chore: initial commit

2026-01-26 11:32:09 +01:00
commit 3a31948f02
4 changed files with 384 additions and 0 deletions
--- a/converter.py
+++ b/converter.py
@@ -0,0 +1,170 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+import json
+
+from bs4 import BeautifulSoup
+
+
+def get_eid(elmt):
+    return elmt["eId"].split("/")[-1]
+
+
+def register_path(elmts, elmt):
+    parents = list(reversed(elmt.find_parents()))
+    in_body = False
+    
+    elmt = elmts
+    path = []
+    for parent in parents:
+        name = parent.name
+        if not in_body:
+            if name == "body":
+                in_body = True
+            continue
+        
+        if not name in elmt["children"]:
+            elmt["children"][name] = {}
+        
+        eid = get_eid(parent)
+        if not eid in elmt["children"][name]:
+            elmt["children"][name][eid] = get_meta(parent)
+        
+        path.append((name, eid))
+        elmt = elmt["children"][name][eid]
+    
+    return path
+
+
+def get_meta(elmt):
+    name = elmt.name
+    meta = {
+        "children": {}
+    }
+    num = elmt.find("num")
+    if num:
+        meta["num"] = get_text(num)
+    
+    title = elmt.find("heading")
+    if title:
+        meta["heading"] = next(title.stripped_strings)
+    
+    return meta
+
+
+def get_text(elmt):
+    children = list(filter(lambda e: e.name in (None, "b", "i"), elmt.children))
+    return "".join(c.get_text().strip() for c in children).strip()
+
+
+def parse_elmt(elmt):
+    name = elmt.name
+    
+    if name is None:
+        return elmt.get_text().strip()
+    
+    if name == "sup":
+        return {
+            "type": "sup",
+            "body": list(map(parse_elmt, elmt.children))
+        }
+    
+    if name == "sub":
+        return {
+            "type": "sub",
+            "body": list(map(parse_elmt, elmt.children))
+        }
+    
+    if name == "blockList":
+        intro = parse_elmt(elmt.find("listIntroduction"))
+        items = []
+        
+        for item in elmt.find_all("item", recursive=False):
+            num = get_text(item.find("num"))
+            item_body = parse_elmt(item)
+            items.append((num, item_body))
+        
+        return {
+            "type": "enum",
+            "intro": intro,
+            "items": items
+        }
+    
+    body = []
+    for child in elmt.children:
+        if child.name != "num":
+            body.append(parse_elmt(child))
+    
+    body = remove_empty(body)
+    return body
+
+
+def remove_empty(body):
+    body = list(filter(lambda p: not isinstance(p, str) or p.strip(), body))
+    if len(body) == 1:
+        return body[0]
+    return body
+
+
+def convert(in_path, out_path):
+    with open(in_path, "r") as f:
+        bs = BeautifulSoup(f.read(), "xml")
+    
+    elmts = {
+        "children": {
+            "article": {}
+        }
+    }
+    for t in bs.find_all("transitional"):
+        t.extract()
+    
+    for p in bs.find_all("proviso"):
+        p.extract()
+    
+    for article in bs.find_all("article"):
+        pars = []
+        for par in article.find_all("paragraph"):
+            num = par.find("num")
+            if num is None:
+                num = "1"
+            else:
+                num = get_text(num)
+            
+            text = par.find("content")
+            for note in text.find_all("authorialNote"):
+                note.extract()
+            
+            body = []
+            
+            for child in text.children:
+                body.append(parse_elmt(child))
+            
+            body = remove_empty(body)
+            
+            pars.append({
+                "num": num,
+                "content": body
+            })
+        
+        path = register_path(elmts, article)
+        eid = get_eid(article)
+        articles = elmts["children"]["article"]
+        if eid not in articles:
+            articles[eid] = []
+        articles[eid].append({
+            "num": get_text(article.num),
+            "path": path,
+            "children": pars
+        })
+    
+    with open(out_path, "w") as f:
+        json.dump(elmts, f)
+
+
+if __name__ == "__main__":
+    paths = [
+        ("./raw/RS-311.0-01072024-FR.xml", "./law/code_penal.json"),
+    ]
+    
+    for in_path, out_path in paths:
+        print(f"{in_path} -> {out_path}")
+        convert(in_path, out_path)