replace cif::read_file with read_cif_gz in python bindings

and in two examples. In Python, cif.read_file() and gemmi.read_small_structure() can now read gzipped files.
project-gemmi · Feb 13, 2025 · 4579695 · 4579695
1 parent 8c41feb
commit 4579695
Show file tree

Hide file tree

Showing 5 changed files with 19 additions and 22 deletions.
diff --git a/examples/torsion_in_ring.cpp b/examples/torsion_in_ring.cpp
@@ -1,6 +1,6 @@
 // prints torsion restraints that are part of a ring
 
-#include <gemmi/cif.hpp>
+#include <gemmi/read_cif.hpp>
 #include <gemmi/chemcomp.hpp>
 //#include <gemmi/to_cif.hpp>  // for write_cif_to_stream
 #include <stdio.h>
@@ -9,7 +9,7 @@ namespace cif = gemmi::cif;
 using AtomId = gemmi::Restraints::AtomId;
 
 void check_torsions(const char* file_path) {
-  cif::Document doc = cif::read_file(file_path);
+  cif::Document doc = gemmi::read_cif_gz(file_path);
   for (cif::Block& block : doc.blocks) {
     if (block.name.empty() && block.name == "comp_list")
       continue;

diff --git a/examples/with_bgl.cpp b/examples/with_bgl.cpp
@@ -10,7 +10,7 @@
 #include <boost/graph/adjacency_list.hpp>
 #include <boost/graph/vf2_sub_graph_iso.hpp>
 #include <boost/graph/mcgregor_common_subgraphs.hpp>
-#include <gemmi/cif.hpp>             // for cif::read_file
+#include <gemmi/read_cif.hpp>        // for read_cif_gz
 #include <gemmi/chemcomp.hpp>        // for ChemComp, make_chemcomp_from_block
 
 struct AtomVertex {
@@ -41,7 +41,7 @@ Graph make_graph(const gemmi::ChemComp& cc) {
 }
 
 gemmi::ChemComp make_chemcomp(const char* path) {
-  gemmi::cif::Document doc = gemmi::cif::read_file(path);
+  gemmi::cif::Document doc = gemmi::read_cif_gz(path);
   // assuming the component description is in the last block of the file
   return gemmi::make_chemcomp_from_block(doc.blocks.back());
 }
@@ -83,7 +83,7 @@ void count_automorphisms_of_SO3() {
 //
 // Example output:
 //
-//  $ ./with_bgl ccd/M10.cif monomers/m/M10.cif 
+//  $ ./with_bgl ccd/M10.cif monomers/m/M10.cif
 //  isomorphic!
 //    O4 -> O9
 //    O9 -> O4
@@ -162,7 +162,7 @@ void check_subgraph_isomorphism(const char* cif1, const char* cif2) {
 //     CAE -> C7
 //     OAI -> O8
 //     CAD -> C1
-//   
+//
 //   real	0m0.012s
 //   user	0m0.008s
 //   sys	0m0.004s

diff --git a/python/cif.cpp b/python/cif.cpp
@@ -1,6 +1,7 @@
 // Copyright 2017 Global Phasing Ltd.
 
 #include <sstream>
+#include "gemmi/cif.hpp"  // for parse_input
 #include "gemmi/cifdoc.hpp"
 #include "gemmi/to_cif.hpp"
 #include "gemmi/to_json.hpp"
@@ -59,6 +60,16 @@ T& add_to_vector(std::vector<T>& vec, const T& new_item, int pos) {
   return vec[pos];
 }
 
+// cif parsing without checking for missing values and duplicates
+void cif_parse_string(Document& doc, const std::string& data) {
+  tao::pegtl::memory_input<> in(data, "string");
+  parse_input(doc, in);
+}
+void cif_parse_file(Document& doc, const std::string& filename) {
+  GEMMI_CIF_FILE_INPUT(in, filename);
+  parse_input(doc, in);
+}
+
 }  // anonymous namespace
 
 // for delitem_slice

diff --git a/python/common.h b/python/common.h
@@ -58,10 +58,6 @@ void add_write(nb::module_& m, nb::class_<gemmi::Structure>& structure);
 // defined in align.cpp
 void add_assign_label_seq_id(nb::class_<gemmi::Structure>& structure);
 
-// defined in read.cpp
-void cif_parse_string(gemmi::cif::Document& doc, const std::string& data);
-void cif_parse_file(gemmi::cif::Document& doc, const std::string& filename);
-
 // convert pythonic index to C++ index
 inline int c_index(int index, size_t size) {
   if (index < 0)

diff --git a/python/read.cpp b/python/read.cpp
@@ -7,7 +7,6 @@
 
 #include "gemmi/numb.hpp"
 #include "gemmi/cifdoc.hpp"
-#include "gemmi/cif.hpp"
 #include "gemmi/mmcif.hpp"         // for make_structure_from_block, ...
 #include "gemmi/pdb.hpp"           // for read_pdb_string
 #include "gemmi/gz.hpp"            // for estimate_uncompressed_size
@@ -24,7 +23,7 @@ using namespace gemmi;
 NB_MAKE_OPAQUE(std::vector<SmallStructure::Site>)
 
 void add_cif_read(nb::module_& cif) {
-  cif.def("read_file", &cif::read_file, nb::arg("filename"),
+  cif.def("read_file", &read_cif_gz, nb::arg("filename"),
           "Reads a CIF file copying data into Document.");
   cif.def("read", &read_cif_or_mmjson_gz,
           nb::arg("filename"), "Reads normal or gzipped CIF file.");
@@ -101,7 +100,7 @@ void add_read_structure(nb::module_& m) {
 
   // from smcif.hpp
   m.def("read_small_structure", [](const std::string& path) {
-          cif::Block block = cif::read_file(path).sole_block();
+          cif::Block block = read_cif_gz(path).sole_block();
           return new SmallStructure(make_small_structure_from_block(block));
         }, nb::arg("path"), "Reads a small molecule CIF file.");
   m.def("make_small_structure_from_block", &make_small_structure_from_block,
@@ -180,12 +179,3 @@ void add_small(nb::module_& m) {
         nb::arg("st"), nb::arg("n")=0);
 }
 
-// used in cif.cpp
-void cif_parse_string(cif::Document& doc, const std::string& data) {
-  tao::pegtl::memory_input<> in(data, "string");
-  cif::parse_input(doc, in);
-}
-void cif_parse_file(cif::Document& doc, const std::string& filename) {
-  GEMMI_CIF_FILE_INPUT(in, filename);
-  cif::parse_input(doc, in);
-}