Improvements from Softcite

.dockerignore

@@ -0,0 +1,28 @@
+# Build outputs (regenerated inside the builder stage)
+build/
+.gradle/
+out/
+
+# IDE / editor / OS
+.idea/
+.vscode/
+*.iml
+.DS_Store
+
+# Runtime & logs (never needed at build time)
+logs/
+tmp/
+*.log
+
+# Research / developer helpers never copied into the image
+script/
+doc/
+
+# Markdown & meta — not copied into the image
+*.md
+LICENSE
+.claude/
+.github/
+
+# Note: .git/ is intentionally NOT listed — Dockerfile.datastet COPYs it so
+# Gradle can stamp revision.txt from `git rev-parse`.

.github/workflows/ci-build.yml

@@ -4,7 +4,7 @@ on: [push]
 
 concurrency:
   group: gradle
-#  cancel-in-progress: true
+  cancel-in-progress: true
 
 
 jobs:

Dockerfile.datastet

@@ -84,6 +84,15 @@ ENV DATASTET_OPTS="-Djava.library.path=/opt/grobid/grobid-home/lib/lin-64:/usr/l
 
 CMD ["./datastet/bin/datastet", "server", "datastet/resources/config/config.yml"]
 
+# Container-level health probe backed by the model-aware /service/health
+# endpoint, which returns HTTP 503 until every classifier reports "loaded".
+# The 180s start-period covers the worst case with modelPreload=true, where
+# SciBERT + four context classifiers + DataType classifier all load before
+# the first probe. Using python3 (required by JEP/DeLFT so guaranteed
+# present) to avoid a dependency on curl/wget from the base image.
+HEALTHCHECK --interval=30s --timeout=5s --start-period=180s --retries=3 \
+    CMD python3 -c "import urllib.request,sys; sys.exit(0 if urllib.request.urlopen('http://localhost:8060/service/health', timeout=4).status == 200 else 1)" || exit 1
+
 LABEL \
     authors="The contributors" \
     org.label-schema.name="datastet" \

build.gradle

@@ -30,6 +30,29 @@ apply plugin: 'java-library'
 apply plugin: 'base'
 apply plugin: 'com.github.kt3k.coveralls'
 
+def getGitRevision() {
+    def gitRevision = "unknown"
+    try {
+        def result = providers.exec {
+            workingDir = rootProject.rootDir
+            commandLine 'git', 'describe', '--tags', '--always', '--first-parent'
+        }
+        gitRevision = result.standardOutput.asText.get().trim()
+    } catch (Exception e) {
+        println "Could not get Git revision: ${e}"
+    }
+    return gitRevision
+}
+
+project.ext.gitRevision = getGitRevision()
+
+tasks.register('collectGitRevision') {
+    project.ext.gitRevision = getGitRevision()
+    doLast {
+        println "Git revision: ${project.ext.gitRevision}"
+    }
+}
+
 group = "org.grobid.datastet"
 version = '0.9.0'
 
@@ -281,6 +304,15 @@ artifacts {
     archives shadowJar
 }
 
+processResources {
+    filesMatching(["version.txt", "revision.txt"]) {
+        expand(
+            project_version: project.property('version') ?: "unknown",
+            project_revision: rootProject.ext.gitRevision
+        )
+    }
+}
+
 task copyModels(type: Copy) {
     from "${rootDir}/resources/models"
     include "**/*.wapiti"

src/main/java/org/grobid/core/engines/DatasetDisambiguator.java

@@ -146,7 +146,6 @@ public boolean checkIfAlive() {
     public void ensureCustomizationReady() {
         boolean result = false;
         URL url = null;
-        CloseableHttpResponse response = null;
         try {
             if ((nerd_port != null) && (nerd_port.length() > 0))
                 if (nerd_port.equals("443"))
@@ -157,24 +156,15 @@ public void ensureCustomizationReady() {
                 url = new URL("http://" + nerd_host + "/service/customisation/dataset");
 
             LOGGER.debug("Calling: " + url.toString());
-//System.out.println("Calling: " + url.toString());
-            CloseableHttpClient httpClient = HttpClients.createDefault();
             HttpGet get = new HttpGet(url.toString());
-            Scanner in = null;
-            try {
-                response = httpClient.execute(get);
-//System.out.println(response.getStatusLine());
+            try (CloseableHttpClient httpClient = HttpClients.createDefault();
+                 CloseableHttpResponse response = httpClient.execute(get)) {
                 int code = response.getStatusLine().getStatusCode();
                 if (code != 200) {
                     LOGGER.info("Failed customization lookup service: HTTP error code : " + code + " - the customization will be loaded");
                 } else {
                     result = true;
                 }
-            } finally {
-                if (in != null)
-                    in.close();
-                if (response != null)
-                    response.close();
             }
         } catch (MalformedURLException e) {
             LOGGER.warn("entity-fishing URL is malformed, customization skipped");
@@ -196,44 +186,34 @@ public void ensureCustomizationReady() {
                     url = new URL("http://" + nerd_host + "/service/customisations");
 
                 LOGGER.debug("Calling: " + url.toString());
-//System.out.println("Calling: " + url.toString());
                 // load the dataset customisation
                 File cutomisationFile = new File("resources/config/customisation-dataset.json");
                 cutomisationFile = new File(cutomisationFile.getAbsolutePath());
 
                 String json = FileUtils.readFileToString(cutomisationFile, "UTF-8");
 
-                CloseableHttpClient httpClient = HttpClients.createDefault();
                 HttpPost post = new HttpPost(url.toString());
 
-                //StringBody stringValue = new StringBody(json, ContentType.MULTIPART_FORM_DATA);
-                //StringBody stringName = new StringBody("dataset", ContentType.MULTIPART_FORM_DATA);
                 MultipartEntityBuilder builder = MultipartEntityBuilder.create();
                 builder.setMode(HttpMultipartMode.BROWSER_COMPATIBLE);
                 builder.addTextBody("value", json);
                 builder.addTextBody("name", "dataset");
-                //builder.addPart("value", stringValue);
-                //builder.addPart("name", stringName);
                 HttpEntity entity = builder.build();
-                try {
-                    post.setEntity(entity);
-                    response = httpClient.execute(post);
-//System.out.println(response.getStatusLine());
+                post.setEntity(entity);
 
+                try (CloseableHttpClient httpClient = HttpClients.createDefault();
+                     CloseableHttpResponse response = httpClient.execute(post)) {
                     int code = response.getStatusLine().getStatusCode();
                     if (code != 200) {
                         LOGGER.error("Failed loading dataset customisation: HTTP error code : " + code);
                     } else {
                         LOGGER.info("Dataset customisation loaded");
                     }
-                } finally {
-                    if (response != null)
-                        response.close();
                 }
             } catch (MalformedURLException e) {
-                e.printStackTrace();
+                LOGGER.warn("MalformedURLException while loading dataset customisation", e);
             } catch (IOException e) {
-                e.printStackTrace();
+                LOGGER.warn("I/O error while loading dataset customisation", e);
             }
         }
     }
@@ -489,8 +469,6 @@ public String runNerd(List<Dataset> entities, List<LayoutToken> subtokens, Strin
                     url = new URL("http://" + nerd_host + ":" + nerd_port + "/service/" + RESOURCEPATH);
             else
                 url = new URL("http://" + nerd_host + "/service/" + RESOURCEPATH);
-//System.out.println("calling... " + url.toString());
-            CloseableHttpClient httpClient = HttpClients.createDefault();
             HttpPost post = new HttpPost(url.toString());
             //post.addHeader("Content-Type", "application/json");
             //post.addHeader("Accept", "application/json");
@@ -567,32 +545,23 @@ public String runNerd(List<Dataset> entities, List<LayoutToken> subtokens, Strin
             builder.addPart("query", stringBody);
             HttpEntity entity = builder.build();
 
-            CloseableHttpResponse response = null;
-            Scanner in = null;
-            try {
-                //post.setEntity(new UrlEncodedFormEntity(params));
-                post.setEntity(entity);
-                response = httpClient.execute(post);
-                // System.out.println(response.getStatusLine());
-
+            post.setEntity(entity);
+            try (CloseableHttpClient httpClient = HttpClients.createDefault();
+                 CloseableHttpResponse response = httpClient.execute(post)) {
                 int code = response.getStatusLine().getStatusCode();
                 if (code != 200) {
                     LOGGER.warn("entity-fishing annotation returned HTTP " + code + ", disambiguation skipped");
                     return null;
                 }
 
                 HttpEntity entityResp = response.getEntity();
-                in = new Scanner(entityResp.getContent());
-                while (in.hasNext()) {
-                    output.append(in.next());
-                    output.append(" ");
+                try (Scanner in = new Scanner(entityResp.getContent())) {
+                    while (in.hasNext()) {
+                        output.append(in.next());
+                        output.append(" ");
+                    }
                 }
                 EntityUtils.consume(entityResp);
-            } finally {
-                if (in != null)
-                    in.close();
-                if (response != null)
-                    response.close();
             }
         } catch (MalformedURLException e) {
             LOGGER.warn("entity-fishing URL is malformed, disambiguation skipped");