From f62b083d623ba5acc20411fbbba2b5c600550ba8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=C5=81ukasz=20Szeremeta?= <l.szeremeta.dev@gmail.com>
Date: Thu, 8 Apr 2021 20:12:43 +0200
Subject: [PATCH] Add --base option

---
 README.md                                     |  1 +
 .../java/pl/edu/uwb/ii/sdfeater/Molecule.java | 27 ++++++++++++++++---
 .../java/pl/edu/uwb/ii/sdfeater/SDFEater.java |  9 +++++++
 3 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index c9c1dca..4fa081e 100644
--- a/README.md
+++ b/README.md
@@ -93,6 +93,7 @@ Running SDFEater without parameters displays help.
 * `-i,--input <arg>` - input SDF file path (required)
 * `-f,--format <arg>` - output format (e.g. `cypher`, `jsonld`, `cvme`, `smiles`, `inchi`) (required; full list below)
 * `-s,--subject <arg>` - subject type (`iri`, `uuid`, `bnode`; `iri` by default; for all formats excluding cypher, cvme, smiles, inchi)
+* `-b,--base <arg>` - molecule subject base for 'iri' subject type ('https://example.com/molecule#entity' by default)
 
 Remember about the appropriate file path when using Docker image. Suppose you mounted your local directory `/home/user/input` under `/app/input` and the path to the SDF file you want to use in SDFEater is `/home/user/input/file.sdf`. In this case, enter the path `/app/input/file.sdf` or `input/file.sdf` as the value of the `-i` argument.
 
diff --git a/src/main/java/pl/edu/uwb/ii/sdfeater/Molecule.java b/src/main/java/pl/edu/uwb/ii/sdfeater/Molecule.java
index 7738cb9..764a610 100644
--- a/src/main/java/pl/edu/uwb/ii/sdfeater/Molecule.java
+++ b/src/main/java/pl/edu/uwb/ii/sdfeater/Molecule.java
@@ -71,6 +71,12 @@ class Molecule {
      * Stores all properties of the chemical molecule
      */
     private final Map<String, List<String>> properties = new HashMap<>();
+
+    /**
+     * Subject base of molecule
+     */
+    String subjectBase = "https://example.com/molecule#entity";
+
     private UUID uuid;
 
     Molecule() {
@@ -332,7 +338,7 @@ void addToJenaModel(SDFEater.Subject subject) {
         Resource me = ResourceFactory.createResource();
 
         if (subject == SDFEater.Subject.iri) {
-            me = ResourceFactory.createResource("https://example.com/molecule#entity" + createID());
+            me = ResourceFactory.createResource(subjectBase + createID());
         } else if (subject == SDFEater.Subject.uuid) {
             me = ResourceFactory.createResource("urn:uuid:" + uuid);
         } else if (subject == SDFEater.Subject.bnode) {
@@ -401,7 +407,7 @@ StringBuilder constructJSONLDMolecule(SDFEater.Subject subject) {
 
         output_str.append("    {\n");
         if (subject == SDFEater.Subject.iri) {
-            output_str.append("      \"@id\" : \"https://example.com/molecule#entity" + createID() + "\",\n");
+            output_str.append("      \"@id\" : \"" + subjectBase + createID() + "\",\n");
         } else if (subject == SDFEater.Subject.uuid) {
             output_str.append("      \"@id\" : \"urn:uuid:" + uuid + "\",\n");
         } else if (subject == SDFEater.Subject.bnode) {
@@ -499,7 +505,14 @@ void printRDFaMolecule(SDFEater.Subject subject) {
         if (output_str.length() > 0) {
             if (subject == SDFEater.Subject.iri) {
                 String mID = createID();
-                System.out.println("    <div typeof='schema:MolecularEntity' about='https://example.com/molecule#entity" + mID + "' id='entity" + mID + "'>");
+                System.out.print("    <div typeof='schema:MolecularEntity' about='" + subjectBase + mID + "'");
+
+                if (subjectBase.contains("#")){
+                    System.out.print(" id='" +  subjectBase.substring(subjectBase.lastIndexOf('#') + 1) + mID + "'");
+                }
+
+                System.out.print(">\n");
+
             } else if (subject == SDFEater.Subject.uuid) {
                 System.out.println("    <div typeof='schema:MolecularEntity' about='urn:uuid:" + uuid + "'>");
             } else if (subject == SDFEater.Subject.bnode) {
@@ -559,7 +572,13 @@ void printMicrodataMolecule(SDFEater.Subject subject) {
         if (output_str.length() > 0) {
             if (subject == SDFEater.Subject.iri) {
                 String mID = createID();
-                System.out.println("    <div itemscope itemtype='http://schema.org/MolecularEntity' itemid='https://example.com/molecule#entity" + mID + "' id='entity" + mID + "'>");
+                System.out.print("    <div itemscope itemtype='http://schema.org/MolecularEntity' itemid='" + subjectBase + mID + "'");
+
+                if (subjectBase.contains("#")){
+                    System.out.print(" id='" +  subjectBase.substring(subjectBase.lastIndexOf('#') + 1) + mID + "'");
+                }
+
+                System.out.print(">\n");
             } else if (subject == SDFEater.Subject.uuid) {
                 System.out.println("    <div itemscope itemtype='http://schema.org/MolecularEntity' itemid='urn:uuid:" + uuid + "'>");
             } else if (subject == SDFEater.Subject.bnode) {
diff --git a/src/main/java/pl/edu/uwb/ii/sdfeater/SDFEater.java b/src/main/java/pl/edu/uwb/ii/sdfeater/SDFEater.java
index 50c9d5f..9a07143 100644
--- a/src/main/java/pl/edu/uwb/ii/sdfeater/SDFEater.java
+++ b/src/main/java/pl/edu/uwb/ii/sdfeater/SDFEater.java
@@ -90,6 +90,9 @@ public static void main(String[] args) {
         Option subject = new Option("s", "subject", true, "subject type (iri, uuid, bnode; iri by default); for all formats excluding cypher, cvme, smiles, inchi");
         subject.setRequired(false);
         options.addOption(subject);
+        Option base = new Option("b", "base", true, "molecule subject base for 'iri' subject type ('" + molecule.subjectBase + "' by default)");
+        base.setRequired(false);
+        options.addOption(base);
         CommandLineParser parser = new DefaultParser();
         HelpFormatter formatter = new HelpFormatter();
         CommandLine cmd;
@@ -111,6 +114,12 @@ public static void main(String[] args) {
                     initializeJenaModel();
                     break;
             }
+
+            // replace default base molecule IRI
+            if (cmd.getOptionValue("subject", Subject.iri.toString()).equals(Subject.iri.toString())) {
+                molecule.subjectBase = cmd.getOptionValue("base", molecule.subjectBase);
+            }
+
             file.parse(molecule, Format.valueOf(cmd.getOptionValue("format")), Subject.valueOf(cmd.getOptionValue("subject", Subject.iri.toString())));
         } catch (IllegalArgumentException e) {
             System.err.println("Incorrect option selected");