charset="utf-8"))
body.attrib.update(pandoc_body.attrib)
body.extend(pandoc_body[:])
# ----------------------------------------------------------------------
info("Add JQuery")
head.extend(jquery(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Velocity")
head.extend(velocity(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Clipboard.js")
head.extend(clipboard(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Highlight.js")
head.extend(highlight(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Google Fonts support")
head.extend(
google_fonts(["Alegreya", "Alegreya SC"], standalone=standalone))
# ----------------------------------------------------------------------
info("Add Mathjax support")
head.extend(mathjax(standalone=standalone))
# ----------------------------------------------------------------------
info("Add Font Awesome support")
head.extend(font_awesome(standalone=standalone))
# ----------------------------------------------------------------------
info("Add artdoc css & js files")
head.extend(artdoc())
# ----------------------------------------------------------------------
info("Setting language to english (required for hyphens)")
html.set("lang", "en")
# ----------------------------------------------------------------------
info("Ensure ids uniqueness")
id_count = {}
for elt in html.iter():
_id = elt.get("id")
if _id is not None:
count = id_count.get(_id, 0)
if count > 0:
elt.set("id", _id + "-" + str(count))
id_count[_id] = count + 1
# ----------------------------------------------------------------------
info("Turning headers into self-links")
sections = html.cssselect("section")
for section in sections:
id_ = section.get("id")
heading = None
if len(section):
first = section[0]
if first.tag in "h1 h2 h3 h4 h5 h6".split():
heading = first
if id_ and heading is not None:
contents = [heading.text or ""] + heading[:]
heading.text, heading[:] = None, []
href = {"href": "#" + id_}
link = HTML.a(href, *contents)
heading.insert(0, link)
# ----------------------------------------------------------------------
# TODO: deal with metadata & insert a document header with:
# - title,
# - date (format: Month Day, Year), autoformat, autogen ?
# - author(s) (with mail & affiliation when available ?).
# Assume custom metadata or parse the author field ?
# Representation of multiple authors ? MMm eLIFEsciences use
# popup for author info. Ex: http://elifesciences.org/content/4/e06356 !
# here, use hints from http://dtd.nlm.nih.gov/book/tag-library/:
#
# - name (don't be more precise)
# - affiliation (concatenate)
# - address ???
# - email --> Font Awesome Icon
# - url / uri ?
# - form of ID ? (like HAL ? or ZBlatt ?)
# TODO: look at the rendering of
# http://kieranhealy.org/blog/archives/2014/01/23/plain-text/:
# - small grey date on top, bold title, bold author name,
# italics affiliation, repeat.
metadata = get_metadata(str(doc))
items = []
date = parse_html(metadata.get("date"))
if date is not None:
items.append(HTML.p({"class": "date"}, *date))
# def textify(item):
# if isinstance(item, basestring):
# return item
# elif hasattr(item, "text"):
# return item.text
# else:
# return "".join([textify(it) or "" for it in item])
title = parse_html(metadata.get("title"))
title_id = None
if title is not None:
#title_id = textify(title).lower().replace(" ", "-")
items.append(
HTML.h1({"class": "title"}, HTML.a({"href": "#"}, *title)))
head.insert(0, HTML.title(*title))
authors = metadata.get("author") or []
for author in authors:
if isinstance(author, basestring):
name = parse_html(author)
email = None
affiliation = None
else:
name = parse_html(author.get("name"))
email = parse_html(author.get("email"))
affiliation = parse_html(author.get("affiliation"))
if name is not None:
if email is not None:
name = [HTML.a({"href": "mailto:" + email[0]}, *name)]
name = HTML.p({"class": "author"}, *name)
items.append(name)
if affiliation is not None:
affiliation = HTML.p({"class": "affiliation"},
*affiliation)
items.append(affiliation)
header_attr = {"class": "main"}
# if title_id is not None:
# header_attr["id"] = title_id
header = HTML.header(header_attr, *items)
# print("HEADER", lxml.html.tostring(header))
body.insert(0, header)
# print("BODY", lxml.html.tostring(body))
# print("HTML", lxml.html.tostring(html))
# ----------------------------------------------------------------------
info("Generate the standalone HTML file")
html_str = lxml.html.tostring(html,
encoding="utf-8",
doctype="<!DOCTYPE html>")
doc.with_suffix(".html").open("wb").write(html_str)
sys.exit(0)