defter 0.4.0

//! The public profile's languages. The wording follows the English and German
//! locales of commonmeta.org (dragoman's `ui/src/locales`), so a person's page
//! reads the same on both.

/// A language the public profile is available in.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum Lang {
    En,
    De,
}

/// The words on the public profile.
pub struct Text {
    pub person: &'static str,
    pub given_name: &'static str,
    pub family_name: &'static str,
    pub additional_names: &'static str,
    pub websites: &'static str,
    pub country: &'static str,
    pub affiliations: &'static str,
    pub started: &'static str,
    pub ended: &'static str,
    pub filters: &'static str,
    pub facet_year: &'static str,
    pub facet_type: &'static str,
    pub open_all: &'static str,
    pub close_all: &'static str,
    pub clear_filters: &'static str,
    pub search: &'static str,
    pub search_placeholder: &'static str,
    pub copy_citation: &'static str,
    pub content: &'static str,
    pub published: &'static str,
    pub in_container: &'static str,
    pub license: &'static str,
    pub no_results: &'static str,
    pub no_works: &'static str,
    pub pagination: &'static str,
    pub previous_page: &'static str,
    pub next_page: &'static str,
    pub language: &'static str,
    pub apply: &'static str,
    pub featured_only: &'static str,
    pub clear: &'static str,
    pub export_format: &'static str,
    pub citation_style: &'static str,
    pub formatted_citation: &'static str,
    pub export: &'static str,
    pub copy: &'static str,
    pub copied: &'static str,
}

const EN: Text = Text {
    person: "Person",
    given_name: "Given name",
    family_name: "Family name",
    additional_names: "Also known as",
    websites: "Websites & social links",
    country: "Country",
    affiliations: "Affiliations",
    started: "Started",
    ended: "Ended at",
    filters: "Filters",
    facet_year: "Publication year",
    facet_type: "Content type",
    open_all: "Open all",
    close_all: "Close all",
    clear_filters: "Clear filters",
    search: "Search",
    search_placeholder: "Search these works",
    copy_citation: "Copy citation",
    content: "Content",
    published: "published",
    in_container: "in",
    license: "Distributed under the terms of the",
    no_results: "No results found.",
    no_works: "No works on this page.",
    pagination: "Pagination",
    previous_page: "Previous page",
    next_page: "Next page",
    language: "Language",
    apply: "Apply",
    featured_only: "Featured only",
    clear: "Clear",
    export_format: "Export format",
    citation_style: "Citation style",
    formatted_citation: "Formatted Citation",
    export: "Export",
    copy: "Copy",
    copied: "Copied",
};

const DE: Text = Text {
    person: "Person",
    given_name: "Vorname",
    family_name: "Familienname",
    additional_names: "Auch bekannt als",
    websites: "Webseiten & soziale Netzwerke",
    country: "Land",
    affiliations: "Zugehörigkeiten",
    started: "Beginn",
    ended: "Ende",
    filters: "Filter",
    facet_year: "Erscheinungsjahr",
    facet_type: "Inhaltstyp",
    open_all: "Alle öffnen",
    close_all: "Alle schließen",
    clear_filters: "Filter zurücksetzen",
    search: "Suche",
    search_placeholder: "Diese Werke durchsuchen",
    copy_citation: "Zitat kopieren",
    content: "Inhalt",
    published: "veröffentlicht",
    in_container: "in",
    license: "Veröffentlicht unter den Bedingungen der",
    no_results: "Keine Ergebnisse gefunden.",
    no_works: "Keine Werke auf dieser Seite.",
    pagination: "Seitennummerierung",
    previous_page: "Vorherige Seite",
    next_page: "Nächste Seite",
    language: "Sprache",
    apply: "Anwenden",
    featured_only: "Nur hervorgehobene",
    clear: "Löschen",
    export_format: "Exportformat",
    citation_style: "Zitierstil",
    formatted_citation: "Formatiertes Zitat",
    export: "Exportieren",
    copy: "Kopieren",
    copied: "Kopiert",
};

/// The words in `templates/base.html`.
pub struct LayoutText {
    pub lang: &'static str,
    pub toggle_dark: &'static str,
    pub sign_out: &'static str,
    pub tagline: &'static str,
    pub built_with: &'static str,
    pub by: &'static str,
    pub software: &'static str,
    pub mit_license: &'static str,
}

pub const LAYOUT_EN: LayoutText = LayoutText {
    lang: "en",
    toggle_dark: "Toggle dark mode",
    sign_out: "Sign out",
    tagline: "Self-hosted Zotero sync and file storage",
    built_with: "Built with",
    by: "by",
    software: "software provided under an",
    mit_license: "MIT license",
};

const LAYOUT_DE: LayoutText = LayoutText {
    lang: "de",
    toggle_dark: "Dunkelmodus umschalten",
    sign_out: "Abmelden",
    tagline: "Selbst gehostete Zotero-Synchronisierung und Dateiablage",
    built_with: "Erstellt mit",
    by: "von",
    software: "Software bereitgestellt unter einer",
    mit_license: "MIT-Lizenz",
};

const MONTHS_EN: [&str; 12] = [
    "January",
    "February",
    "March",
    "April",
    "May",
    "June",
    "July",
    "August",
    "September",
    "October",
    "November",
    "December",
];

const MONTHS_DE: [&str; 12] = [
    "Januar",
    "Februar",
    "März",
    "April",
    "Mai",
    "Juni",
    "Juli",
    "August",
    "September",
    "Oktober",
    "November",
    "Dezember",
];

/// German names of Zotero's item types. English ones are derived from the type.
const ITEM_TYPES_DE: &[(&str, &str)] = &[
    ("artwork", "Kunstwerk"),
    ("audioRecording", "Tonaufnahme"),
    ("bill", "Gesetzentwurf"),
    ("blogPost", "Blogbeitrag"),
    ("book", "Buch"),
    ("bookSection", "Buchkapitel"),
    ("case", "Gerichtsentscheidung"),
    ("computerProgram", "Software"),
    ("conferencePaper", "Konferenzbeitrag"),
    ("dataset", "Datensatz"),
    ("dictionaryEntry", "Wörterbucheintrag"),
    ("document", "Dokument"),
    ("email", "E-Mail"),
    ("encyclopediaArticle", "Lexikonartikel"),
    ("film", "Film"),
    ("forumPost", "Forumsbeitrag"),
    ("hearing", "Anhörung"),
    ("instantMessage", "Sofortnachricht"),
    ("interview", "Interview"),
    ("journalArticle", "Zeitschriftenartikel"),
    ("letter", "Brief"),
    ("magazineArticle", "Magazinartikel"),
    ("manuscript", "Manuskript"),
    ("map", "Karte"),
    ("newspaperArticle", "Zeitungsartikel"),
    ("patent", "Patent"),
    ("podcast", "Podcast"),
    ("preprint", "Preprint"),
    ("presentation", "Präsentation"),
    ("radioBroadcast", "Radiosendung"),
    ("report", "Bericht"),
    ("standard", "Norm"),
    ("statute", "Gesetz"),
    ("thesis", "Abschlussarbeit"),
    ("tvBroadcast", "Fernsehsendung"),
    ("videoRecording", "Videoaufnahme"),
    ("webpage", "Webseite"),
];

/// The languages of works most often seen, by primary subtag.
const LANGUAGES: &[(&str, &str, &str)] = &[
    ("ar", "Arabic", "Arabisch"),
    ("cs", "Czech", "Tschechisch"),
    ("da", "Danish", "Dänisch"),
    ("de", "German", "Deutsch"),
    ("el", "Greek", "Griechisch"),
    ("en", "English", "Englisch"),
    ("es", "Spanish", "Spanisch"),
    ("fi", "Finnish", "Finnisch"),
    ("fr", "French", "Französisch"),
    ("he", "Hebrew", "Hebräisch"),
    ("hu", "Hungarian", "Ungarisch"),
    ("it", "Italian", "Italienisch"),
    ("ja", "Japanese", "Japanisch"),
    ("ko", "Korean", "Koreanisch"),
    ("nb", "Norwegian", "Norwegisch"),
    ("nl", "Dutch", "Niederländisch"),
    ("no", "Norwegian", "Norwegisch"),
    ("pl", "Polish", "Polnisch"),
    ("pt", "Portuguese", "Portugiesisch"),
    ("ru", "Russian", "Russisch"),
    ("sv", "Swedish", "Schwedisch"),
    ("tr", "Turkish", "Türkisch"),
    ("zh", "Chinese", "Chinesisch"),
];

impl Lang {
    /// What the language switcher offers.
    pub const ALL: [Lang; 2] = [Lang::En, Lang::De];

    pub fn parse(code: &str) -> Option<Lang> {
        match code.to_ascii_lowercase().as_str() {
            "en" => Some(Lang::En),
            "de" => Some(Lang::De),
            _ => None,
        }
    }

    /// The first language in an `Accept-Language` header that is available,
    /// else English. Browsers list their languages by preference.
    pub fn from_accept_language(header: &str) -> Lang {
        header
            .split(',')
            .filter_map(|tag| {
                let tag = tag.split(';').next()?.trim();
                Lang::parse(tag.split('-').next()?)
            })
            .next()
            .unwrap_or(Lang::En)
    }

    pub fn code(&self) -> &'static str {
        match self {
            Lang::En => "en",
            Lang::De => "de",
        }
    }

    /// The language's name, in that language.
    pub fn name(&self) -> &'static str {
        match self {
            Lang::En => "English",
            Lang::De => "Deutsch",
        }
    }

    /// The locale formatted citations are written in.
    pub fn citation_locale(self) -> &'static str {
        match self {
            Lang::En => "en-US",
            Lang::De => "de-DE",
        }
    }

    pub fn text(self) -> &'static Text {
        match self {
            Lang::En => &EN,
            Lang::De => &DE,
        }
    }

    pub fn layout(self) -> &'static LayoutText {
        match self {
            Lang::En => &LAYOUT_EN,
            Lang::De => &LAYOUT_DE,
        }
    }

    /// `1043` → "1,043" or "1.043".
    pub fn count(&self, n: usize) -> String {
        let separator = match self {
            Lang::En => ',',
            Lang::De => '.',
        };
        let digits = n.to_string();
        let mut out = String::with_capacity(digits.len() + digits.len() / 3);
        for (i, digit) in digits.chars().enumerate() {
            if i > 0 && (digits.len() - i).is_multiple_of(3) {
                out.push(separator);
            }
            out.push(digit);
        }
        out
    }

    /// "1 work", "1,043 works".
    pub fn works(self, n: usize) -> String {
        let noun = match (self, n) {
            (Lang::En, 1) => "work",
            (Lang::En, _) => "works",
            (_, 1) => "Werk",
            (_, _) => "Werke",
        };
        format!("{} {noun}", self.count(n))
    }

    /// "page 1 of 105".
    pub fn page_of(self, page: usize, pages: usize) -> String {
        let (page, pages) = (self.count(page), self.count(pages));
        match self {
            Lang::En => format!("page {page} of {pages}"),
            Lang::De => format!("Seite {page} von {pages}"),
        }
    }

    /// A partial date for people: `2015` → "2015", `2015-08` → "August 2015",
    /// `2021-08-01` → "August 1, 2021" or "1. August 2021". Anything else, such
    /// as Zotero's free-form dates, stays as it is.
    pub fn date(self, date: &str) -> String {
        let parts: Vec<&str> = date.trim().split('-').collect();
        let months = match self {
            Lang::En => &MONTHS_EN,
            Lang::De => &MONTHS_DE,
        };
        let month = parts
            .get(1)
            .and_then(|m| m.parse::<usize>().ok())
            .and_then(|m| months.get(m.wrapping_sub(1)));
        let day = parts.get(2).and_then(|d| d.parse::<u32>().ok());
        match (self, parts[0], month, day) {
            (Lang::En, year, Some(month), Some(day)) => format!("{month} {day}, {year}"),
            (_, year, Some(month), Some(day)) => format!("{day}. {month} {year}"),
            (_, year, Some(month), None) => format!("{month} {year}"),
            (_, year, _, _) => year.to_string(),
        }
    }

    pub fn date_range(self, start: &str, end: &str) -> String {
        let text = self.text();
        match (start.trim().is_empty(), end.trim().is_empty()) {
            (false, false) => format!("{} – {}", self.date(start), self.date(end)),
            (false, true) => format!("{} {}", text.started, self.date(start)),
            (true, false) => format!("{} {}", text.ended, self.date(end)),
            (true, true) => String::new(),
        }
    }

    /// `journalArticle` → "Journal article" or "Zeitschriftenartikel".
    pub fn item_type(self, item_type: &str) -> String {
        if self != Lang::En
            && let Some((_, label)) = ITEM_TYPES_DE.iter().find(|(t, _)| *t == item_type)
        {
            return label.to_string();
        }
        let mut label = String::new();
        for (i, c) in item_type.chars().enumerate() {
            if i == 0 {
                label.extend(c.to_uppercase());
            } else if c.is_uppercase() {
                label.push(' ');
                label.extend(c.to_lowercase());
            } else {
                label.push(c);
            }
        }
        label
    }

    /// The name of a work's language: `en-US` → "English". Zotero's field is
    /// free text, so anything else stays as it is.
    pub fn language(self, language: &str) -> String {
        let language = language.trim();
        let primary = language
            .split('-')
            .next()
            .unwrap_or("")
            .to_ascii_lowercase();
        match LANGUAGES.iter().find(|(code, _, _)| *code == primary) {
            Some((_, en, _)) if self == Lang::En => en.to_string(),
            Some((_, _, de)) => de.to_string(),
            None => language.to_string(),
        }
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn picks_a_language_from_the_browser() {
        let accept = Lang::from_accept_language;
        assert_eq!(accept("de-DE,de;q=0.9,en;q=0.8"), Lang::De);
        assert_eq!(accept("de-AT,de;q=0.9"), Lang::De);
        assert_eq!(accept("fr-FR,fr;q=0.9,de;q=0.8,en;q=0.7"), Lang::De);
        assert_eq!(accept("en-GB,de;q=0.5"), Lang::En);
        assert_eq!(accept("fr"), Lang::En);
        assert_eq!(accept(""), Lang::En);
    }

    #[test]
    fn formats_dates_and_ranges() {
        assert_eq!(Lang::En.date("2015"), "2015");
        assert_eq!(Lang::En.date("2015-08"), "August 2015");
        assert_eq!(Lang::En.date("2021-08-01"), "August 1, 2021");
        assert_eq!(Lang::De.date("2021-08-01"), "1. August 2021");
        assert_eq!(Lang::De.date("2015-03"), "März 2015");
        assert_eq!(Lang::De.date("2021-01-05"), "5. Januar 2021");
        assert_eq!(Lang::En.date("May 2021"), "May 2021");
        assert_eq!(
            Lang::En.date_range("2021-08-01", ""),
            "Started August 1, 2021"
        );
        assert_eq!(
            Lang::De.date_range("2021-08-01", ""),
            "Beginn 1. August 2021"
        );
        assert_eq!(
            Lang::En.date_range("2015-08", "2021-07-31"),
            "August 2015 – July 31, 2021"
        );
        assert_eq!(Lang::En.date_range("", ""), "");
    }

    #[test]
    fn counts_in_the_page_language() {
        assert_eq!(Lang::En.count(1043), "1,043");
        assert_eq!(Lang::De.count(1043), "1.043");
        assert_eq!(Lang::En.count(999), "999");
        assert_eq!(Lang::En.works(1), "1 work");
        assert_eq!(Lang::De.works(1043), "1.043 Werke");
        assert_eq!(Lang::De.page_of(1, 105), "Seite 1 von 105");
    }

    #[test]
    fn names_types_and_languages() {
        assert_eq!(Lang::En.item_type("blogPost"), "Blog post");
        assert_eq!(Lang::De.item_type("blogPost"), "Blogbeitrag");
        assert_eq!(Lang::De.item_type("unknownType"), "Unknown type");
        assert_eq!(Lang::En.language("en-US"), "English");
        assert_eq!(Lang::De.language("en"), "Englisch");
        assert_eq!(Lang::En.language("Klingon"), "Klingon");
    }
}