OneGov · BreathingFlesh · Sep 23, 2025 · Sep 16, 2025 · Sep 16, 2025 · Sep 16, 2025
diff --git a/src/onegov/landsgemeinde/forms/agenda.py b/src/onegov/landsgemeinde/forms/agenda.py
@@ -1,8 +1,20 @@
 from __future__ import annotations
-from datetime import datetime
 
+import re
+import zipfile
+import os
+import base64
+import gzip
 import pytz
 
+from bs4 import BeautifulSoup
+from datetime import datetime
+from markupsafe import Markup
+from io import BytesIO
+from pathlib import Path
+from tempfile import TemporaryDirectory
+
+from onegov.form import Form
 from onegov.form.fields import TagsField
 from onegov.form.fields import TimeField
 from onegov.form.fields import UploadField
@@ -27,11 +39,11 @@
 from wtforms.validators import Optional
 from wtforms.validators import ValidationError
 
-
 from typing import Any
 from typing import TYPE_CHECKING
 if TYPE_CHECKING:
     from onegov.landsgemeinde.request import LandsgemeindeRequest
+    from onegov.landsgemeinde.collections import AgendaItemCollection
 
 
 class AgendaItemForm(NamedFileForm):
@@ -202,3 +214,179 @@ def populate_obj(self, obj: AgendaItem) -> None:  # type:ignore[override]
             tz = pytz.timezone('Europe/Zurich')
             now = datetime.now(tz=tz).time()
             obj.start_time = now
+
+
+class AgendaItemUploadForm(Form):
+
+    request: LandsgemeindeRequest
+
+    agenda_item_zip = UploadField(
+        label=_('Agenda Item ZIP'),
+        fieldset=_('Import'),
+        validators=[
+            WhitelistedMimeType({'application/zip'}),
+            FileSizeLimit(100 * 1024 * 1024)
+        ]
+    )
+
+    def get_html_dir(self,
+                     temp: TemporaryDirectory[str],
+                     field: UploadField) -> str | None:
+
+        temp_path = Path(temp.name)
+        zip_content = None
+
+        if isinstance(field.data, dict) and 'data' in field.data:
+            encoded_data = field.data['data']
+            decoded_data = base64.b64decode(encoded_data)
+
+            if decoded_data[:2] == b'\x1f\x8b':
+                decompressed_data = gzip.decompress(decoded_data)
+                zip_content = BytesIO(decompressed_data)
+            else:
+                zip_content = BytesIO(decoded_data)
+
+        with zipfile.ZipFile(
+            zip_content, 'r') as zip_ref:  # type:ignore
+            zip_ref.extractall(temp_path)
+
+        html_dir = None
+        for root, dirs, files in os.walk(temp_path):
+            if 'html' in dirs:
+                html_dir = os.path.join(root, 'html')
+                break
+
+        return html_dir
+
+    def import_agenda_item(
+            self, collection: AgendaItemCollection) -> AgendaItem:
+
+        temp = TemporaryDirectory()
+        html_dir = self.get_html_dir(temp, self.agenda_item_zip)
+        html_path = Path(html_dir)  # type:ignore
+        html_files = sorted(
+            [f for f in html_path.glob('*.html')
+             if f.name != 'combined_clean.html'],
+            key=lambda f: [int(s) if s.isdigit() else -1
+                           for s in re.findall(r'\d+|\D+', f.name)]
+        )
+
+        combined_html = BeautifulSoup(
+            '<html><head><meta charset="utf-8"></head><body></body></html>',
+            'html.parser')
+        title = ''
+
+        for file_path in html_files:
+            with open(file_path, encoding='utf-8') as f:
+                soup = BeautifulSoup(f.read(), 'html.parser')
+
+            all_paragraphs = []
+
+            if soup.body is None:
+                continue
+            textframes = soup.body.find_all('div', recursive=False)
+            for textframe in textframes[1:] if textframes else []:
+                for p_tag in textframe.find_all('p'):
+                    p_class = ' '.join(p_tag.get('class', []))
+                    spans_text = []
+                    parent = p_tag.find_parent()
+
+                    parent_inline_style = parent.get('style', ''
+                                                     ) if parent else ''
+                    if not parent_inline_style:
+                        p_class = 'table'
+
+                    regex = re.compile(r'^_idTextSpan\d+')
+                    for span in p_tag.find_all('span', id=regex):
+                        text = span.get_text().strip()
+                        if text:
+                            spans_text.append(text)
+
+                    if spans_text:
+                        all_paragraphs.append({
+                            'class': p_class,
+                            'text': ' '.join(spans_text),
+                            'is_list_item': 'Aufz-hlung' in p_class
+                        })
+
+            i = 0
+            while i < len(all_paragraphs):
+                para = all_paragraphs[i]
+                if combined_html.body is None:
+                    break
+
+                # Check if this is the start of a list
+                if para['is_list_item']:
+                    ul_element = combined_html.new_tag('ul')
+
+                    # Add this and all consecutive list items to the <ul>
+                    while (i < len(all_paragraphs)
+                           and all_paragraphs[i]['is_list_item']):
+                        li_element = combined_html.new_tag('li')
+                        li_element.string = all_paragraphs[
+                            i]['text'].strip()  # type:ignore
+                        ul_element.append(li_element)
+                        i += 1
+                    combined_html.body.append(ul_element)
+
+                else:
+                    # Regular paragraph or heading
+                    p_class = str(para['class'])
+                    p_text = para['text']
+
+                    # Determine heading level based on class
+                    tag_type = 'p'
+                    if '_01-Titel' in p_class:
+                        title = p_text.strip()  # type:ignore
+                        i += 1
+                        continue
+                    elif '_02-Titel' in p_class:
+                        tag_type = 'h2'
+                    elif '03-Titel' in p_class in p_class:
+                        tag_type = 'h3'
+                    elif '04-' in p_class in p_class:
+                        tag_type = 'h4'
+                    elif 'Vorlage_Vorlage-Titel' in p_class:
+                        tag_type = 'h2'
+                    elif p_class == 'table':
+                        if i > 0 and all_paragraphs[i-1]['class'] != 'table':
+                            element = combined_html.new_tag('p')
+                            element.string = '--- TABELLE HIER EINFÜGEN ---'
+                            combined_html.body.append(element)
+                        i += 1
+                        continue
+
+                    element = combined_html.new_tag(tag_type)
+                    element.string = p_text.strip()  # type:ignore
+                    combined_html.body.append(element)
+                    i += 1
+
+        cleaned_html = Markup(combined_html.prettify())  # nosec: B704
+
+        query = self.request.session.query(func.max(AgendaItem.number))
+        query = query.filter(AgendaItem.assembly_id == self.model.assembly.id)
+        next_number = (query.scalar() or 0) + 1
+
+        agenda_item = collection.add(
+            text=cleaned_html,
+            number=next_number,
+            state='draft',
+            title=title,
+            assembly_id=self.model.assembly.id
+        )
+
+        temp.cleanup()
+        return agenda_item
+
+    def validate_agenda_item_zip(
+            self,
+            field: UploadField) -> None:
+        if not field.data:
+            raise ValidationError(_('No file uploaded.'))
+
+        temp = TemporaryDirectory()
+        html_dir = self.get_html_dir(temp, field)
+        if not html_dir:
+            raise ValidationError(
+                _('No html directory found in the zip file.'))
+        temp.cleanup()
diff --git a/src/onegov/landsgemeinde/layouts/assembly.py b/src/onegov/landsgemeinde/layouts/assembly.py
@@ -130,7 +130,12 @@ def editbar_links(self) -> list[Link | LinkGroup] | None:
                             text=_('Agenda item'),
                             url=self.request.link(items, 'new'),
                             attrs={'class': 'check-list-link'}
-                        )
+                        ),
+                        Link(
+                            text=_('ZIP Upload'),
+                            url=self.request.link(items, 'new-import'),
+                            attrs={'class': 'ticket-archive'}
+                        ),
                     ]
                 )
             ]

diff --git a/src/onegov/landsgemeinde/locale/de_CH/LC_MESSAGES/onegov.landsgemeinde.po b/src/onegov/landsgemeinde/locale/de_CH/LC_MESSAGES/onegov.landsgemeinde.po
@@ -125,6 +125,18 @@ msgstr "Nummer wird bereits verwendet."
 msgid "Invalid timestamp."
 msgstr "Ungültiger Zeitstempel."
 
+msgid "Agenda Item ZIP"
+msgstr "Traktandum ZIP"
+
+msgid "Import"
+msgstr "Importieren"
+
+msgid "No html directory found in the zip file."
+msgstr "Keine HTML-Dateien im ZIP-Archiv gefunden."
+
+msgid "No file uploaded."
+msgstr "Keine Datei hochgeladen."
+
 msgid "Date"
 msgstr "Datum"
 
@@ -293,6 +305,9 @@ msgstr "Möchten Sie diese Landsgemeinde wirklich löschen?"
 msgid "Delete assembly"
 msgstr "Landsgemeinde löschen"
 
+msgid "ZIP Upload"
+msgstr "ZIP-Upload"
+
 #, python-format
 msgid "Extraodinary ${assembly_type} from ${date}"
 msgstr "Ausserordentliche ${assembly_type} vom ${date}"
@@ -419,6 +434,9 @@ msgstr "Neu"
 msgid "New agenda item"
 msgstr "Neues Traktandum"
 
+msgid "Imported a new agenda item"
+msgstr "Neues Traktandum importiert"
+
 msgid "Your changes were saved"
 msgstr "Ihre Änderungen wurden gespeichert"
 
@@ -446,12 +464,6 @@ msgstr "Neue Wortmeldung hinzugefügt"
 msgid "New votum"
 msgstr "Neue Wortmeldung"
 
-#~ msgid "Added a new assembly"
-#~ msgstr "Neue Landsgemeinde hinzugefügt"
-
-#~ msgid "New assembly"
-#~ msgstr "Neue Landsgemeinde"
-
 #, python-format
 #~ msgid "Extraodinary assembly from ${date}"
 #~ msgstr "Ausserordentliche Landsgemeinde vom ${date}"

diff --git a/src/onegov/landsgemeinde/views/agenda.py b/src/onegov/landsgemeinde/views/agenda.py
@@ -10,6 +10,7 @@
 from onegov.landsgemeinde import LandsgemeindeApp
 from onegov.landsgemeinde.collections import AgendaItemCollection
 from onegov.landsgemeinde.forms import AgendaItemForm
+from onegov.landsgemeinde.forms.agenda import AgendaItemUploadForm
 from onegov.landsgemeinde.layouts import AgendaItemCollectionLayout
 from onegov.landsgemeinde.layouts import AgendaItemLayout
 from onegov.landsgemeinde.models import AgendaItem
@@ -62,6 +63,35 @@ def add_agenda_item(
     }
 
 
+@LandsgemeindeApp.form(
+    model=AgendaItemCollection,
+    name='new-import',
+    template='form.pt',
+    permission=Private,
+    form=AgendaItemUploadForm
+)
+def import_agenda_item(
+    self: AgendaItemCollection,
+    request: LandsgemeindeRequest,
+    form: AgendaItemUploadForm
+) -> RenderData | Response:
+
+    if form.submitted(request):
+        agenda_item = form.import_agenda_item(self)
+        request.success(_('Imported a new agenda item'))
+        return redirect(request.link(agenda_item, 'edit'))
+
+    layout = AgendaItemCollectionLayout(self, request)
+    layout.breadcrumbs.append(Link(_('Import'), '#'))
+    layout.editbar_links = []
+
+    return {
+        'layout': layout,
+        'title': _('New agenda item'),
+        'form': form,
+    }
+
+
 @LandsgemeindeApp.html(
     model=AgendaItem,
     template='agenda_item.pt',