Merge pull request #311 from korpling/feature/fix-saltxml-spans

Give correct name in xlsx importer to segmentation token.

CHANGELOG.md

@@ -5,6 +5,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Fixed
+
+- `xlsx` importer did not give the correct node name to segmentation token. Due
+  to this inconsistency, span annotations on segmentation nodes where not
+  connected to the segmentation token.
+
 ## [0.16.0] - 2024-09-02
 
 ### Added

src/exporter/exmaralda.rs

@@ -133,10 +133,7 @@ impl Exporter for ExportExmaralda {
                 };
             let doc_path = output_path.join(format!(
                 "{}.{extension}",
-                doc_name
-                    .split(|c| c == '/' || c == '\\')
-                    .last()
-                    .unwrap_or(&doc_name) // This always has a last
+                doc_name.split(['/', '\\']).last().unwrap_or(&doc_name) // This always has a last
             ));
             if let Some(doc_parent) = doc_path.as_path().parent() {
                 fs::create_dir_all(doc_parent)?;

src/importer/xlsx.rs

@@ -231,7 +231,7 @@ impl ImportSpreadsheet {
                         let overlapped_base_tokens: &[String] =
                             &base_tokens[base_token_start..base_token_end]; // TODO check indices
 
-                        let node_name = if token_annos.contains(name) {
+                        let node_name = if name == tok_name || token_annos.contains(name) {
                             format!("{}#{}_{}-{}", &doc_path, tok_name, start_row, end_row_excl)
                         } else {
                             format!(