def repl_func(m):
"""
Replaces <CIT ID=0 /> with <REF ID=cit0 REFID=ref0 />
"""
cit_id=m.group(1) # this is the actual unique identifier of a citation in the document
ref_id=doc.citation_by_id[cit_id]["ref_id"] # this is the id of the reference the cit cites
return u'<REF ID="'+safe_unicode(cit_id)+u'" REFID="'+safe_unicode(ref_id)+'" />'
def repl_func(m):
"""
Replaces <CIT ID=0 /> with <REF ID=cit0 REFID=ref0 />
"""
cit_id = m.group(
1
) # this is the actual unique identifier of a citation in the document
ref_id = doc.citation_by_id[cit_id][
"ref_id"] # this is the id of the reference the cit cites
return u'<REF ID="' + safe_unicode(
cit_id) + u'" REFID="' + safe_unicode(ref_id) + '" />'
def saveSapientaXML(doc,filename):
"""
Exports a SciDocJSON to a file in Sapienta sort-of SciXML format
"""
header_depth=0
def dumpMetadata(doc):
"""
"""
lines=[]
lines.append('<mode2 hasDoc="yes" name="'+doc.metadata["filename"].replace(".xml","")+'" version="597"/>')
lines.append("<METADATA>")
if doc["metadata"].has_key("fileno"):
lines.append("<FILENO>"+doc.metadata["fileno"]+"</FILENO>\n")
lines.append("<FILENAME>"+doc.metadata["filename"]+"</FILENAME>\n")
lines.append("<APPEARED>")
if "conference" in doc.metadata:
lines.append("<CONFERENCE>"+doc.metadata["conference"]+"</CONFERENCE>\n")
if "journal" in doc.metadata:
lines.append("<JOURNAL>"+doc.metadata["journal"]+"</JOURNAL>\n")
lines.append(u"<CURRENT_AUTHORLIST>")
for author in doc.metadata["authors"]:
bits=author.split(",")
rest=" ".join(bits[1:])
lines.append(u"<CURRENT_AUTHOR>")
## lines.append("<FIRSTNAME>%s</FIRSTNAME>" % rest)
rest=fixAmpersandsHTML(rest+u" ")
bits[0]=fixAmpersandsHTML(bits[0]+u" ").strip()
lines.append(rest+u"<CURRENT_SURNAME>%s</CURRENT_SURNAME>" % bits[0])
lines.append(u"</CURRENT_AUTHOR>")
lines.append(u"</CURRENT_AUTHORLIST>")
## lines.append("<SURNAMES>")
## for author in doc["authors"]:
##
## lines.append("</SURNAMES>")
lines.append(u"<YEAR>"+doc.metadata["year"]+"</YEAR>")
lines.append(u"</APPEARED>")
if doc["metadata"].has_key("revisionhistory"):
lines.append(u"<REVISIONHISTORY>"+doc.metadata["revisionhistory"]+u"</REVISIONHISTORY>\n")
lines.append(u"</METADATA>")
return lines
def dumpAbstract(doc):
"""
"""
lines=[]
lines.append(u"<ABSTRACT>")
if "content" in doc.abstract:
for element_id in doc.abstract["content"]:
s=doc.element_by_id[element_id]
if s["type"]=="s":
lines.append(dumpSentence(doc,s, True))
if s["type"]=="p":
lines.extend(dumpParagraph(doc,s, True))
lines.append(u"</ABSTRACT>")
return lines
def dumpSentence(doc,s,in_abstract=False):
"""
Fixes all the contents of the sentence, returns a string of proper XML
"""
def repl_func(m):
"""
Replaces <CIT ID=0 /> with <REF ID=cit0 REFID=ref0 />
"""
cit_id=m.group(1) # this is the actual unique identifier of a citation in the document
ref_id=doc.citation_by_id[cit_id]["ref_id"] # this is the id of the reference the cit cites
return u'<REF ID="'+safe_unicode(cit_id)+u'" REFID="'+safe_unicode(ref_id)+'" />'
in_abstract=False # hack because of sapienta difference vs SciXML
if in_abstract:
xmltag=u"A-S"
else:
xmltag=u"s"
s_id=re.match(r"\w+?(\d+)",s["id"],re.IGNORECASE)
assert(s_id)
res=u"<"+xmltag+u' sid="%s"' % str(int(s_id.group(1)))
# !TODO change this to export more markers, such as CFC and whatever
if "az" in s:
res+=u'AZ="%s"' % s["az"]
text=s["text"]
text=re.sub(r"<CIT ID=(.*?)\s?/>",repl_func, text)
res+=">"+text+"</"+xmltag+">"
return res
def dumpParagraph(doc,paragraph, in_abstract=False):
"""
"""
lines=[]
# fix for empty paragraphs: should it be here?
if len(paragraph["content"]) == 0:
return lines
if not in_abstract:
lines.append(u"<P>")
for element in paragraph["content"]:
element=doc.element_by_id[element]
if element["type"]=="s":
lines.append(dumpSentence(doc,element, in_abstract))
if not in_abstract:
lines.append(u"</P>")
return lines
def dumpSection(doc,section, header_depth):
"""
"""
header_depth+=1
lines=[]
lines.append(u'<DIV depth="'+str(header_depth)+'">')
lines.append(u"<HEADER>%s</HEADER>" % section["header"])
for element in section["content"]:
element=doc.element_by_id[element]
if element["type"]=="section":
lines.extend(dumpSection(doc,element, header_depth))
elif element["type"]=="p":
lines.extend(dumpParagraph(doc,element))
lines.append(u"</DIV>")
header_depth-=1
return lines
def dumpBody(doc):
"""
"""
lines=[]
lines.append(u"<BODY>")
for section in doc.allsections:
if section != doc.abstract:
lines.extend(dumpSection(doc,section,0))
lines.append(u"</BODY>")
return lines
def dumpRefAuthor(doc,author):
"""
"""
lines=[]
lines.append(u"<AUTHOR>")
if isinstance(author,basestring):
lines.append(author)
elif isinstance(author,dict):
lines.append(author["given"])
lines.append(u"<SURNAME>%s</SURNAME>" % author["family"])
lines.append(u"</AUTHOR>")
return lines
def dumpRefAuthors(doc,ref):
"""
"""
lines=[]
lines.append("<AUTHORLIST>")
for author in ref["authors"]:
lines.extend(dumpRefAuthor(doc,author))
lines.append("</AUTHORLIST>")
return lines
def dumpReference(doc,ref):
"""
<REFERENCE ID="cit1">
<AUTHORLIST>
<AUTHOR>L.<SURNAME>Que, Jr.</SURNAME></AUTHOR>
<AUTHOR>R. Y. N.<SURNAME>Ho</SURNAME></AUTHOR>
</AUTHORLIST>
<TITLE></TITLE>
<JOURNAL>Chem. Rev.<YEAR>1996</YEAR>962607-2624</JOURNAL></REFERENCE>
"""
lines=[]
lines.append('<REFERENCE ID="%s">' % ref["id"])
lines.extend(dumpRefAuthors(doc,ref))
lines.append("<TITLE>%s</TITLE>" % ref["title"])
lines.append("<YEAR>%s</YEAR>" % ref["year"])
if "journal" in ref:
lines.append("<JOURNAL>%s</JOURNAL>" % ref["journal"])
lines.append("</REFERENCE>")
return lines
def dumpReferences(doc):
"""
"""
lines=[]
lines.append("<REFERENCELIST>")
for ref in doc.data["references"]:
lines.extend(dumpReference(doc,ref))
lines.append("</REFERENCELIST>")
return lines
lines=[]
lines.append(u'<?xml version="1.0" encoding="UTF-8" standalone="no"?>')
## lines.append('<!DOCTYPE PAPER SYSTEM "paper-structure-annotation.dtd">')
lines.append(u"<PAPER>")
lines.extend(dumpMetadata(doc))
lines.extend(dumpAbstract(doc))
lines.extend(dumpBody(doc))
lines.extend(dumpReferences(doc))
lines.append(u"</PAPER>")
lines2=[]
for line in lines:
lines2.append(safe_unicode(line))
lines=lines2
## file_str = StringIO()
## file_str.writelines(lines)
## full_str=file_str.getvalue()
full_str=u"".join(lines)
## full_str=fixAmpersandsHTML(full_str)
f=codecs.open(filename,"w", encoding="utf-8",errors="ignore")
f.write(full_str)
f.close()
def saveSapientaXML(doc, filename):
"""
Exports a SciDocJSON to a file in Sapienta sort-of SciXML format
"""
header_depth = 0
def dumpMetadata(doc):
"""
"""
lines = []
lines.append('<mode2 hasDoc="yes" name="' +
doc.metadata["filename"].replace(".xml", "") +
'" version="597"/>')
lines.append("<METADATA>")
if "fileno" in doc["metadata"]:
lines.append("<FILENO>" + doc.metadata["fileno"] + "</FILENO>\n")
lines.append("<FILENAME>" + doc.metadata["filename"] + "</FILENAME>\n")
lines.append("<APPEARED>")
if "conference" in doc.metadata:
lines.append("<CONFERENCE>" + doc.metadata["conference"] +
"</CONFERENCE>\n")
if "journal" in doc.metadata:
lines.append("<JOURNAL>" + doc.metadata["journal"] +
"</JOURNAL>\n")
lines.append(u"<CURRENT_AUTHORLIST>")
for author in doc.metadata["authors"]:
bits = author.split(",")
rest = " ".join(bits[1:])
lines.append(u"<CURRENT_AUTHOR>")
## lines.append("<FIRSTNAME>%s</FIRSTNAME>" % rest)
rest = fixAmpersandsHTML(rest + u" ")
bits[0] = fixAmpersandsHTML(bits[0] + u" ").strip()
lines.append(rest +
u"<CURRENT_SURNAME>%s</CURRENT_SURNAME>" % bits[0])
lines.append(u"</CURRENT_AUTHOR>")
lines.append(u"</CURRENT_AUTHORLIST>")
## lines.append("<SURNAMES>")
## for author in doc["authors"]:
##
## lines.append("</SURNAMES>")
lines.append(u"<YEAR>" + doc.metadata["year"] + "</YEAR>")
lines.append(u"</APPEARED>")
if "revisionhistory" in doc["metadata"]:
lines.append(u"<REVISIONHISTORY>" +
doc.metadata["revisionhistory"] +
u"</REVISIONHISTORY>\n")
lines.append(u"</METADATA>")
return lines
def dumpAbstract(doc):
"""
"""
lines = []
lines.append(u"<ABSTRACT>")
if "content" in doc.abstract:
for element_id in doc.abstract["content"]:
s = doc.element_by_id[element_id]
if s["type"] == "s":
lines.append(dumpSentence(doc, s, True))
if s["type"] == "p":
lines.extend(dumpParagraph(doc, s, True))
lines.append(u"</ABSTRACT>")
return lines
def dumpSentence(doc, s, in_abstract=False):
"""
Fixes all the contents of the sentence, returns a string of proper XML
"""
def repl_func(m):
"""
Replaces <CIT ID=0 /> with <REF ID=cit0 REFID=ref0 />
"""
cit_id = m.group(
1
) # this is the actual unique identifier of a citation in the document
ref_id = doc.citation_by_id[cit_id][
"ref_id"] # this is the id of the reference the cit cites
return u'<REF ID="' + safe_unicode(
cit_id) + u'" REFID="' + safe_unicode(ref_id) + '" />'
in_abstract = False # hack because of sapienta difference vs SciXML
if in_abstract:
xmltag = u"A-S"
else:
xmltag = u"s"
s_id = re.match(r"\w+?(\d+)", s["id"], re.IGNORECASE)
assert (s_id)
res = u"<" + xmltag + u' sid="%s"' % str(int(s_id.group(1)))
# !TODO change this to export more markers, such as CFC and whatever
if "az" in s:
res += u'AZ="%s"' % s["az"]
text = s["text"]
text = re.sub(r"<CIT ID=(.*?)\s?/>", repl_func, text)
res += ">" + text + "</" + xmltag + ">"
return res
def dumpParagraph(doc, paragraph, in_abstract=False):
"""
"""
lines = []
# fix for empty paragraphs: should it be here?
if len(paragraph["content"]) == 0:
return lines
if not in_abstract:
lines.append(u"<P>")
for element in paragraph["content"]:
element = doc.element_by_id[element]
if element["type"] == "s":
lines.append(dumpSentence(doc, element, in_abstract))
if not in_abstract:
lines.append(u"</P>")
return lines
def dumpSection(doc, section, header_depth):
"""
"""
header_depth += 1
lines = []
lines.append(u'<DIV depth="' + str(header_depth) + '">')
lines.append(u"<HEADER>%s</HEADER>" % section["header"])
for element in section["content"]:
element = doc.element_by_id[element]
if element["type"] == "section":
lines.extend(dumpSection(doc, element, header_depth))
elif element["type"] == "p":
lines.extend(dumpParagraph(doc, element))
lines.append(u"</DIV>")
header_depth -= 1
return lines
def dumpBody(doc):
"""
"""
lines = []
lines.append(u"<BODY>")
for section in doc.allsections:
if section != doc.abstract:
lines.extend(dumpSection(doc, section, 0))
lines.append(u"</BODY>")
return lines
def dumpRefAuthor(doc, author):
"""
"""
lines = []
lines.append(u"<AUTHOR>")
if isinstance(author, six.string_types):
lines.append(author)
elif isinstance(author, dict):
lines.append(author["given"])
lines.append(u"<SURNAME>%s</SURNAME>" % author["family"])
lines.append(u"</AUTHOR>")
return lines
def dumpRefAuthors(doc, ref):
"""
"""
lines = []
lines.append("<AUTHORLIST>")
for author in ref["authors"]:
lines.extend(dumpRefAuthor(doc, author))
lines.append("</AUTHORLIST>")
return lines
def dumpReference(doc, ref):
"""
<REFERENCE ID="cit1">
<AUTHORLIST>
<AUTHOR>L.<SURNAME>Que, Jr.</SURNAME></AUTHOR>
<AUTHOR>R. Y. N.<SURNAME>Ho</SURNAME></AUTHOR>
</AUTHORLIST>
<TITLE></TITLE>
<JOURNAL>Chem. Rev.<YEAR>1996</YEAR>962607-2624</JOURNAL></REFERENCE>
"""
lines = []
lines.append('<REFERENCE ID="%s">' % ref["id"])
lines.extend(dumpRefAuthors(doc, ref))
lines.append("<TITLE>%s</TITLE>" % ref["title"])
lines.append("<YEAR>%s</YEAR>" % ref["year"])
if "journal" in ref:
lines.append("<JOURNAL>%s</JOURNAL>" % ref["journal"])
lines.append("</REFERENCE>")
return lines
def dumpReferences(doc):
"""
"""
lines = []
lines.append("<REFERENCELIST>")
for ref in doc.data["references"]:
lines.extend(dumpReference(doc, ref))
lines.append("</REFERENCELIST>")
return lines
lines = []
lines.append(u'<?xml version="1.0" encoding="UTF-8" standalone="no"?>')
## lines.append('<!DOCTYPE PAPER SYSTEM "paper-structure-annotation.dtd">')
lines.append(u"<PAPER>")
lines.extend(dumpMetadata(doc))
lines.extend(dumpAbstract(doc))
lines.extend(dumpBody(doc))
lines.extend(dumpReferences(doc))
lines.append(u"</PAPER>")
lines2 = []
for line in lines:
lines2.append(safe_unicode(line))
lines = lines2
## file_str = StringIO()
## file_str.writelines(lines)
## full_str=file_str.getvalue()
full_str = u"".join(lines)
## full_str=fixAmpersandsHTML(full_str)
f = codecs.open(filename, "w", encoding="utf-8", errors="ignore")
f.write(full_str)
f.close()