From 335a711bca3ee2ee227a5680be8fb03bde82e90c Mon Sep 17 00:00:00 2001
From: Patrick Golden <ptgolden@email.unc.edu>
Date: Tue, 16 Dec 2025 09:56:52 -0500
Subject: [PATCH] Add option to split output according to predicates

This commit adds a `--split-with-predicates` flag which changes the
behavior of the `--split` flag. Typically, the output is split along a
subject prefix and an object prefix, e.g. `MONDO-to-NCIT.sssom.tsv`.
When the `--split-with-predicates` flag is passed, it also includes the
CURIE of the relation in the name of the output split file, e.g.
`MONDO-skos_exactMatch-NCIT.sssom.tsv`. (Note that the colon in the
CURIE has been replaced by an underscore, since colons cannot appear in
filenames in Windows).

This (almost) matches the behavior of sssom-py:

<https://github.com/mapping-commons/sssom-py/blob/ac0f769386d35e2b1c961cac59221c816143bf07/src/sssom/parsers.py#L1033-L1035>

The difference being that sssom-py only includes the local name of the
predicate, while this pull request includes the full CURIE.
---
 .../obofoundry/sssom/cli/SimpleCLI.java       | 23 ++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/cli/src/main/java/org/incenp/obofoundry/sssom/cli/SimpleCLI.java b/cli/src/main/java/org/incenp/obofoundry/sssom/cli/SimpleCLI.java
index 29c9bf8..c12b41d 100644
--- a/cli/src/main/java/org/incenp/obofoundry/sssom/cli/SimpleCLI.java
+++ b/cli/src/main/java/org/incenp/obofoundry/sssom/cli/SimpleCLI.java
@@ -198,6 +198,11 @@ private static class OutputOptions {
                 description = "Split the set along subject and object prefix names and write the split sets in the specified directory.")
         String splitDirectory;
 
+        @Option(names = "--split-with-predicates",
+                description = "When splitting, include the predicate CURIE in the split identifier.")
+        boolean splitWithPredicates;
+
+
         @Option(names = { "-c", "--force-cardinality" },
                 hidden = true,
                 description = "Include mapping cardinality values.")
@@ -646,7 +651,7 @@ private void writeOutput(MappingSet set) {
         }
 
         if ( outputOpts.splitDirectory != null ) {
-            writeSplitSet(set, outputOpts.splitDirectory);
+            writeSplitSet(set, outputOpts.splitDirectory, outputOpts.splitWithPredicates);
             return; // Skip writing the full set when writing splits
         }
         boolean stdout = outputOpts.file.equals("-");
@@ -660,7 +665,7 @@ private void writeOutput(MappingSet set) {
         }
     }
 
-    private void writeSplitSet(MappingSet ms, String directory) {
+    private void writeSplitSet(MappingSet ms, String directory, boolean splitWithPredicates) {
         File dir = new File(directory);
         if ( !dir.isDirectory() && !dir.mkdirs() ) {
             helper.error("cannot create directory %s", directory);
@@ -677,7 +682,19 @@ private void writeSplitSet(MappingSet ms, String directory) {
             String subjectPrefixName = pm.getPrefixName(mapping.getSubjectId());
             String objectPrefixName = pm.getPrefixName(mapping.getObjectId());
             if ( subjectPrefixName != null && objectPrefixName != null ) {
-                String splitId = subjectPrefixName + "-to-" + objectPrefixName;
+                String splitId;
+                if (splitWithPredicates) {
+                    String predicatePrefixName = pm.getPrefixName(mapping.getPredicateId());
+                    if (predicatePrefixName != null) {
+                        splitId = subjectPrefixName + "-" + pm.shortenIdentifier(mapping.getPredicateId()) + "-" + objectPrefixName;
+                        splitId = splitId.replace(":", "_");
+                    } else {
+                        splitId = subjectPrefixName + "-to-" + objectPrefixName;
+                    }
+                } else {
+                    splitId = subjectPrefixName + "-to-" + objectPrefixName;
+                }
+
                 mappingsBySplit.computeIfAbsent(splitId, k -> new ArrayList<Mapping>()).add(mapping);
             }
         }