From ff78eb7e6d2f5e5dfe70ccf31724394b106b9d1b Mon Sep 17 00:00:00 2001
From: Simon Martens <simon.martens@mailbox.org>
Date: Wed, 5 Feb 2025 20:12:43 +0100
Subject: [PATCH] Bugfix: Kommentare in meta.xml erhalten

---
 data/xml/meta.xml              | 48 +++++++++++++++---------------
 data/xsd/meta.xsd              | 10 +++++++
 scripts/transform-meta-sort.py | 54 ++++++++++++++++++++++++----------
 3 files changed, 73 insertions(+), 39 deletions(-)

diff --git a/data/xml/meta.xml b/data/xml/meta.xml
index 94207de..fdf8ee7 100644
--- a/data/xml/meta.xml
+++ b/data/xml/meta.xml
@@ -16,7 +16,6 @@
 			<isProofread value="true" />
 			<isDraft value="false" />
 		</letterDesc>
-
 		<letterDesc letter="2">
 			<sent>
 				<date when="1767-10-11">Dorpat (Tartu), 11. Oktober 1767</date>
@@ -63,7 +62,7 @@
 
 		<letterDesc letter="5">
 			<sent>
-				<date when="1767-11-24" notBefore="1767-11-24">Tarwast (Tarvastu), nach 24. Januar 1768</date>
+				<date notBefore="1767-11-24" when="1767-11-24">Tarwast (Tarvastu), nach 24. Januar 1768</date>
 				<location ref="1" />
 				<person ref="1" />
 				<person ref="7" />
@@ -93,7 +92,7 @@
 
 		<letterDesc letter="7">
 			<sent>
-				<date when="1770-06-15" notBefore="1770-06-01" notAfter="1770-07-01">Dorpat (Tartu),
+				<date notBefore="1770-06-01" notAfter="1770-07-01" when="1770-06-15">Dorpat (Tartu),
 					wahrscheinlich Mitte 1770</date>
 				<location ref="1" />
 				<person ref="5" />
@@ -109,7 +108,7 @@
 
 		<letterDesc letter="8">
 			<sent>
-				<date when="1772-05-28" notBefore="1772-05-15" notAfter="1772-05-30">Fort Louis, Ende Mai
+				<date notBefore="1772-05-15" notAfter="1772-05-30" when="1772-05-28">Fort Louis, Ende Mai
 					1772</date>
 				<location ref="4" />
 				<person ref="1" />
@@ -373,7 +372,7 @@
 			<isDraft value="false" />
 		</letterDesc>
 
-		<letterDesc letter="27">
+		<letterDesc letter="27">			<!-- Handelt es sich hier um einen Entwurf? -->
 			<sent>
 				<date when="1772-12-10">Landau, 10. Dezember 1772</date>
 				<location ref="5" />
@@ -727,7 +726,7 @@
 			<hasOriginal value="false" />
 			<isProofread value="true" />
 			<isDraft value="false" />
-
+			<!-- Handelt es sich hier um einen Entwurf? -->
 		</letterDesc>
 
 		<letterDesc letter="52">
@@ -758,7 +757,7 @@
 			<isDraft value="false" />
 		</letterDesc>
 
-		<letterDesc letter="54">
+		<letterDesc letter="54">			<!-- Französischer Brief -->
 			<sent>
 				<date when="1775-06-20">Straßburg, Wahrscheinlich Juni 1775</date>
 				<location ref="7" />
@@ -1418,7 +1417,7 @@
 			<hasOriginal value="true" />
 			<isProofread value="true" />
 			<isDraft value="false" />
-
+			<!-- Teile des Briefs sind auf französisch geschrieben. -->
 		</letterDesc>
 
 		<letterDesc letter="101">
@@ -1451,7 +1450,7 @@
 
 		<letterDesc letter="103">
 			<sent>
-				<date when="1776-01-21" notBefore="1776-01-21" notAfter="1776-01-22">Straßburg, 21./22.
+				<date notBefore="1776-01-21" notAfter="1776-01-22" when="1776-01-21">Straßburg, 21./22.
 					Januar 1776</date>
 				<location ref="7" />
 				<person ref="1" />
@@ -1552,7 +1551,7 @@
 			<sent>
 				<date when="1776-02-09">Auf dem Weg von Hannover nach Kassel?, 9. Februar 1776</date>
 				<location ref="34" />
-
+				<!-- Der Ort kann nicht genau bestimmt werden -->				<!-- Für die Strecke extra Ort -->
 				<person ref="19" />
 			</sent>
 			<received>
@@ -1567,7 +1566,7 @@
 			<sent>
 				<date when="1776-02-12">Auf dem Weg von Hannover nach Kassel?, Februar 1776</date>
 				<location ref="34" />
-
+				<!-- Der Ort kann nicht genau bestimmt werden -->
 				<person ref="19" />
 			</sent>
 			<received>
@@ -1826,7 +1825,7 @@
 				<person ref="10" />
 			</received>
 			<hasOriginal value="false" />
-
+			<!-- Exzerptheft von Lavaters Hand; Abschrift? -->
 			<isProofread value="true" />
 			<isDraft value="false" />
 		</letterDesc>
@@ -2176,7 +2175,7 @@
 				<person ref="10" />
 			</received>
 			<hasOriginal value="false" />
-
+			<!-- Hier handelt es sich um das Exzerptheft von Lavaters Hand -->
 			<isProofread value="true" />
 			<isDraft value="false" />
 		</letterDesc>
@@ -2767,7 +2766,8 @@
 			<received>
 				<person ref="42" />
 				<person ref="43" />
-
+				<!-- Wie wird der Fall ausgezeichnet, dass unbestimmt ist, welche der beiden Personen als
+				receiver in Frage kommt? -->
 			</received>
 			<hasOriginal value="true" />
 			<isProofread value="true" />
@@ -3165,7 +3165,7 @@
 			<hasOriginal value="true" />
 			<isProofread value="true" />
 			<isDraft value="false" />
-
+			<!-- Dieser Brief ist zugleich ein Entwurf -->
 		</letterDesc>
 
 		<letterDesc letter="225">
@@ -3298,12 +3298,12 @@
 			<sent>
 				<date when="1776-09-20">Weimar [tatsächlich wohl aus Kochberg], 20. September 1776</date>
 				<location ref="8" />
-
+				<!-- Eigentlicher Ort? -->
 				<person ref="1" />
 			</sent>
 			<received>
 				<person ref="49" />
-
+				<!-- Wird sein Bruder auch genannt? -->				<!-- Onkel von Jakob Lenz benennen? -->
 			</received>
 			<hasOriginal value="true" />
 			<isProofread value="true" />
@@ -3468,7 +3468,7 @@
 			<sent>
 				<date when="1776-10-26">Wohl Berka oder Kochberg, Oktober 1776</date>
 				<location ref="14" />
-
+				<!-- Oder Kochberg -->				<!-- Extra location für den Fall -->
 				<person ref="1" />
 			</sent>
 			<received>
@@ -3483,7 +3483,7 @@
 			<sent>
 				<date when="1776-10-27">Wohl Berka oder Kochberg, Oktober/November 1776</date>
 				<location ref="14" />
-
+				<!-- Oder Kochberg -->
 				<person ref="1" />
 			</sent>
 			<received>
@@ -3595,19 +3595,19 @@
 		<letterDesc letter="255">
 			<sent>
 				<date when="1776-11-23">Weimar [wohl Berka], 23. November 1776</date>
-
+				<!-- Weimar [wohl Berka] -->
 				<location ref="8" />
 				<person ref="1" />
 			</sent>
 			<received>
 				<person ref="42" />
-
+				<!-- Unklar welche Laudth den Brief empfangen hat -->
 				<person ref="43" />
 			</received>
 			<hasOriginal value="true" />
 			<isProofread value="true" />
 			<isDraft value="true" />
-
+			<!-- Wohl Entwurf -->
 		</letterDesc>
 
 		<letterDesc letter="256">
@@ -3822,6 +3822,7 @@
 
 		<letterDesc letter="271">
 			<sent>
+				<!-- Ort fehlt -->
 				<date when="1777-01-31">Januar/Februar 1777</date>
 				<person ref="1" />
 			</sent>
@@ -3863,6 +3864,7 @@
 
 		<letterDesc letter="274">
 			<sent>
+				<!-- Kommentar in der Datierung -->
 				<date when="1777-04-09">Emmendingen, [vor 22. April 1777], warum nicht 9. April 1777</date>
 				<location ref="12" />
 				<person ref="1" />
@@ -3936,7 +3938,7 @@
 				<date when="1777-05-12">Schinznach/Zürich, 12.-15. Mai 1777</date>
 				<location ref="11" />
 				<person ref="10" />
-
+				<!-- Wer ist Absender Br.? hier heißt das Brief. Der Brief ist von und an Lavater -->
 			</sent>
 			<received>
 				<person ref="10" />
diff --git a/data/xsd/meta.xsd b/data/xsd/meta.xsd
index f788f69..540bb84 100644
--- a/data/xsd/meta.xsd
+++ b/data/xsd/meta.xsd
@@ -58,9 +58,19 @@
 									</xsd:sequence>
 									<xsd:attribute name="letter" type="xsd:nonNegativeInteger" use="required" />
 								</xsd:complexType>
+
 							</xsd:element>
 						</xsd:sequence>
 					</xsd:complexType>
+					<xsd:unique name="uniqueIndexLetter">
+						<xsd:annotation>
+							<xsd:documentation>Pflicht: Jeder Akteur hat eine eindeutige ID.</xsd:documentation>
+						</xsd:annotation>
+
+						<xsd:selector xpath="lenz:letterDesc" />
+						<xsd:field xpath="@letter" />
+					</xsd:unique>
+
 				</xsd:element>
 			</xsd:sequence>
 		</xsd:complexType>
diff --git a/scripts/transform-meta-sort.py b/scripts/transform-meta-sort.py
index e3b9eed..499f94f 100644
--- a/scripts/transform-meta-sort.py
+++ b/scripts/transform-meta-sort.py
@@ -1,32 +1,54 @@
-import xml.etree.ElementTree as ET
+#!/usr/bin/env python3
 
-def transform_dates_to_sort_text(input_file, output_file):
-    tree = ET.parse(input_file)
+import sys
+from lxml import etree
+
+def transform_sort_to_date(input_file, output_file):
+    """
+    Transform an XML file:
+     - Move <date>'s "value" attr as text to <sort>
+     - Remove that <date> element
+     - Rename <sort> to <date> and rename its @value to @when
+     - Preserve comments
+    """
+    # Use a parser that keeps comments
+    parser = etree.XMLParser(remove_comments=False)
+    tree = etree.parse(input_file, parser)
     root = tree.getroot()
 
-    # Find all <letterDesc> elements under <descriptions>
+    # If your XML has namespaces, you'll need namespace-aware searches.
+    # This example assumes no default namespace is used for 'descriptions' or 'letterDesc'.
     descriptions = root.find("descriptions")
     if descriptions is not None:
         for letter_desc in descriptions.findall("letterDesc"):
             sent = letter_desc.find("sent")
             if sent is not None:
-                # Locate the <date> and <sort> elements
-                date_el = sent.find("date")
+                # Locate <date> and <sort> inside <sent>
+                old_date_el = sent.find("date")
                 sort_el = sent.find("sort")
 
-                if date_el is not None and sort_el is not None:
-                    # Move the date attribute's value into sort_el's text
-                    date_value = date_el.get("value")
+                # 1) Move the old <date> @value into <sort>.text and remove <date>
+                if old_date_el is not None and sort_el is not None:
+                    date_value = old_date_el.get("value")
                     if date_value:
                         sort_el.text = date_value
+                    sent.remove(old_date_el)
 
-                    # Remove the <date> element from <sent>
-                    sent.remove(date_el)
+                # 2) Rename <sort> to <date>, and rename @value to @when
+                if sort_el is not None:
+                    sort_el.tag = "date"  # rename element
+                    old_value = sort_el.attrib.pop("value", None)
+                    if old_value is not None:
+                        sort_el.set("when", old_value)
 
-    # Write the modified tree to a new file
-    tree.write(output_file, encoding="UTF-8", xml_declaration=True)
+    # Write the modified tree, preserving comments
+    tree.write(output_file, encoding="UTF-8", xml_declaration=True, pretty_print=True)
 
 if __name__ == "__main__":
-    # Usage example:
-    # transform_dates_to_sort_text("input.xml", "output.xml")
-    transform_dates_to_sort_text("../data/xml/meta.xml", "output.xml")
+    if len(sys.argv) != 3:
+        print("Usage: {} INPUT.xml OUTPUT.xml".format(sys.argv[0]))
+        sys.exit(1)
+
+    input_file = sys.argv[1]
+    output_file = sys.argv[2]
+    transform_sort_to_date(input_file, output_file)