def get_license_text(license, name):
raw_name = name.strip()
name = remove_prefix(remove_prefix(escape_latex(name).strip(), "File:"),
"Datei:")
name = name.replace("\\_", "\\allowbreak\\_")
name = name.replace("-", "\\allowbreak-")
if license:
return "Abb. \\arabic{imagelabel}: \\protect\\href{https://commons.wikimedia.org/wiki/%s}{\\textbf{%s}} by " % (
raw_name, name) + ", ".join(
license["authors"]
) + " \\textit{(" + license["shortname"] + ")}"
else:
return "Abb. \\arabic{imagelabel}: could not get licensing information!"
def transform_template(self, obj):
if obj["name"].startswith("#lst:"):
article_name = remove_prefix(obj["name"], "#lst:")
article = self.api.get_content(article_name)
section_name = obj["params"]["1"]
begin = r"\<section\s+begin\=[\"']?" + re.escape(section_name) + \
r"[\"']?\s*\/>"
end = r"\<section\s+end\=[\"']?" + re.escape(section_name) + \
r"[\"']?\s*\/\>"
section = re.search(begin + "(.*)" + end, article, re.DOTALL)
if section:
section = section.group(1).strip()
content = parse_content(self.api, self.title, section)
return {"type": "included_section", "content": content}
else:
message = "section '{}' of '{}' cannot be included" \
.format(section_name, article_name)
return {"type": "error", "message": message}
else:
raise NotInterested()
def normalize(self, obj, mode):
try:
formula = self.api.normalize_formula(obj["formula"], mode)
if mode == "tex":
formula = remove_prefix(formula, "{\\begin{aligned}")
formula = remove_suffix(formula, "\\end{aligned}}")
except ValueError:
message = "Wrong formatted formula"
# TODO: current_section was not set for this class
log_parser_error(message, obj)
return {"type": "error", "message": message}
return merge(obj, {"formula": formula})
def generate_sitemap_nodes(sitemap_text):
"""Generator for all node specifications in a sitemap source code. It
yields dictionaries of the form:
{ "code": code, "depth": depth, "children": [] }
Thereby `code` is a string representation of the node and `depth` is a
number corresponding to the node's depth. The higher the depth is, the
deeper the node need to be included in the final tree.
"""
# In MediaWiki the maximal depth of a headline is 6 (as in HTML).
# For list elements this maximal header depth is added so that list
# elements will always be included under a headline node.
max_headline_depth = 6
headline_re = re.compile(
r"""(={1,%s}) # Equal signs of the headline
(.*) # code defining the node
\1 # Repeatation of the equal signs
""" % max_headline_depth, re.X)
list_re = re.compile(
r"""([*]+) # asteriks of a list element
(.*) # code defining a sitemap node
""", re.X)
for line in sitemap_text.splitlines():
for regex, depth_start in ((headline_re, 0), (list_re,
max_headline_depth)):
match = regex.fullmatch(line.strip())
if match:
yield {
"code": match.group(2).strip(),
"depth": depth_start + len(match.group(1)),
"children": []
}
for an_type, an_prefix in SITEMAP_ANNOTATIONS.items():
if line.startswith(an_prefix):
value = remove_prefix(line, an_prefix).strip()
yield {
"type": "annotation",
"value": value,
"annotation_type": an_type
}
def transform_element(self, obj):
if lookup(obj, "attrs", "about") in self._template_ids:
return None
check(obj, "attrs", "typeof").of([
"mw:Transclusion", "mw:Transclusion mw:Video/Thumb",
"mw:Transclusion mw:Image"
])
template = json.loads(obj["attrs"]["data-mw"])["parts"][0]
try:
template = template["template"]
except (TypeError, KeyError):
return {
"type": "error",
"message": "Template spans over several HTML elements."
}
name = template["target"]["wt"].strip()
# labeled section transclusion needs unchanged case.
if not name.startswith("#lst:"):
name = name.lower()
if name != "(!":
# Template includes a table afterwards
self._template_ids.add(obj["attrs"]["about"])
name = remove_prefix(name, ":mathe für nicht-freaks: vorlage:")
params = template["params"]
params = {k: v["wt"] for k, v in params.items()}
params = {key: self.parse_parameter_value(name, key, value) \
for key, value in params.items()
if not params.get(key + "-noprint", False)}
# TODO: Find better solution
if params.get("noprint", False):
return None
return {"type": "template", "name": name, "params": params}
def test_remove_prefix(self):
self.assertEqual(remove_prefix("aa", "a"), "a")
self.assertEqual(remove_prefix("aa", ""), "aa")
self.assertEqual(remove_prefix("aa", "aaa"), "aa")
self.assertEqual(remove_prefix("a 4 2", "a 4"), " 2")
self.assertEqual(remove_prefix("", "a 4"), "")
def transform_template(self, obj):
for bname, tname, param_names in BOXSPEC:
if obj["name"] == tname:
params = {
k: self(obj["params"].get(v, None))
for k, v in param_names.items()
}
return merge(params, {"type": bname})
if obj["name"] == "liste":
if "liste" in obj["params"]:
sublist = obj["params"]["liste"][0]
assert sublist["type"] == "list"
items = sublist["items"]
ordered = sublist["ordered"]
else:
items = [{
"type": "listitem",
"content": self(x)
} for x in obj["params"]["item_list"]]
ordered = obj["params"].get("type", "") == "ol"
return {
"type": "list",
"items": items,
"ordered": ordered,
"spacing": obj["params"].get("abstand", None)
}
elif obj["name"] == "formel":
formula = obj["params"].get("1", [])
if len(formula) == 1 and \
lookup(formula, 0, "type") == "inlinemath":
formula = formula[0]["formula"]
if formula.startswith(
"\\begin{align}") and formula.endswith(
"\\end{align}"):
formula = remove_prefix(formula, "\\begin{align}")
formula = remove_suffix(formula, "\\end{align}")
formula = "\\begin{align}" + formula + "\\end{align}"
return {"type": "equation", "formula": formula}
else:
message = "Wrong formatted equation"
details = "Equation source code must be completely contained in just one <math></math>.\n (use \\text{this is not math} macro instead)"
log_parser_error(message, obj, details,
self.current_section)
return {"type": "error", "message": message}
elif obj["name"] == "(!":
return None
elif obj["name"].startswith("#invoke:"):
# Template is header or footer
return None
elif obj["name"] == "noprint":
return None
elif obj["name"] == "todo":
message = "Todo-Message in MediaWiki code."
details = "Check if this TODO shoud be completed for a book release."
log_parser_error(message, obj, details, self.current_section)
return {"type": "error", "message": message}
else:
message = "Parsing of template `{}`".format(obj["name"])
log_parser_error(message, obj, position=self.current_section)
return {
"type": "notimplemented",
"target": obj,
"message": message
}
def canonical_image_name(name):
name = remove_prefix(name, "./")
name = remove_prefix(name, "Datei:")
name = remove_prefix(name, "File:")
return "File:" + name