parse full urhg

src/main.rs

@@ -44,5 +44,6 @@ fn main() {
     env_logger::init();
     let builder = LawBuilder::new("UrhG");
 
+    println!("{:#?}", builder);
     println!("{:#?}", builder.toc());
 }

src/overview/mod.rs

@@ -17,14 +17,14 @@ fn current_date() -> String {
 ///
 /// # Errors
 /// Fails if `ureq` can't create a connection, probably because there's no internet connection? (Or RIS is not online.)
-fn fetch_page(overview_id: usize) -> Result<String, Error> {
+fn fetch_page(overview_id: usize, page: usize) -> Result<String, Error> {
     Ok(
         ureq::post("https://data.bka.gv.at/ris/api/v2.6/Bundesrecht")
             .send_form(&[
                 ("Applikation", "BrKons"),
                 ("Gesetzesnummer", &format!("{}", overview_id)),
                 ("DokumenteProSeite", "OneHundred"),
-                ("Seitennummer", &format!("{}", 1)),
+                ("Seitennummer", &format!("{}", page)),
                 ("Fassung.FassungVom", &current_date()),
             ])?
             .into_string()?,
@@ -38,15 +38,26 @@ pub(crate) struct Wrapper {
 }
 
 pub(crate) fn parse(overview_id: usize, builder: &mut LawBuilder) -> Result<(), Error> {
-    let json = fetch_page(overview_id)?;
+    let mut page = 1;
+    loop {
+        let json = fetch_page(overview_id, page)?;
 
-    let wrapper: Wrapper = serde_json::from_str(&json)?;
+        let wrapper: Wrapper = serde_json::from_str(&json)?;
 
-    for par in wrapper.ogd_search_result.get_par().into_iter().skip(1) {
-        // skip bc. first one is
-        // always not relevant for
-        // me :-)
-        crate::par::parse(&par, builder).unwrap();
+        for par in wrapper.ogd_search_result.get_par().into_iter().skip(1) {
+            // skip bc. first one is
+            // always not relevant for
+            // me :-)
+            if !crate::par::parse(&par, builder).unwrap() {
+                break;
+            }
+        }
+
+        page += 1;
+
+        if !wrapper.ogd_search_result.has_next_page() {
+            break;
+        }
     }
 
     Ok(())

src/overview/parser.rs

@@ -17,8 +17,15 @@ pub(crate) struct OgdSearchResult {
 }
 
 impl OgdSearchResult {
-    fn has_next_page(&self) -> bool {
-        todo!();
+    pub(crate) fn has_next_page(&self) -> bool {
+        let hits = &self.ogd_document_results.hits;
+        let curr_page_number = hits.page_number;
+        let page_size = hits.page_size;
+        let elements = hits.text;
+
+        let parsed_so_far = curr_page_number * page_size;
+
+        elements > parsed_so_far
     }
 
     pub(crate) fn get_par(&self) -> Vec<String> {

src/par/mod.rs

@@ -8,7 +8,7 @@ fn fetch_page(url: &str) -> Result<String, Error> {
     Ok(ureq::get(url).call()?.into_string()?)
 }
 
-pub(crate) fn parse(url: &str, builder: &mut LawBuilder) -> Result<(), Error> {
+pub(crate) fn parse(url: &str, builder: &mut LawBuilder) -> Result<bool, Error> {
     info!("Parsing {url}");
     let xml = fetch_page(url)?;
     let xml = xml.replace("<gdash />", "-"); // used e.g. in §11 Abs. 3 UrhG
@@ -26,9 +26,14 @@ pub(crate) fn parse(url: &str, builder: &mut LawBuilder) -> Result<(), Error> {
     ); // 1. Verwertungsrechte. before § 14
 
     let xml = xml.replace("<i>.</i>", "."); // e.g. § 37d Abs. 4 (last point)...
+
+    // Artikel 18 UrhG
+    let xml = xml.replace("<n><i>", "");
+    let xml = xml.replace("</i></n>", "");
+
     debug!("{xml}");
 
-    let risdok = Risdok::from_str(&xml, builder)?;
+    let continue_parsing = Risdok::from_str(&xml, builder)?;
 
-    Ok(())
+    Ok(continue_parsing)
 }

src/par/parser.rs

@@ -9,32 +9,33 @@ use crate::{
 #[derive(Debug, PartialEq)]
 pub(crate) struct Risdok {
     metadaten: Metadaten,
-    nutzdaten: Nutzdaten,
     layoutdaten: Layoutdaten,
 }
 
 impl Risdok {
-    pub(crate) fn parse(n: Node, builder: &mut LawBuilder) -> Self {
+    pub(crate) fn parse(n: Node, builder: &mut LawBuilder) -> bool {
         assert!(n.tag_name().name() == "risdok");
 
         let mut c = n.children();
 
-        let ret = Self {
-            metadaten: Metadaten::parse(c.next().unwrap()),
-            nutzdaten: Nutzdaten::parse(c.next().unwrap(), builder),
-            layoutdaten: Layoutdaten::parse(c.next().unwrap()),
-        };
+        let metadaten = Metadaten::parse(c.next().unwrap());
+        let nutzdaten = Nutzdaten::parse(c.next().unwrap(), builder);
+        if !nutzdaten {
+            return false;
+        }
+        let layoutdaten = Layoutdaten::parse(c.next().unwrap());
 
         assert_eq!(c.next(), None);
 
-        ret
+        true
     }
 
-    pub(crate) fn from_str(xml: &str, builder: &mut LawBuilder) -> Result<Self, Error> {
+    pub(crate) fn from_str(xml: &str, builder: &mut LawBuilder) -> Result<bool, Error> {
         let doc = roxmltree::Document::parse(&xml)?;
         let root = doc.root();
         assert_eq!(root.children().into_iter().count(), 1);
-        Ok(Self::parse(root.children().next().unwrap(), builder))
+        let continue_parsing = Self::parse(root.children().next().unwrap(), builder);
+        Ok(continue_parsing)
     }
 }
 
@@ -53,23 +54,23 @@ impl Metadaten {
 #[derive(Debug, PartialEq)]
 pub(crate) struct Nutzdaten {}
 impl Nutzdaten {
-    pub(crate) fn parse(n: Node, builder: &mut LawBuilder) -> Self {
+    pub(crate) fn parse(n: Node, builder: &mut LawBuilder) -> bool {
         assert!(n.tag_name().name() == "nutzdaten");
 
         let mut c = n.children();
 
-        Abschnitt::parse(c.next().unwrap(), builder);
+        let ret = Abschnitt::parse(c.next().unwrap(), builder);
 
         assert_eq!(c.next(), None);
 
-        Self {}
+        ret
     }
 }
 
 #[derive(Debug, PartialEq)]
 pub(crate) struct Abschnitt;
 impl Abschnitt {
-    pub(crate) fn parse(n: Node, builder: &mut LawBuilder) {
+    pub(crate) fn parse(n: Node, builder: &mut LawBuilder) -> bool {
         assert!(n.tag_name().name() == "abschnitt");
 
         let mut c = n.children().peekable();
@@ -102,6 +103,9 @@ impl Abschnitt {
                 Some(child) => {
                     if Ueberschrift::test(&child, "g1") {
                         let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "g1");
+                        if ueberschrift.content.trim().starts_with("Artikel") {
+                            return false;
+                        }
                         builder.new_header(&ueberschrift.content);
                     } else if Ueberschrift::test(&child, "g2") {
                         let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "g2");
@@ -230,6 +234,8 @@ impl Abschnitt {
         }
 
         assert_eq!(c.next(), None);
+
+        true
     }
 }