func (doc *Document) Authors() (authors []finc.Author) {
for _, ra := range doc.Author {
authors = append(authors, finc.Author{
FirstName: AuthorReplacer.Replace(span.UnescapeTrim(ra.Given)),
LastName: AuthorReplacer.Replace(span.UnescapeTrim(ra.Family)),
})
}
return authors
}
// CombinedTitle returns a longish title.
func (doc *Document) CombinedTitle() string {
if len(doc.Title) > 0 {
if len(doc.Subtitle) > 0 {
return span.UnescapeTrim(fmt.Sprintf("%s : %s", strings.Join(doc.Title, " "), strings.Join(doc.Subtitle, " ")))
}
return span.UnescapeTrim(strings.Join(doc.Title, " "))
}
if len(doc.Subtitle) > 0 {
return span.UnescapeTrim(strings.Join(doc.Subtitle, " "))
}
return ""
}
// ToIntermediateSchema converts a crossref document into IS.
func (doc *Document) ToIntermediateSchema() (*finc.IntermediateSchema, error) {
var err error
output := finc.NewIntermediateSchema()
output.Date, err = doc.Issued.Date()
if err != nil {
return output, err
}
output.RawDate = output.Date.Format("2006-01-02")
if doc.URL == "" {
return output, errNoURL
}
output.RecordID = doc.RecordID()
if len(output.RecordID) > span.KeyLengthLimit {
return output, span.Skip{Reason: fmt.Sprintf("ID_TOO_LONG %s", output.RecordID)}
}
if output.Date.After(Future) {
return output, span.Skip{Reason: fmt.Sprintf("TOO_FUTURISTIC %s", output.RecordID)}
}
if doc.Type == "journal-issue" {
return output, span.Skip{Reason: fmt.Sprintf("JOURNAL_ISSUE %s", output.RecordID)}
}
output.ArticleTitle = doc.CombinedTitle()
if len(output.ArticleTitle) == 0 {
return output, span.Skip{Reason: fmt.Sprintf("NO_ATITLE %s", output.RecordID)}
}
for _, title := range ArticleTitleBlocker {
if output.ArticleTitle == title {
return output, span.Skip{Reason: fmt.Sprintf("BLOCKED_ATITLE %s", output.RecordID)}
}
}
for _, p := range ArticleTitleCleanerPatterns {
output.ArticleTitle = p.ReplaceAllString(output.ArticleTitle, "")
}
// refs. #8428
if len(output.ArticleTitle) > 32000 {
return output, span.Skip{Reason: fmt.Sprintf("TOO_LONG_TITLE %s", output.RecordID)}
}
output.DOI = doc.DOI
output.Format = Formats.LookupDefault(doc.Type, DefaultFormat)
output.Genre = Genres.LookupDefault(doc.Type, "unknown")
output.ISSN = doc.ISSN
output.Issue = doc.Issue
output.Languages = []string{"eng"}
output.Publishers = append(output.Publishers, doc.Publisher)
output.RefType = RefTypes.LookupDefault(doc.Type, "GEN")
output.SourceID = SourceID
output.Subjects = doc.Subjects
output.Type = doc.Type
output.URL = append(output.URL, doc.URL)
output.Volume = doc.Volume
if len(doc.ContainerTitle) > 0 {
output.JournalTitle = span.UnescapeTrim(doc.ContainerTitle[0])
} else {
return output, span.Skip{Reason: fmt.Sprintf("NO_JTITLE %s", output.RecordID)}
}
if len(doc.Subtitle) > 0 {
output.ArticleSubtitle = span.UnescapeTrim(doc.Subtitle[0])
}
output.Authors = doc.Authors()
// TODO(miku): do we need a config for these things?
// Maybe a generic filter (in js?) that will gather exclusion rules?
// if len(output.Authors) == 0 {
// return output, span.Skip{Reason: fmt.Sprintf("NO_AUTHORS %s", output.RecordID)}
// }
pi := doc.PageInfo()
output.StartPage = fmt.Sprintf("%d", pi.StartPage)
output.EndPage = fmt.Sprintf("%d", pi.EndPage)
output.Pages = pi.RawMessage
output.PageCount = fmt.Sprintf("%d", pi.PageCount())
if doc.Publisher == "" {
output.MegaCollection = fmt.Sprintf("X-U (CrossRef)")
} else {
output.MegaCollection = fmt.Sprintf("%s (CrossRef)", doc.Publisher)
}
return output, nil
}
// ShortTitle returns the first main title only.
func (doc *Document) ShortTitle() (s string) {
if len(doc.Title) > 0 {
s = span.UnescapeTrim(doc.Title[0])
}
return
}