/** Check whether the source is in the correct format for this importer. */
@Override
public boolean isRecognizedFormat(InputStream stream) throws IOException {
// Our strategy is to look for the "PY <year>" line.
BufferedReader in = new BufferedReader(ImportFormatReader.getReaderDefaultEncoding(stream));
// Pattern pat1 = Pattern.compile("PY: \\d{4}");
Pattern pat1 = Pattern.compile("Record.*INSPEC.*");
// was PY \\\\d{4}? before
String str;
while ((str = in.readLine()) != null) {
// Inspec and IEEE seem to have these strange " - " between key and value
// str = str.replace(" - ", "");
// System.out.println(str);
if (pat1.matcher(str).find()) {
return true;
}
}
return false;
}
/** Check whether the source is in the correct format for this importer. */
@Override
public boolean isRecognizedFormat(InputStream stream) throws IOException {
BufferedReader in = new BufferedReader(ImportFormatReader.getReaderDefaultEncoding(stream));
String str;
int i = 0;
while (((str = in.readLine()) != null) && (i < 50)) {
/**
* The following line gives false positives for RIS files, so it should not be uncommented.
* The hypen is a characteristic of the RIS format.
*
* <p>str = str.replace(" - ", "")
*/
if (IsiImporter.ISI_PATTERN.matcher(str).find()) {
return true;
}
i++;
}
return false;
}
@Override
public void endElement(String uri, String localName, String qName) {
if ("PubmedArticle".equals(localName)) {
// bibitems.add( new Bibitem(null, makeBibtexString(), Globals.nextKey(),"-1" ) );
// check if year ="" then give medline date instead
if ("".equals(year)) {
if (!"".equals(MedlineDate)) {
// multi-year date format
// System.out.println(MedlineDate);
year = MedlineDate.substring(0, 4);
// Matcher m = Pattern.compile("\\b[0-9]{4}\\b").matcher(MedlineDate);
// if(m.matches())
// year = m.group();
}
}
// Build a string from the collected keywords:
StringBuilder sb = new StringBuilder();
for (Iterator<String> iterator = descriptors.iterator(); iterator.hasNext(); ) {
String s = iterator.next();
sb.append(s);
if (iterator.hasNext()) {
sb.append(MedlineHandler.KEYWORD_SEPARATOR);
}
}
String keywords = sb.toString();
BibEntry b =
new BibEntry(
IdGenerator.next(), // Globals.DEFAULT_BIBTEXENTRY_ID,
EntryTypes.getTypeOrDefault(
"article")); // id assumes an existing database so don't create one here
if (!"".equals(author)) {
b.setField(
"author",
MedlineHandler.htmlConverter.formatUnicode(
ImportFormatReader.expandAuthorInitials(author)));
// b.setField("author",Util.replaceSpecialCharacters(ImportFormatReader.expandAuthorInitials(author)));
author = "";
}
if (!"".equals(title)) {
b.setField("title", MedlineHandler.htmlConverter.formatUnicode(title));
}
// if (!title.equals("")) b.setField("title",Util.replaceSpecialCharacters(title));
if (!"".equals(journal)) {
b.setField("journal", journal);
}
if (!"".equals(year)) {
b.setField("year", year);
}
// PENDING [email protected] 2005-05-27 : added call to fixPageRange
if (!"".equals(page)) {
b.setField("pages", fixPageRange(page));
}
if (!"".equals(volume)) {
b.setField("volume", volume);
}
if (!"".equals(language)) {
b.setField("language", language);
}
if (!"".equals(pst)) {
b.setField("medline-pst", pst);
}
if (!"".equals(abstractText)) {
b.setField("abstract", abstractText.replaceAll("%", "\\\\%"));
}
if (!"".equals(keywords)) {
b.setField("keywords", keywords);
}
if (!"".equals(month)) {
b.setField("month", month);
}
// if (!url.equals("")) b.setField("url",url);
if (!"".equals(number)) {
b.setField("number", number);
}
if (!"".equals(doi)) {
b.setField("doi", doi);
b.setField("url", "http://dx.doi.org/" + doi);
}
if (!"".equals(pii)) {
b.setField("pii", pii);
}
if (!"".equals(pmc)) {
b.setField("pmc", pmc);
}
if (!"".equals(affiliation)) {
b.setField("institution", affiliation.replaceAll("#", "\\\\#"));
}
// PENDING [email protected] 2005-05-27 : added "pmid" bibtex field
// Older references do not have doi entries, but every
// medline entry has a unique pubmed ID (aka primary ID).
// Add a bibtex field for the pubmed ID for future use.
if (!"".equals(pubmedid)) {
b.setField("pmid", pubmedid);
}
bibitems.add(b);
abstractText = "";
author = "";
title = "";
journal = "";
keywords = "";
doi = "";
pii = "";
pmc = "";
year = "";
forename = "";
lastName = "";
suffix = "";
abstractText = "";
affiliation = "";
pubmedid = "";
majorTopic = "";
minorTopics = "";
month = "";
volume = "";
language = "";
pst = "";
lastname = "";
suffix = "";
initials = "";
number = "";
page = "";
String medlineID = "";
String url = "";
MedlineDate = "";
descriptors.clear();
} else if ("ArticleTitle".equals(localName)) {
inTitle = false;
} else if ("PubDate".equals(localName)) {
inPubDate = false;
} else if ("Year".equals(localName)) {
inYear = false;
} else if ("PMID".equals(localName)) {
inPubMedID = false;
} else if ("MedlineDate".equals(localName)) {
inMedlineDate = false;
} else if ("MedlineTA".equals(localName)) {
inJournal = false;
} // journal name
else if ("Month".equals(localName)) {
inMonth = false;
} else if ("Volume".equals(localName)) {
inVolume = false;
} else if ("Language".equals(localName)) {
inLanguage = false;
} else if ("PublicationStatus".equals(localName)) {
inPst = false;
} else if ("AuthorList".equals(localName)) {
author = join(authors.toArray(), " and ");
inAuthorList = false;
} else if ("Author".equals(localName)) {
// forename sometimes has initials with " " in middle: is pattern [A-Z] [A-Z]
// when above is the case replace it with initials
if ((forename.length() == 3) && (forename.charAt(1) == ' ')) {
forename = initials;
}
// Put together name with last name first, and enter suffix in between if present:
if (lastname.indexOf(" ") > 0) {
author = "{" + lastname + "}";
} else {
author = lastname;
}
if (!suffix.isEmpty()) {
author = author + ", " + suffix;
}
if (!forename.isEmpty()) {
author = author + ", " + forename;
}
// author = initials + " " + lastname;
authors.add(author);
inAuthor = false;
forename = "";
initials = "";
lastname = "";
suffix = "";
} else if ("DescriptorName".equals(localName)) {
inDescriptorName = false;
} else if ("QualifierName".equals(localName)) {
inQualifierName = false;
} else if ("MeshHeading".equals(localName)) {
inMeshHeader = false;
if ("".equals(minorTopics)) {
descriptors.add(majorTopic);
} else {
descriptors.add(majorTopic + ", " + minorTopics);
}
} else if ("LastName".equals(localName)) {
inLastName = false;
} else if ("Suffix".equals(localName)) {
inSuffix = false;
} else if ("ForeName".equals(localName) || "FirstName".equals(localName)) {
inForename = false;
} else if ("Issue".equals(localName)) {
inIssue = false;
} else if ("MedlinePgn".equals(localName)) {
inMedlinePgn = false;
} // pagenumber
else if ("URL".equals(localName)) {
inUrl = false;
} else if ("Initials".equals(localName)) {
// initials= '.' + initials + '.';
inInitials = false;
} else if ("AbstractText".equals(localName)) {
inAbstractText = false;
} else if ("Affiliation".equals(localName)) {
inAffiliation = false;
} else if ("ArticleId".equals(localName)) {
if (inDoi) {
inDoi = false;
} else if (inPii) {
inPii = false;
} else if (inPmc) {
inPmc = false;
}
}
}
/** Parse the entries in the source, and return a List of BibtexEntry objects. */
@Override
public List<BibtexEntry> importEntries(InputStream stream, OutputPrinter status)
throws IOException {
ArrayList<BibtexEntry> bibItems = new ArrayList<BibtexEntry>();
BufferedReader in = new BufferedReader(ImportFormatReader.getReaderDefaultEncoding(stream));
String line;
HashMap<String, String> hm = new HashMap<String, String>();
HashMap<String, StringBuffer> lines = new HashMap<String, StringBuffer>();
StringBuffer previousLine = null;
while ((line = in.readLine()) != null) {
if (line.isEmpty()) {
continue; // ignore empty lines, e.g. at file
}
// end
// entry delimiter -> item complete
if (line.equals("------")) {
String[] type = new String[2];
String[] pages = new String[2];
String country = null;
String address = null;
String titleST = null;
String titleTI = null;
Vector<String> comments = new Vector<String>();
// add item
for (Map.Entry<String, StringBuffer> entry : lines.entrySet()) {
if (entry.getKey().equals("AU")) {
hm.put("author", entry.getValue().toString());
} else if (entry.getKey().equals("TI")) {
titleTI = entry.getValue().toString();
} else if (entry.getKey().equals("ST")) {
titleST = entry.getValue().toString();
} else if (entry.getKey().equals("YP")) {
hm.put("year", entry.getValue().toString());
} else if (entry.getKey().equals("VL")) {
hm.put("volume", entry.getValue().toString());
} else if (entry.getKey().equals("NB")) {
hm.put("number", entry.getValue().toString());
} else if (entry.getKey().equals("PS")) {
pages[0] = entry.getValue().toString();
} else if (entry.getKey().equals("PE")) {
pages[1] = entry.getValue().toString();
} else if (entry.getKey().equals("KW")) {
hm.put("keywords", entry.getValue().toString());
} else if (entry.getKey().equals("RT")) {
type[0] = entry.getValue().toString();
} else if (entry.getKey().equals("SB")) {
comments.add("Subject: " + entry.getValue());
} else if (entry.getKey().equals("SA")) {
comments.add("Secondary Authors: " + entry.getValue());
} else if (entry.getKey().equals("NT")) {
hm.put("note", entry.getValue().toString());
} else if (entry.getKey().equals("PB")) {
hm.put("publisher", entry.getValue().toString());
} else if (entry.getKey().equals("TA")) {
comments.add("Tertiary Authors: " + entry.getValue());
} else if (entry.getKey().equals("TT")) {
comments.add("Tertiary Title: " + entry.getValue());
} else if (entry.getKey().equals("ED")) {
hm.put("edition", entry.getValue().toString());
} else if (entry.getKey().equals("TW")) {
type[1] = entry.getValue().toString();
} else if (entry.getKey().equals("QA")) {
comments.add("Quaternary Authors: " + entry.getValue());
} else if (entry.getKey().equals("QT")) {
comments.add("Quaternary Title: " + entry.getValue());
} else if (entry.getKey().equals("IS")) {
hm.put("isbn", entry.getValue().toString());
} else if (entry.getKey().equals("AB")) {
hm.put("abstract", entry.getValue().toString());
} else if (entry.getKey().equals("AD")) {
address = entry.getValue().toString();
} else if (entry.getKey().equals("LG")) {
hm.put("language", entry.getValue().toString());
} else if (entry.getKey().equals("CO")) {
country = entry.getValue().toString();
} else if (entry.getKey().equals("UR") || entry.getKey().equals("AT")) {
String s = entry.getValue().toString().trim();
hm.put(
s.startsWith("http://") || s.startsWith("ftp://") ? "url" : "pdf",
entry.getValue().toString());
} else if (entry.getKey().equals("C1")) {
comments.add("Custom1: " + entry.getValue());
} else if (entry.getKey().equals("C2")) {
comments.add("Custom2: " + entry.getValue());
} else if (entry.getKey().equals("C3")) {
comments.add("Custom3: " + entry.getValue());
} else if (entry.getKey().equals("C4")) {
comments.add("Custom4: " + entry.getValue());
} else if (entry.getKey().equals("C5")) {
comments.add("Custom5: " + entry.getValue());
} else if (entry.getKey().equals("C6")) {
comments.add("Custom6: " + entry.getValue());
} else if (entry.getKey().equals("DE")) {
hm.put("annote", entry.getValue().toString());
} else if (entry.getKey().equals("CA")) {
comments.add("Categories: " + entry.getValue());
} else if (entry.getKey().equals("TH")) {
comments.add("Short Title: " + entry.getValue());
} else if (entry.getKey().equals("SE")) {
hm.put("chapter", entry.getValue().toString());
// else if (entry.getKey().equals("AC"))
// hm.put("",entry.getValue().toString());
// else if (entry.getKey().equals("LP"))
// hm.put("",entry.getValue().toString());
}
}
String bibtexType = "misc";
// to find type, first check TW, then RT
for (int i = 1; i >= 0 && bibtexType.equals("misc"); --i) {
if (type[i] == null) {
continue;
}
type[i] = type[i].toLowerCase();
if (type[i].contains("article")) {
bibtexType = "article";
} else if (type[i].contains("journal")) {
bibtexType = "article";
} else if (type[i].contains("book section")) {
bibtexType = "inbook";
} else if (type[i].contains("book")) {
bibtexType = "book";
} else if (type[i].contains("conference")) {
bibtexType = "inproceedings";
} else if (type[i].contains("proceedings")) {
bibtexType = "inproceedings";
} else if (type[i].contains("report")) {
bibtexType = "techreport";
} else if (type[i].contains("thesis") && type[i].contains("master")) {
bibtexType = "mastersthesis";
} else if (type[i].contains("thesis")) {
bibtexType = "phdthesis";
}
}
// depending on bibtexType, decide where to place the titleRT and
// titleTI
if (bibtexType.equals("article")) {
if (titleST != null) {
hm.put("journal", titleST);
}
if (titleTI != null) {
hm.put("title", titleTI);
}
} else if (bibtexType.equals("inbook")) {
if (titleST != null) {
hm.put("booktitle", titleST);
}
if (titleTI != null) {
hm.put("title", titleTI);
}
} else {
if (titleST != null) {
hm.put("booktitle", titleST); // should not
}
// happen, I
// think
if (titleTI != null) {
hm.put("title", titleTI);
}
}
// concatenate pages
if (pages[0] != null || pages[1] != null) {
hm.put(
"pages",
(pages[0] != null ? pages[0] : "") + (pages[1] != null ? "--" + pages[1] : ""));
}
// concatenate address and country
if (address != null) {
hm.put("address", address + (country != null ? ", " + country : ""));
}
if (!comments.isEmpty()) { // set comment if present
StringBuilder s = new StringBuilder();
for (int i = 0; i < comments.size(); ++i) {
s.append(i > 0 ? "; " : "").append(comments.elementAt(i));
}
hm.put("comment", s.toString());
}
BibtexEntry b =
new BibtexEntry(DEFAULT_BIBTEXENTRY_ID, BibtexEntryTypes.getEntryType(bibtexType));
b.setField(hm);
bibItems.add(b);
hm.clear();
lines.clear();
previousLine = null;
continue;
}
// new key
if (line.startsWith("--") && line.length() >= 7 && line.substring(4, 7).equals("-- ")) {
lines.put(line.substring(2, 4), previousLine = new StringBuffer(line.substring(7)));
continue;
}
// continuation (folding) of previous line
if (previousLine == null) {
return null;
}
previousLine.append(line.trim());
}
return bibItems;
}
/** Parse the entries in the source, and return a List of BibEntry objects. */
@Override
public List<BibEntry> importEntries(InputStream stream, OutputPrinter status) throws IOException {
ArrayList<BibEntry> bibitems = new ArrayList<>();
StringBuilder sb = new StringBuilder();
String str;
try (BufferedReader in =
new BufferedReader(ImportFormatReader.getReaderDefaultEncoding(stream))) {
while ((str = in.readLine()) != null) {
if (str.length() < 2) {
continue;
}
if (str.indexOf("Record") == 0) {
sb.append("__::__").append(str);
} else {
sb.append("__NEWFIELD__").append(str);
}
}
}
String[] entries = sb.toString().split("__::__");
String type = "";
HashMap<String, String> h = new HashMap<>();
for (String entry : entries) {
if (entry.indexOf("Record") != 0) {
continue;
}
h.clear();
String[] fields = entry.split("__NEWFIELD__");
for (String s : fields) {
// System.out.println(fields[j]);
String f3 = s.substring(0, 2);
String frest = s.substring(5);
if ("TI".equals(f3)) {
h.put("title", frest);
} else if ("PY".equals(f3)) {
h.put("year", frest);
} else if ("AU".equals(f3)) {
h.put(
"author",
AuthorList.fixAuthor_lastNameFirst(frest.replace(",-", ", ").replace(";", " and ")));
} else if ("AB".equals(f3)) {
h.put("abstract", frest);
} else if ("ID".equals(f3)) {
h.put("keywords", frest);
} else if ("SO".equals(f3)) {
int m = frest.indexOf('.');
if (m >= 0) {
String jr = frest.substring(0, m);
h.put("journal", jr.replace("-", " "));
frest = frest.substring(m);
m = frest.indexOf(';');
if (m >= 5) {
String yr = frest.substring(m - 5, m);
h.put("year", yr);
frest = frest.substring(m);
m = frest.indexOf(':');
if (m >= 0) {
String pg = frest.substring(m + 1).trim();
h.put("pages", pg);
h.put("volume", frest.substring(1, m));
}
}
}
} else if ("RT".equals(f3)) {
frest = frest.trim();
if ("Journal-Paper".equals(frest)) {
type = "article";
} else if ("Conference-Paper".equals(frest)
|| "Conference-Paper; Journal-Paper".equals(frest)) {
type = "inproceedings";
} else {
type = frest.replace(" ", "");
}
}
}
BibEntry b =
new BibEntry(
DEFAULT_BIBTEXENTRY_ID,
EntryTypes.getTypeOrDefault(type)); // id assumes an existing database so don't
// create one here
b.setField(h);
bibitems.add(b);
}
return bibitems;
}
/** Parse the entries in the source, and return a List of BibEntry objects. */
@Override
public List<BibEntry> importEntries(InputStream stream, OutputPrinter status) throws IOException {
if (stream == null) {
throw new IOException("No stream given.");
}
ArrayList<BibEntry> bibitems = new ArrayList<>();
StringBuilder sb = new StringBuilder();
BufferedReader in = new BufferedReader(ImportFormatReader.getReaderDefaultEncoding(stream));
// Pattern fieldPattern = Pattern.compile("^AU |^TI |^SO |^DT |^C1 |^AB
// |^ID |^BP |^PY |^SE |^PY |^VL |^IS ");
String str;
while ((str = in.readLine()) != null) {
if (str.length() < 3) {
continue;
}
// begining of a new item
if ("PT ".equals(str.substring(0, 3))) {
sb.append("::").append(str);
} else {
String beg = str.substring(0, 3).trim();
// I could have used the fieldPattern regular expression instead
// however this seems to be
// quick and dirty and it works!
if (beg.length() == 2) {
sb.append(" ## "); // mark the begining of each field
sb.append(str);
} else {
sb.append("EOLEOL"); // mark the end of each line
sb.append(str.trim()); // remove the initial spaces
}
}
}
String[] entries = sb.toString().split("::");
HashMap<String, String> hm = new HashMap<>();
// skip the first entry as it is either empty or has document header
for (String entry : entries) {
String[] fields = entry.split(" ## ");
if (fields.length == 0) {
fields = entry.split("\n");
}
String Type = "";
String PT = "";
String pages = "";
hm.clear();
for (String field : fields) {
// empty field don't do anything
if (field.length() <= 2) {
continue;
}
String beg = field.substring(0, 2);
String value = field.substring(3);
if (value.startsWith(" - ")) {
value = value.substring(3);
}
value = value.trim();
if ("PT".equals(beg)) {
if (value.startsWith("J")) {
PT = "article";
} else {
PT = value;
}
Type = "article"; // make all of them PT?
} else if ("TY".equals(beg)) {
if ("JOUR".equals(value)) {
Type = "article";
} else if ("CONF".equals(value)) {
Type = "inproceedings";
}
} else if ("JO".equals(beg)) {
hm.put("booktitle", value);
} else if ("AU".equals(beg)) {
String author = IsiImporter.isiAuthorsConvert(value.replaceAll("EOLEOL", " and "));
// if there is already someone there then append with "and"
if (hm.get("author") != null) {
author = hm.get("author") + " and " + author;
}
hm.put("author", author);
} else if ("TI".equals(beg)) {
hm.put("title", value.replaceAll("EOLEOL", " "));
} else if ("SO".equals(beg) || "JA".equals(beg)) {
hm.put("journal", value.replaceAll("EOLEOL", " "));
} else if ("ID".equals(beg) || "KW".equals(beg)) {
value = value.replaceAll("EOLEOL", " ");
String existingKeywords = hm.get("keywords");
if ((existingKeywords == null) || existingKeywords.contains(value)) {
existingKeywords = value;
} else {
existingKeywords += ", " + value;
}
hm.put("keywords", existingKeywords);
} else if ("AB".equals(beg)) {
hm.put("abstract", value.replaceAll("EOLEOL", " "));
} else if ("BP".equals(beg) || "BR".equals(beg) || "SP".equals(beg)) {
pages = value;
} else if ("EP".equals(beg)) {
int detpos = value.indexOf(' ');
// tweak for IEEE Explore
if ((detpos != -1) && !value.substring(0, detpos).trim().isEmpty()) {
value = value.substring(0, detpos);
}
pages = pages + "--" + value;
} else if ("PS".equals(beg)) {
pages = IsiImporter.parsePages(value);
} else if ("AR".equals(beg)) {
pages = value;
} else if ("IS".equals(beg)) {
hm.put("number", value);
} else if ("PY".equals(beg)) {
hm.put("year", value);
} else if ("VL".equals(beg)) {
hm.put("volume", value);
} else if ("PU".equals(beg)) {
hm.put("publisher", value);
} else if ("DI".equals(beg)) {
hm.put("doi", value);
} else if ("PD".equals(beg)) {
String month = IsiImporter.parseMonth(value);
if (month != null) {
hm.put("month", month);
}
} else if ("DT".equals(beg)) {
Type = value;
if ("Review".equals(Type)) {
Type = "article"; // set "Review" in Note/Comment?
} else if (Type.startsWith("Article")
|| Type.startsWith("Journal")
|| "article".equals(PT)) {
Type = "article";
} else {
Type = "misc";
}
} else if ("CR".equals(beg)) {
hm.put("CitedReferences", value.replaceAll("EOLEOL", " ; ").trim());
} else {
// Preserve all other entries except
if ("ER".equals(beg) || "EF".equals(beg) || "VR".equals(beg) || "FN".equals(beg)) {
continue;
}
hm.put(beg.toLowerCase(), value);
}
}
if (!"".equals(pages)) {
hm.put("pages", pages);
}
// Skip empty entries
if (hm.isEmpty()) {
continue;
}
BibEntry b = new BibEntry(DEFAULT_BIBTEXENTRY_ID, EntryTypes.getTypeOrDefault(Type));
// id assumes an existing database so don't
// Remove empty fields:
List<Object> toRemove = new ArrayList<>();
for (Map.Entry<String, String> field : hm.entrySet()) {
String content = field.getValue();
if ((content == null) || content.trim().isEmpty()) {
toRemove.add(field.getKey());
}
}
for (Object aToRemove : toRemove) {
hm.remove(aToRemove);
}
// Polish entries
IsiImporter.processSubSup(hm);
IsiImporter.processCapitalization(hm);
b.setField(hm);
bibitems.add(b);
}
return bibitems;
}
