star-hitran

commit 2c37705a0739bdfbc1c77fd4df177ff7eda1277f
parent 0d1212b9bba3dfa7589772b0beca126530e1a852
Author: Vincent Forest <vincent.forest@meso-star.com>
Date:   Fri, 26 Jun 2026 16:03:22 +0200

Improve detection of duplicate molecules

Verifies whether the molecule name matches the expected name for its
identifier. Thus, once the analysis is successfully completed, if a
molecule is already registered under the corresponding identifier, it
means it is a duplicate.

Note that previously, when a conflict was detected, accessing a NULL
pointer caused a segmentation fault. This commit not only makes the
error message clearer but also corrects this invalid read.

Diffstat:
M
src/shtr_isotope_metadata.c
 | 
39
+++++++++++++++++++++++++--------------
M
src/test_shtr_isotope_metadata.c
 | 
18
++++++++++++++++++

2 files changed, 43 insertions(+), 14 deletions(-)
diff --git a/src/shtr_isotope_metadata.c b/src/shtr_isotope_metadata.c
@@ -155,7 +155,6 @@ flush_molecule
 {
   size_t nisotopes = 0;
   size_t ientry = SIZE_MAX;
-  size_t* pimolecule = NULL;
   res_T res = RES_OK;
   ASSERT(metadata && molecule && MOLECULE_IS_VALID(molecule));
 
@@ -195,16 +194,9 @@ flush_molecule
 
   /* Register the molecule */
   if(metadata->molid2idx[molecule->id] >= 0) {
-    const struct molecule* molecule2 = NULL;
-    molecule2 = darray_molecule_cdata_get(&metadata->molecules) + *pimolecule;
-    ERROR(metadata->shtr,
-      "%s: cannot register the %s molecule. "
-      "The %s molecule has the same identifier %i.\n",
-      txtrdr_get_name(txtrdr),
-      str_cget(&molecule->name),
-      str_cget(&molecule2->name),
-      molecule->id);
-    res = RES_OK;
+    ERROR(metadata->shtr, "%s: the molecule %s appears several times.\n",
+      txtrdr_get_name(txtrdr), str_cget(&molecule->name));
+    res = RES_BAD_ARG;
     goto error;
   }
   ASSERT((size_t)((int)ientry) == ientry);
@@ -246,7 +238,7 @@ parse_molecule
   res = str_set(&molecule->name, name);
   if(res != RES_OK) {
     ERROR(metadata->shtr,
-      "%s:%lu: error seting the molecule name `%s' -- %s.\n",
+      "%s:%lu: error setting the molecule name `%s' -- %s.\n",
       txtrdr_get_name(txtrdr), (unsigned long)txtrdr_get_line_num(txtrdr),
       name, res_to_cstr(res));
     goto error;
@@ -263,14 +255,33 @@ parse_molecule
 
   id[len-1] = '\0'; /* Rm trailing parenthesis */
   res = cstr_to_int(id+1/*Rm leading parenthesis*/, &molecule->id);
-  if(res != RES_OK || !MOLECULE_IS_VALID(molecule)) {
+  if(res != RES_OK) {
     id[len-1] = ')'; /* Re-add the trailing parenthesis */
-    ERROR(metadata->shtr, "%s:%lu: invalid molecule identifier `%s'.\n",
+    ERROR(metadata->shtr,
+      "%s:%lu: invalid molecule identifier `%s'.\n",
       txtrdr_get_name(txtrdr), (unsigned long)txtrdr_get_line_num(txtrdr), id);
     res = RES_BAD_ARG;
     goto error;
   }
 
+  if(!MOLECULE_IS_VALID(molecule)) {
+    ERROR(metadata->shtr,
+      "%s:%lu: the `%s (%d)' molecule is not supported.\n",
+      txtrdr_get_name(txtrdr), (unsigned long)txtrdr_get_line_num(txtrdr),
+      name, molecule->id);
+    res = RES_BAD_ARG;
+    goto error;
+  }
+
+  if(strcmp(name, shtr_molecule_cstr(molecule->id))) {
+    ERROR(metadata->shtr,
+      "%s:%li: the molecule %d is named `%s' instead of `%s'.\n",
+      txtrdr_get_name(txtrdr), (unsigned long)txtrdr_get_line_num(txtrdr),
+      molecule->id, name, shtr_molecule_cstr(molecule->id));
+    res = RES_BAD_ARG;
+    goto error;
+  }
+
   tk = strtok_r(NULL, " \t", &tk_ctx);
   if(tk) {
     WARN(metadata->shtr, "%s:%lu: unexpected text `%s'.\n",
diff --git a/src/test_shtr_isotope_metadata.c b/src/test_shtr_isotope_metadata.c
@@ -559,6 +559,24 @@ test_load_failures(struct shtr* shtr)
   rewind(fp);
   CHK(shtr_isotope_metadata_load_stream(shtr, fp, NULL, &mdata) == RES_MEM_ERR);
   CHK(fclose(fp) == 0);
+
+  /* Duplicate molecule */
+  CHK(fp = tmpfile());
+  fprintf(fp, "Molecule # Iso Abundance Q(296K) gj Molar Mass(g)\n");
+  molecule_print(fp, &H2O);
+  molecule_print(fp, &H2O);
+  rewind(fp);
+  CHK(shtr_isotope_metadata_load_stream(shtr, fp, NULL, &mdata) == RES_BAD_ARG);
+  CHK(fclose(fp) == 0);
+
+  /* Invalid molecule name */
+  CHK(fp = tmpfile());
+  fprintf(fp, "Comment line\n");
+  fprintf(fp, "H_2O (1)\n");
+  isotope_print(fp, &H2O_isotopes[0]);
+  rewind(fp);
+  CHK(shtr_isotope_metadata_load_stream(shtr, fp, NULL, &mdata) == RES_BAD_ARG);
+  CHK(fclose(fp) == 0);
 }
 
 static void