Add SaltXML importer

CHANGELOG.md

@@ -12,6 +12,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - `map` manipulator can now add annotated spans and copy values from existing
   annotations. The copied values can be manipulated using regular expressions and
   replacement values.
+- Addes `saltxml` import format
 
 ### Fixed
 

Cargo.toml

@@ -1,5 +1,8 @@
 [package]
-authors = ["Thomas Krause <[email protected]>", "Martin Klotz <[email protected]>"]
+authors = [
+    "Thomas Krause <[email protected]>",
+    "Martin Klotz <[email protected]>",
+]
 description = "Converts linguistic data formats based on the graphANNIS data model as intermediate representation and can apply consistency tests."
 edition = "2018"
 homepage = "https://github.com/korpling/annatto/"
@@ -11,7 +14,7 @@ version = "0.14.0"
 [dependencies]
 ansi_term = "0.12"
 anyhow = "1.0"
-clap = {version = "4.0", features = ["derive", "env"]}
+clap = { version = "4.0", features = ["derive", "env"] }
 console = "0.15"
 csv = "1.1"
 documented = "0.3.0"
@@ -27,32 +30,34 @@ lazy_static = "1.4.0"
 linked-hash-map = "0.5.6"
 log = "0.4"
 normpath = "1.1"
-ordered-float = {version = "4.1", default-features = false}
+ordered-float = { version = "4.1", default-features = false }
 pathdiff = "0.2"
 percent-encoding = "2.3.1"
 pest = "2.7"
 pest_derive = "2.0"
-quick-xml = "0.31"
+quick-xml = "0.34"
 rayon = "1.1"
 regex = "1.10"
+roxmltree = "0.20.0"
 serde = "1.0"
 serde_derive = "1.0"
 struct-field-names-as-array = "0.3.0"
-strum = {version = "0.26.2", features = ["derive"]}
-tabled = {version = "0.15", features = ["ansi"]}
+strum = { version = "0.26.2", features = ["derive"] }
+tabled = { version = "0.15", features = ["ansi"] }
 tempfile = "3"
 termimad = "0.29.1"
 text-splitter = "0.6.3"
 thiserror = "1.0"
 toml = "0.8.0"
-tracing-subscriber = {version = "0.3", features = ["env-filter"]}
+tracing-subscriber = { version = "0.3", features = ["env-filter"] }
 umya-spreadsheet = "~1.1.1"
+url = "2.5.2"
 xml-rs = "0.8"
 zip = "0.6.6"
 
 [dev-dependencies]
 assert_cmd = "2.0.11"
-insta = {version = "1.26.0", features = ["toml", "filters"]}
+insta = { version = "1.26.0", features = ["toml", "filters"] }
 pretty_assertions = "1.3"
 
 # Compile some of the dependencies in release mode if when we are ourself in
@@ -82,7 +87,12 @@ ci = "github"
 # The installers to generate for each app
 installers = []
 # Target platforms to build apps for (Rust target-triple syntax)
-targets = ["aarch64-apple-darwin", "x86_64-apple-darwin", "x86_64-unknown-linux-gnu", "x86_64-pc-windows-msvc"]
+targets = [
+    "aarch64-apple-darwin",
+    "x86_64-apple-darwin",
+    "x86_64-unknown-linux-gnu",
+    "x86_64-pc-windows-msvc",
+]
 # The preferred cargo-dist version to use in CI (Cargo.toml SemVer syntax)
 cargo-dist-version = "0.16.0"
 # Publish jobs to run in CI

docs/README.md

@@ -1,5 +1,5 @@
-| Type             | Modules                                                                                                                                                                                                                                                                                                                                                                                                                                    |
-|------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| Import formats   | [conllu](importers/conllu.md), [exmaralda](importers/exmaralda.md), [graphml](importers/graphml.md), [meta](importers/meta.md), [none](importers/none.md), [opus](importers/opus.md), [path](importers/path.md), [ptb](importers/ptb.md), [relannis](importers/relannis.md), [textgrid](importers/textgrid.md), [toolbox](importers/toolbox.md), [treetagger](importers/treetagger.md), [xlsx](importers/xlsx.md), [xml](importers/xml.md) |
-| Export formats   | [graphml](exporters/graphml.md), [exmaralda](exporters/exmaralda.md), [sequence](exporters/sequence.md), [textgrid](exporters/textgrid.md), [xlsx](exporters/xlsx.md)                                                                                                                                                                                                                                                                      |
-| Graph operations | [check](graph_ops/check.md), [collapse](graph_ops/collapse.md), [visualize](graph_ops/visualize.md), [enumerate](graph_ops/enumerate.md), [link](graph_ops/link.md), [map](graph_ops/map.md), [revise](graph_ops/revise.md), [chunk](graph_ops/chunk.md), [split](graph_ops/split.md), [none](graph_ops/none.md)                                                                                                                           |
+| Type             | Modules                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
+|------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| Import formats   | [conllu](importers/conllu.md), [exmaralda](importers/exmaralda.md), [graphml](importers/graphml.md), [meta](importers/meta.md), [none](importers/none.md), [opus](importers/opus.md), [path](importers/path.md), [ptb](importers/ptb.md), [relannis](importers/relannis.md), [saltxml](importers/saltxml.md), [textgrid](importers/textgrid.md), [toolbox](importers/toolbox.md), [treetagger](importers/treetagger.md), [xlsx](importers/xlsx.md), [xml](importers/xml.md) |
+| Export formats   | [graphml](exporters/graphml.md), [exmaralda](exporters/exmaralda.md), [sequence](exporters/sequence.md), [textgrid](exporters/textgrid.md), [xlsx](exporters/xlsx.md)                                                                                                                                                                                                                                                                                                       |
+| Graph operations | [check](graph_ops/check.md), [collapse](graph_ops/collapse.md), [visualize](graph_ops/visualize.md), [enumerate](graph_ops/enumerate.md), [link](graph_ops/link.md), [map](graph_ops/map.md), [revise](graph_ops/revise.md), [chunk](graph_ops/chunk.md), [split](graph_ops/split.md), [none](graph_ops/none.md)                                                                                                                                                            |

docs/exporters/saltxml.md

@@ -0,0 +1,6 @@
+# saltxml (exporter)
+
+Exports Excel Spreadsheets where each line is a token, the other columns are
+spans and merged cells can be used for spans that cover more than one token.
+
+*No Configuration*

docs/importers/saltxml.md

@@ -0,0 +1,14 @@
+# saltxml (importer)
+
+Imports the SaltXML format used by Pepper (<https://corpus-tools.org/pepper/>).
+SaltXML is an XMI serialization of the [Salt model](https://raw.githubusercontent.com/korpling/salt/master/gh-site/doc/salt_modelGuide.pdf).
+
+## Configuration
+
+###  missing_anno_ns_from_layer
+
+If `true`, use the layer name as fallback for the namespace annotations
+if none is given. This is consistent with how the ANNIS tree visualizer
+handles annotations without any namespace. If `false`, use the
+`default_ns` namespace as fallback.
+

src/exporter/graphml.rs

@@ -238,17 +238,17 @@ fn media_vis(graph: &AnnotationGraph) -> Result<Vec<Visualizer>, Box<dyn std::er
                             layer: None,
                             vis_type: "audio".to_string(),
                             display_name: "audio".to_string(),
-                            visibility: "hidden".to_string(),
+                            visibility: "preloaded".to_string(),
                             mappings: None,
                         });
                     }
-                    "mp4" | "avi" | "mov" => {
+                    "mp4" | "avi" | "mov" | "webm" => {
                         vis.push(Visualizer {
                             element: "node".to_string(),
                             layer: None,
                             vis_type: "video".to_string(),
                             display_name: "video".to_string(),
-                            visibility: "hidden".to_string(),
+                            visibility: "preloaded".to_string(),
                             mappings: None,
                         });
                     }
@@ -350,35 +350,36 @@ fn node_annos_vis(graph: &AnnotationGraph) -> Result<Visualizer, Box<dyn std::er
     mappings.insert("annos".to_string(), node_names);
     mappings.insert("escape_html".to_string(), "false".to_string());
 
-    let more_than_one_ordering = order_names.len() > 1;
-    let ordered_nodes_are_identical = {
-        more_than_one_ordering && {
-            let ordering_components =
-                graph.get_all_components(Some(AnnotationComponentType::Ordering), None);
-            let node_sets = ordering_components
-                .iter()
-                .map(|c| {
-                    if let Some(strge) = graph.get_graphstorage(c) {
-                        strge
-                            .source_nodes()
-                            .filter_map(|r| if let Ok(n) = r { Some(n) } else { None })
-                            .collect::<BTreeSet<u64>>()
-                    } else {
-                        BTreeSet::default()
-                    }
-                })
-                .collect_vec();
-            let mut all_same = true;
-            //for i in 1..node_sets.len()
-            for (a, b) in node_sets.into_iter().tuple_windows() {
-                all_same &= matches!(a.cmp(&b), Ordering::Equal);
-            }
-            all_same
+    let ordered_components_contain_identical_nodes = if order_names.len() > 1 {
+        let ordering_components =
+            graph.get_all_components(Some(AnnotationComponentType::Ordering), None);
+        let node_sets = ordering_components
+            .iter()
+            .map(|c| {
+                if let Some(strge) = graph.get_graphstorage(c) {
+                    strge
+                        .source_nodes()
+                        .filter_map(|r| if let Ok(n) = r { Some(n) } else { None })
+                        .collect::<BTreeSet<u64>>()
+                } else {
+                    BTreeSet::default()
+                }
+            })
+            .collect_vec();
+        let mut all_same = true;
+        //for i in 1..node_sets.len()
+        for (a, b) in node_sets.into_iter().tuple_windows() {
+            all_same &= matches!(a.cmp(&b), Ordering::Equal);
         }
+        all_same
+    } else {
+        // There is only one ordering component
+        true
     };
+
     mappings.insert(
         "hide_tok".to_string(),
-        (!ordered_nodes_are_identical).to_string(),
+        (!ordered_components_contain_identical_nodes).to_string(),
     );
     mappings.insert("show_ns".to_string(), "false".to_string());
     Ok(Visualizer {