From 05865c3d9bd2f3eb46dc7de7d69693c3c113eabe Mon Sep 17 00:00:00 2001
From: "Maarten L. Hekkelman" <maarten@hekkelman.com>
Date: Tue, 28 Jan 2025 15:51:40 +0100
Subject: [PATCH] Fixes for dictionary loading

---
 changelog                |  3 +++
 include/cif++/parser.hpp | 12 +++++++++++-
 src/compound.cpp         |  9 ++++-----
 test/unit-v2-test.cpp    |  2 +-
 4 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/changelog b/changelog
index 5656b3c..ea02532 100644
--- a/changelog
+++ b/changelog
@@ -2,6 +2,9 @@ Version 7.0.9
 - Using cif::file::load_dictionary it is now possible to
   load a dictionary along with its extensions in one go.
   E.g. file.load_dictionary("mmcif_pdbx;dssp-extension")
+- Fix in compound factory to avoid errors with lower case
+  compound id's
+- Fix sac_parser's index to be case insensitive
 
 Version 7.0.8
 - Fix PDB Remark 3 parser
diff --git a/include/cif++/parser.hpp b/include/cif++/parser.hpp
index 8244f51..594e713 100644
--- a/include/cif++/parser.hpp
+++ b/include/cif++/parser.hpp
@@ -73,7 +73,17 @@ class sac_parser
 {
   public:
 	/** @cond */
-	using datablock_index = std::map<std::string, std::size_t>;
+	struct iless_op
+	{
+		bool operator()(std::string a, std::string b) const
+		{
+			to_upper(a);
+			to_upper(b);
+			return a < b;
+		}
+	};
+
+	using datablock_index = std::map<std::string, std::size_t, iless_op>;
 
 	virtual ~sac_parser() = default;
 	/** @endcond */
diff --git a/src/compound.cpp b/src/compound.cpp
index f9e8c31..529aca9 100644
--- a/src/compound.cpp
+++ b/src/compound.cpp
@@ -296,8 +296,6 @@ class compound_factory_impl : public std::enable_shared_from_this<compound_facto
 
 	compound *get(std::string id)
 	{
-		cif::to_upper(id);
-
 		std::shared_lock lock(mMutex);
 
 		compound *result = nullptr;
@@ -318,7 +316,8 @@ class compound_factory_impl : public std::enable_shared_from_this<compound_facto
 				break;
 		}
 
-		if (result == nullptr and m_missing.count(id) == 0)
+		if (result == nullptr and
+			find_if(m_missing.begin(), m_missing.end(), [&id](const std::string &m_id) { return cif::iequals(id, m_id); }) == m_missing.end())
 		{
 			for (auto impl = shared_from_this(); impl; impl = impl->m_next)
 			{
@@ -328,7 +327,7 @@ class compound_factory_impl : public std::enable_shared_from_this<compound_facto
 			}
 
 			if (result == nullptr)
-				m_missing.insert(id);
+				m_missing.emplace_back(id);
 		}
 
 		return result;
@@ -361,7 +360,7 @@ class compound_factory_impl : public std::enable_shared_from_this<compound_facto
 	cif::parser::datablock_index m_index;
 
 	std::vector<compound *> m_compounds;
-	std::set<std::string> m_missing;
+	std::vector<std::string> m_missing;
 	std::shared_ptr<compound_factory_impl> m_next;
 };
 
diff --git a/test/unit-v2-test.cpp b/test/unit-v2-test.cpp
index 8c02c32..a76bf50 100644
--- a/test/unit-v2-test.cpp
+++ b/test/unit-v2-test.cpp
@@ -3482,7 +3482,7 @@ TEST_CASE("compound_test_1")
 	cif::compound_factory::instance().push_dictionary(gTestDir / "REA_v2.cif");
 	auto compound = cif::compound_factory::instance().create("REA_v2");
 	REQUIRE(compound != nullptr);
-	REQUIRE(compound->id() == "REA_v2");
+	REQUIRE(cif::iequals(compound->id(), "REA_v2"));
 }
 
 // --------------------------------------------------------------------