more cleaning

src/paragraph/parser/abschnitt.rs

@@ -8,15 +8,12 @@ use crate::paragraph::parser::{
     Absatz, AbsatzAbs, Content, Fzinhalt, Kzinhalt, Liste, Table, Ueberschrift,
 };
 
-#[derive(Debug, PartialEq)]
+#[derive(Debug, PartialEq, Default)]
-#[derive(Default)]
 pub(crate) struct Abschnitt {
     metadata: HashMap<String, String>,
     pub(crate) cont: bool,
 }
 
-
-
 impl Abschnitt {
     pub(crate) fn parse(n: Node, builder: &mut LawBuilder) -> Abschnitt {
         assert!(n.tag_name().name() == "abschnitt");
@@ -26,53 +23,12 @@ impl Abschnitt {
 
         Self::skip_static_fields(&mut c);
 
-        ret.handle_headers(&mut c, builder);
+        ret.handle_metadata(&mut c, builder);
 
-        while let Some(child) = c.peek() {
+        if !ret.handle_headers(&mut c, builder) {
-            // Schiffahrtsgesetz: stop @ anlagen (for now)
+            return ret;
-            if Ueberschrift::test(child, "anlage") {
-                return ret;
-            }
-            if Ueberschrift::test(child, "g1") {
-                let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "g1");
-                if ueberschrift.content.trim().starts_with("Artikel") {
-                    return ret;
-                }
-                builder.new_header(&ueberschrift.content);
-            } else if Ueberschrift::test(child, "g2") {
-                let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "g2");
-                builder.new_desc(&ueberschrift.content);
-            } else if Ueberschrift::test(child, "g1min") {
-                let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "g1min");
-                builder.new_header(&ueberschrift.content);
-            } else if Ueberschrift::test(child, "art") {
-                let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "art");
-                if ueberschrift.content.trim().starts_with("Artikel") {
-                    return ret;
-                }
-            } else {
-                break;
-            }
         }
 
-        if let Some(child) = c.peek() {
-            if Ueberschrift::test(child, "para") {
-                builder
-                    .new_next_para_header(&Ueberschrift::parse(c.next().unwrap(), "para").content);
-            }
-        }
-        // e.g. § 405 abgb has two para (of diseased paragraph)
-        if let Some(child) = c.peek() {
-            if Ueberschrift::test(child, "para") {
-                builder
-                    .new_next_para_header(&Ueberschrift::parse(c.next().unwrap(), "para").content);
-            }
-        }
-
-        // We have 2 tasks
-        // 1) Get paragraph id
-        // 2) Get content
-
         let mut absatze = Vec::new();
         let absatz = AbsatzAbs::parse(c.next().expect("We need at least one 'Absatz'"));
         let par_id = absatz
@@ -194,7 +150,7 @@ impl Abschnitt {
 
     // There are paragraph-specific meta-data at the top of each xml file. We parse those. When we
     // encounter the title "Text" the real content starts, we stop parsing meta data.
-    fn handle_headers(&mut self, c: &mut Peekable<Children>, builder: &mut LawBuilder) {
+    fn handle_metadata(&mut self, c: &mut Peekable<Children>, builder: &mut LawBuilder) {
         loop {
             let key = Ueberschrift::parse(c.next().unwrap(), "titel").content;
 
@@ -230,4 +186,47 @@ impl Abschnitt {
         Fzinhalt::parse(node.next().unwrap()); // "www.ris.bka.gv.at" and "Seite X von Y"
         Fzinhalt::parse(node.next().unwrap()); // "www.ris.bka.gv.at" and "Seite X von Y"
     }
+
+    // we have optionally headers. Such as "Einleitung", "Von den bürgerlichen Gesetzen üerhaupt,"
+    // etc. If we have headers which indicate that we are done and we want to stop parsing
+    // ("anlage" + "Artikel" we indicate this wish by returning false.
+    fn handle_headers(&self, c: &mut Peekable<Children>, builder: &mut LawBuilder) -> bool {
+        while let Some(child) = c.peek() {
+            // Schiffahrtsgesetz: stop @ anlagen (for now)
+            if Ueberschrift::test(child, "anlage") {
+                return false;
+            }
+            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");
+                builder.new_desc(&ueberschrift.content);
+            } else if Ueberschrift::test(child, "g1min") {
+                let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "g1min");
+                builder.new_header(&ueberschrift.content);
+            } else if Ueberschrift::test(child, "art") {
+                let ueberschrift = Ueberschrift::parse(c.next().unwrap(), "art");
+                if ueberschrift.content.trim().starts_with("Artikel") {
+                    return false;
+                }
+            } else {
+                break;
+            }
+        }
+
+        // while (not if) because we can have two subsequent paraheaders (e.g. § 405 abgb)
+        while let Some(child) = c.peek() {
+            if Ueberschrift::test(child, "para") {
+                builder
+                    .new_next_para_header(&Ueberschrift::parse(c.next().unwrap(), "para").content);
+                continue;
+            }
+            break;
+        }
+        true
+    }
 }