Feature/import graphml warning generic config

CHANGELOG.md

@@ -5,6 +5,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Added
+
+- import `graphml`: When generic config attributes are used, such as `extensions`, which the importer has to ignore, a warning is displayed
+
+### Changed
+
+- internal library for documentation generation had changed, as it also changed in graphannis
+
 ## [0.51.0] - 2026-04-10
 
 ### Added

Cargo.toml

@@ -21,12 +21,12 @@ console = "0.15"
 csv = "1.3"
 encoding_rs = "0.8"
 encoding_rs_io = "0.1.7"
-facet = "0.28.0"
-facet-reflect = "0.28.0"
+facet = "0.46.0"
+facet-reflect = "0.46.0"
 git2 = { version = "0.20.2", default-features = false }
 glob = "0.3"
-graphannis = "4.1.1"
-graphannis-core = "4.1.1"
+graphannis = "4.1.4"
+graphannis-core = "4.1.4"
 graphviz-rust = "0.9.7"
 indicatif = "0.17"
 itertools = "0.12"

src/bin/annatto.rs

@@ -244,7 +244,7 @@ fn list_modules() {
         peek_enum_variants(ReadFrom::SHAPE)
             .unwrap_or_default()
             .iter()
-            .map(|v| v.name.to_lowercase())
+            .map(|v| v.effective_name().to_lowercase())
             .join(", "),
     ];
     table_builder.push_record(import_row);
@@ -254,7 +254,7 @@ fn list_modules() {
         peek_enum_variants(WriteAs::SHAPE)
             .unwrap_or_default()
             .iter()
-            .map(|v| v.name.to_lowercase())
+            .map(|v| v.effective_name().to_lowercase())
             .join(", "),
     ];
     table_builder.push_record(export_row);
@@ -264,7 +264,7 @@ fn list_modules() {
         peek_enum_variants(GraphOp::SHAPE)
             .unwrap_or_default()
             .iter()
-            .map(|v| v.name.to_lowercase())
+            .map(|v| v.effective_name().to_lowercase())
             .join(", "),
     ];
     table_builder.push_record(graph_op_row);
@@ -288,18 +288,18 @@ fn module_info(name: &str) {
     let matching_importers: Vec<_> = peek_enum_variants(ReadFrom::SHAPE)
         .unwrap_or_default()
         .iter()
-        .filter(|m| m.name.to_lowercase() == name.to_lowercase())
+        .filter(|m| m.effective_name().to_lowercase() == name.to_lowercase())
         .collect();
     let matching_exporters: Vec<_> = peek_enum_variants(WriteAs::SHAPE)
         .unwrap_or_default()
         .iter()
-        .filter(|m| m.name.to_lowercase() == name.to_lowercase())
+        .filter(|m| m.effective_name().to_lowercase() == name.to_lowercase())
         .collect();
 
     let matching_graph_ops: Vec<_> = peek_enum_variants(GraphOp::SHAPE)
         .unwrap_or_default()
         .iter()
-        .filter(|m| m.name.to_lowercase() == name.to_lowercase())
+        .filter(|m| m.effective_name().to_lowercase() == name.to_lowercase())
         .collect();
 
     if matching_importers.is_empty()
@@ -333,7 +333,7 @@ fn module_info(name: &str) {
         print_markdown("# Graph operations\n\n");
         for m in matching_graph_ops {
             // The name of the module is taken from the wrapper enum
-            let module_name = m.name.to_lowercase();
+            let module_name = m.effective_name().to_lowercase();
             // Get the inner type wrapped by the graph operations enum and use
             // its documentation and fields
             if let Some(inner_field) = m.data.fields.first().map(|m| m.shape())
@@ -349,7 +349,7 @@ fn module_info(name: &str) {
                     .fields
                     .iter()
                     .map(|f| ModuleConfiguration {
-                        name: f.name.to_lowercase(),
+                        name: f.effective_name().to_lowercase(),
                         description: documentation::clean_string(f.doc),
                     })
                     .collect();

src/bin/documentation_generation/mod.rs

@@ -36,7 +36,7 @@ fn write_module_list_table(
         importers
             .iter()
             .map(|m| {
-                let module_name = m.name.to_lowercase();
+                let module_name = m.effective_name().to_lowercase();
                 let file_stem = module_name.replace("unstable:", "");
                 format!("[{module_name}](importers/{file_stem}.md)")
             })
@@ -49,7 +49,7 @@ fn write_module_list_table(
         exporters
             .iter()
             .map(|m| {
-                let module_name = m.name.to_lowercase();
+                let module_name = m.effective_name().to_lowercase();
                 let file_stem = module_name.replace("unstable:", "");
                 format!("[{module_name}](exporters/{file_stem}.md)")
             })
@@ -62,7 +62,7 @@ fn write_module_list_table(
         graph_ops
             .iter()
             .map(|m| {
-                let module_name = m.name.to_lowercase();
+                let module_name = m.effective_name().to_lowercase();
                 let file_stem = module_name.replace("unstable:", "");
                 format!("[{module_name}](graph_ops/{file_stem}.md)")
             })

src/exporter/table.rs

@@ -314,7 +314,7 @@ impl ExportTable {
                     let mut prefixes = sources.iter().map(|_| "in").collect_vec();
                     prefixes.extend(targets.iter().map(|_| "out"));
                     for ((connected_node_name, component, mut edge_annotations), prefix) in
-                        sources.into_iter().chain(targets.into_iter()).zip(prefixes)
+                        sources.into_iter().chain(targets).zip(prefixes)
                     {
                         let qualified_name = [
                             prefix,

src/importer/graphml.rs

@@ -163,43 +163,35 @@ fn read_graphml<R: std::io::BufRead>(
                 level += 1;
 
                 match e.name().0 {
-                    b"graph" => {
-                        if level == 2 {
-                            in_graph = true;
-                        }
+                    b"graph" if level == 2 => {
+                        in_graph = true;
                     }
-                    b"key" => {
-                        if level == 2 {
-                            add_annotation_key(&mut keys, e.attributes())?;
-                        }
+                    b"key" if level == 2 => {
+                        add_annotation_key(&mut keys, e.attributes())?;
                     }
-                    b"node" => {
-                        if in_graph && level == 3 {
-                            // Get the ID of this node
-                            for att in e.attributes() {
-                                let att = att?;
-                                if att.key.0 == b"id" {
-                                    current_node_id =
-                                        Some(String::from_utf8_lossy(&att.value).to_string());
-                                }
+                    b"node" if in_graph && level == 3 => {
+                        // Get the ID of this node
+                        for att in e.attributes() {
+                            let att = att?;
+                            if att.key.0 == b"id" {
+                                current_node_id =
+                                    Some(String::from_utf8_lossy(&att.value).to_string());
                             }
                         }
                     }
-                    b"edge" => {
-                        if in_graph && level == 3 {
-                            // Get the source and target node IDs
-                            for att in e.attributes() {
-                                let att = att?;
-                                if att.key.0 == b"source" {
-                                    current_source_id =
-                                        Some(String::from_utf8_lossy(&att.value).to_string());
-                                } else if att.key.0 == b"target" {
-                                    current_target_id =
-                                        Some(String::from_utf8_lossy(&att.value).to_string());
-                                } else if att.key.0 == b"label" {
-                                    current_component =
-                                        Some(String::from_utf8_lossy(&att.value).to_string());
-                                }
+                    b"edge" if in_graph && level == 3 => {
+                        // Get the source and target node IDs
+                        for att in e.attributes() {
+                            let att = att?;
+                            if att.key.0 == b"source" {
+                                current_source_id =
+                                    Some(String::from_utf8_lossy(&att.value).to_string());
+                            } else if att.key.0 == b"target" {
+                                current_target_id =
+                                    Some(String::from_utf8_lossy(&att.value).to_string());
+                            } else if att.key.0 == b"label" {
+                                current_component =
+                                    Some(String::from_utf8_lossy(&att.value).to_string());
                             }
                         }
                     }
@@ -281,11 +273,16 @@ impl Importer for GraphMLImporter {
         &self,
         path: &Path,
         step_id: StepID,
-        _config: GenericImportConfiguration,
+        config: GenericImportConfiguration,
         tx: Option<StatusSender>,
     ) -> Result<GraphUpdate, Box<dyn std::error::Error>> {
         let reporter = ProgressReporter::new(tx, step_id, 2)?;
 
+        if config != self.default_configuration() {
+            reporter
+                .warn("Generic configuration keys are currently ignored for GraphML imports.")?;
+        }
+
         // TODO: support multiple GraphML and connected binary files
         // TODO: refactor the graphannis_core create to expose the needed functionality directly
 
@@ -314,11 +311,15 @@ impl Importer for GraphMLImporter {
 
 #[cfg(test)]
 mod tests {
-    use std::path::Path;
+    use std::{path::Path, sync::mpsc};
 
     use insta::assert_snapshot;
+    use itertools::Itertools;
 
-    use crate::{importer::graphml::GraphMLImporter, test_util::import_as_graphml_string};
+    use crate::{
+        importer::{GenericImportConfiguration, Importer, graphml::GraphMLImporter},
+        test_util::import_as_graphml_string,
+    };
 
     #[test]
     fn single_sentence() {
@@ -331,4 +332,29 @@ mod tests {
 
         assert_snapshot!(actual);
     }
+
+    #[test]
+    fn generic_config_warning() {
+        let input_path = Path::new("tests/data/import/graphml/single_sentence.graphml");
+        let import = GraphMLImporter::default();
+        let (tx, rx) = mpsc::channel();
+        let import = import.import_corpus(
+            input_path,
+            crate::StepID {
+                module_name: "test_import".to_string(),
+                path: None,
+            },
+            GenericImportConfiguration::new_with_root_name("custom_root".to_string()),
+            Some(tx),
+        );
+        assert!(import.is_ok());
+        assert_snapshot!(
+            rx.into_iter()
+                .map(|m| match m {
+                    crate::workflow::StatusMessage::Warning(w) => w,
+                    _ => "".to_string(),
+                })
+                .join("\n")
+        );
+    }
 }

src/importer/snapshots/annatto__importer__graphml__tests__generic_config_warning.snap

@@ -0,0 +1,5 @@
+---
+source: src/importer/graphml.rs
+expression: "rx.into_iter().map(|m| match m\n{\n    crate::workflow::StatusMessage::Warning(w) => w, _ => \"\".to_string(),\n}).join(\"\\n\")"
+---
+Generic configuration keys are currently ignored for GraphML imports.

src/importer/text/tokenizer.rs

@@ -325,7 +325,7 @@ impl TreeTaggerTokenizer {
                         }
                     }
                     result.push(Token::new_val(current_token));
-                    result.extend(suffix.into_iter());
+                    result.extend(suffix);
                 }
             }
         }

src/importer/textgrid.rs

@@ -246,8 +246,7 @@ impl DocumentMapper<'_> {
             }
             let mut token_ids = Vec::new();
             let mut result = BTreeMap::new();
-            let mut counter = 1;
-            for (time_range, token_text) in token_sorted_by_time {
+            for (counter, (time_range, token_text)) in (1..).zip(token_sorted_by_time) {
                 let id = map_token(
                     u,
                     &NodeInfo::new(&counter.to_string(), &self.doc_path, &self.text_node_name),
@@ -261,7 +260,6 @@ impl DocumentMapper<'_> {
                 token_ids.push(id.clone());
                 result.insert(time_range.0, id.clone());
                 result.insert(time_range.1, id);
-                counter += 1;
             }
             add_order_relations(u, &token_ids, None)?;
 

src/importer/toolbox.rs

@@ -315,10 +315,8 @@ impl ImportToolBox {
                         };
                     }
                 }
-                (Rule::spaces, _) => {
-                    if build_joint {
-                        join_list.push(entry_or_space.as_str());
-                    }
+                (Rule::spaces, _) if build_joint => {
+                    join_list.push(entry_or_space.as_str());
                 }
                 (Rule::null, false) => {
                     timeline_id += 1;

src/manipulator/snapshots/annatto__manipulator__visualize__tests__dot_single_sentence_full.snap

@@ -17,8 +17,8 @@ digraph G {
   2[shape=box label="single_sentence/zossen#text\n \nannis:node_type=datasource"]
   1[shape=box label="single_sentence/zossen\n \nannis:doc=zossen\nannis:node_type=corpus"]
   0[shape=box label="single_sentence\n \nannis:node_type=corpus"]
-  3[shape=box label="single_sentence/zossen#n1\n \nsyntax:cat=ROOT\nannis:layer=syntax\nannis:node_type=node"]
   4[shape=box label="single_sentence/zossen#n2\n \nsyntax:cat=S\nannis:layer=syntax\nannis:node_type=node"]
+  3[shape=box label="single_sentence/zossen#n1\n \nsyntax:cat=ROOT\nannis:layer=syntax\nannis:node_type=node"]
   5[shape=box label="single_sentence/zossen#n3\n \nsyntax:cat=NP\nannis:layer=syntax\nannis:node_type=node"]
   8[shape=box label="single_sentence/zossen#n4\n \nsyntax:cat=PP\nannis:layer=syntax\nannis:node_type=node"]
   12[shape=box label="single_sentence/zossen#n5\n \nsyntax:cat=NP\nannis:layer=syntax\nannis:node_type=node"]

src/manipulator/snapshots/annatto__manipulator__visualize__tests__dot_single_sentence_limit.snap

@@ -13,8 +13,8 @@ digraph G {
   2[shape=box label="single_sentence/zossen#text\n \nannis:node_type=datasource"]
   1[shape=box label="single_sentence/zossen\n \nannis:doc=zossen\nannis:node_type=corpus"]
   0[shape=box label="single_sentence\n \nannis:node_type=corpus"]
-  3[shape=box label="single_sentence/zossen#n1\n \nsyntax:cat=ROOT\nannis:layer=syntax\nannis:node_type=node"]
   4[shape=box label="single_sentence/zossen#n2\n \nsyntax:cat=S\nannis:layer=syntax\nannis:node_type=node"]
+  3[shape=box label="single_sentence/zossen#n1\n \nsyntax:cat=ROOT\nannis:layer=syntax\nannis:node_type=node"]
   5[shape=box label="single_sentence/zossen#n3\n \nsyntax:cat=NP\nannis:layer=syntax\nannis:node_type=node"]
   8[shape=box label="single_sentence/zossen#n4\n \nsyntax:cat=PP\nannis:layer=syntax\nannis:node_type=node"]
   3 -> 6 [label="annis/inherited-coverage (C)" color=darkgreen fontcolor=darkgreen style=dotted]

src/util/documentation.rs

@@ -66,7 +66,7 @@ pub struct ModuleInfo {
 impl From<&Variant> for ModuleInfo {
     fn from(module: &Variant) -> Self {
         // The name of the module is taken from the wrapper enum
-        let module_name = module.name.to_lowercase();
+        let module_name = module.effective_name().to_lowercase();
         // Get the inner type wrapped by the graph operations enum and use
         // its documentation and fields
         let mut result = Self {
@@ -78,7 +78,7 @@ impl From<&Variant> for ModuleInfo {
             let shape = m.shape();
             if let Some(inner) = shape.inner {
                 // This can be a boxed type
-                inner()
+                inner
             } else {
                 shape
             }
@@ -93,7 +93,7 @@ impl From<&Variant> for ModuleInfo {
                 .fields
                 .iter()
                 .map(|f| ModuleConfiguration {
-                    name: f.name.to_lowercase(),
+                    name: f.effective_name().to_lowercase(),
                     description: clean_string(f.doc),
                 })
                 .collect();

src/workflow.rs

@@ -457,8 +457,7 @@ impl Workflow {
         }
         // Execute all manipulators in sequence
         if let Some(ref manipulators) = self.graph_op {
-            let mut graph_op_position = 1;
-            for desc in manipulators.iter() {
+            for (graph_op_position, desc) in (1..).zip(manipulators.iter()) {
                 let step_id = StepID::from_graphop_step(desc, graph_op_position);
                 let workflow_directory = &desc.workflow_directory;
                 desc.execute(
@@ -473,7 +472,6 @@ impl Workflow {
                     reason: reason.to_string(),
                     manipulator: step_id.to_string(),
                 })?;
-                graph_op_position += 1;
 
                 if let Some(ref tx) = tx {
                     tx.send(crate::workflow::StatusMessage::StepDone { id: step_id })?;