risp/src/overview/parser.rs

use std::collections::HashMap;

use serde::Deserialize;

fn deserialize_string_to_usize<'de, D>(deserializer: D) -> Result<usize, D::Error>
where
    D: serde::Deserializer<'de>,
{
    let s = String::deserialize(deserializer)?;
    s.parse().map_err(serde::de::Error::custom)
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct OgdSearchResult {
    ogd_document_results: OgdDocumentResults,
}

impl OgdSearchResult {
    fn has_next_page(&self) -> bool {
        todo!();
    }

    pub(crate) fn get_par(&self) -> Vec<String> {
        let mut ret = Vec::new();
        for doc_ref in &self.ogd_document_results.ogd_document_reference {
            for urls in &doc_ref
                .data
                .document_list
                .content_reference
                .urls
                .content_url
            {
                if urls.data_type == "Xml" {
                    ret.push(urls.url.clone());
                }
            }
        }
        ret
    }
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct OgdDocumentResults {
    hits: Hits,
    ogd_document_reference: Vec<OgdDocumentReference>,
}

#[derive(Deserialize)]
pub(crate) struct Hits {
    #[serde(
        rename = "@pageNumber",
        deserialize_with = "deserialize_string_to_usize"
    )]
    page_number: usize,
    #[serde(rename = "@pageSize", deserialize_with = "deserialize_string_to_usize")]
    page_size: usize,
    #[serde(rename = "#text", deserialize_with = "deserialize_string_to_usize")]
    text: usize,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct OgdDocumentReference {
    data: Data,
}

#[derive(Deserialize)]
pub(crate) struct Data {
    #[serde(rename = "Metadaten")]
    metadata: Metadata,
    #[serde(rename = "Dokumentliste")]
    document_list: DocumentList,
}

#[derive(Deserialize)]
pub(crate) struct Metadata {
    #[serde(rename = "Technisch")]
    technical: TechnicalMetadata,

    #[serde(rename = "Allgemein")]
    general: GeneralMetadata,

    #[serde(rename = "Bundesrecht")]
    fed: FedMetadata,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct TechnicalMetadata {
    #[serde(rename = "ID")]
    id: String,

    #[serde(rename = "Applikation")]
    application: String,

    organ: String,

    import_timestamp: ImportTimestamp,
}

#[derive(Deserialize)]
pub(crate) struct ImportTimestamp {
    #[serde(rename = "@xsi:nil")]
    xsi_nil: String, //TODO: switch to bool

    #[serde(rename = "@xmlns:xsi")]
    xmlns_xsi: String,
}

#[derive(Deserialize)]
pub(crate) struct GeneralMetadata {
    #[serde(rename = "Geaendert")]
    changed: String, //TODO: switch to YYYY-MM-DD string

    #[serde(rename = "DokumentUrl")]
    document_url: String,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct FedMetadata {
    #[serde(rename = "Kurztitel")]
    short_title: String,

    #[serde(rename = "Titel")]
    title: Option<String>,

    eli: String,

    br_kons: BrKons,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct BrKons {
    kundmachungsorgan: String,
    typ: String,
    #[serde(rename = "Dokumenttyp")]
    documenttype: String,
    artikel_paragraph_anlage: String,
    paragraphnummer: Option<String>,
    stammnorm_publikationsorgan: String,
    stammnorm_bgblnummer: String,
    inkrafttretensdatum: String, //TODO: switch to date
    #[serde(rename = "Indizes")]
    indices: HashMap<String, String>, //e.g. "item": "22/04 Sonstiges Zivilprozess, Außerstreitiges Verfahren"
    #[serde(rename = "Aenderung")]
    change: String,
    anmerkung: Option<String>,
    schlagworte: Option<String>,
    gesetzesnummer: String,
    alte_dokumentnummer: Option<String>,
    #[serde(rename = "GesamteRechtsvorschriftUrl")]
    full_url: String,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct DocumentList {
    content_reference: ContentReference,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct ContentReference {
    content_type: String,
    name: String,
    urls: ContentUrl,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct ContentUrl {
    content_url: Vec<ContentUrlItem>,
}

#[derive(Deserialize)]
#[serde(rename_all = "PascalCase")]
pub(crate) struct ContentUrlItem {
    data_type: String,
    url: String,
}

#[cfg(test)]
mod tests {
    use std::{fs::File, io::Read};

    use super::*;

    #[derive(Deserialize)]
    #[serde(rename_all = "PascalCase")]
    pub(crate) struct Wrapper {
        ogd_search_result: OgdSearchResult,
    }

    #[test]
    fn deserialize_teg_success() {
        let mut file = File::open("data/teg.json").unwrap();
        let mut json = String::new();
        file.read_to_string(&mut json).unwrap();

        let wrapper: serde_json::Result<Wrapper> = serde_json::from_str(&json);
        if wrapper.is_err() {
            let dbg = wrapper.as_ref().err().unwrap();
            println!("{dbg:#?}");
        }

        assert!(wrapper.is_ok());
    }

    #[test]
    fn deserialize_abgb_success() {
        let mut file = File::open("data/abgb.json").unwrap();
        let mut json = String::new();
        file.read_to_string(&mut json).unwrap();

        let wrapper: serde_json::Result<Wrapper> = serde_json::from_str(&json);
        if wrapper.is_err() {
            let dbg = wrapper.as_ref().err().unwrap();
            println!("{dbg:#?}");
        }

        assert!(wrapper.is_ok());
    }

    #[test]
    fn deserialize_urhg_success() {
        let mut file = File::open("data/urhg.json").unwrap();
        let mut json = String::new();
        file.read_to_string(&mut json).unwrap();

        let wrapper: serde_json::Result<Wrapper> = serde_json::from_str(&json);
        if wrapper.is_err() {
            let dbg = wrapper.as_ref().err().unwrap();
            println!("{dbg:#?}");
        }

        assert!(wrapper.is_ok());
    }
}