ddl.cpp: tweak for regexes

Author: wojdyr

src/ddl.cpp

@@ -475,9 +475,9 @@ void Ddl::read_ddl2_block(cif::Block& block) {
         // mmcif_pdbx_v50.dic uses custom flavour of regex:
         // character classes have unescaped \, but recognize \n, \t, etc.
         // Here is a quick fix:
-        std::string::size_type pos = re_str.find("/\\{}");
-        if (pos != std::string::npos)
-          re_str.replace(pos, 4, "/\\\\{}");
+        gemmi::replace_all(re_str, "/\\{}", "/\\\\{}");
+        // in binary, \<newline> is apparently meant to be ignored
+        gemmi::replace_all(re_str, "\\\n", "");
         auto flag = std::regex::awk | std::regex::optimize;
         regexes_.emplace(row.str(0), std::regex(re_str, flag));
       } catch (const std::regex_error& e) {

tests/mmcif_pdbx_v50_frag.dic

@@ -540,8 +540,8 @@ data_mmcif_pdbx.dic
                '[A-Za-z0-9]+(,[A-Za-z0-9]+)*'
 ;              A list of comma separated chain or asym ids.
 ;
-               3x4_matrices         char
-               '((([ \t]*-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? +){3}?(-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? *\n([\t ]*\n)*)){3})*(([ \t]*-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? +){3}?(-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? *\n([\t ]*\n)*)){2}(([ \t]*-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? +){3}(-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? *\n?([\t ]*\n)*))[ \t]*'
+
+3x4_matrices              char   "(((([ \t]*-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? +){3})?(-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? *\n([\t ]*\n)*)){3})*((([ \t]*-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? +){3})?(-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? *\n([\t ]*\n)*)){2}((([ \t]*-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? +){3})(-?(([0-9]+)[.]?|([0-9]*[.][0-9]+))([(][0-9]+[)])?([eE][+-]?[0-9]+)? *\n?([\t ]*\n)*))[ \t]*"
 ;              A set of 3x4 matrices separated by spaces and newlines on each line.
                Final newline optional. Optional spaces at start of lines. Blank lines accepted.
 ;
@@ -556,7 +556,7 @@ data_mmcif_pdbx.dic
    author char "[A-Za-z0-9_]+(( |-|'|\. )[A-Za-z0-9_]+)*( Jr.| III)?, [A-Za-z0-9_]\.(-?[A-Za-z0-9_]+\.)*$" 'Author name in PDB format: Taylor, C.A.'
    orcid_id char "[0-9]{4}-[0-9]{4}-[0-9]{4}-([0-9]{3}X|[0-9]{4})"  'ORCID pattern - dddd-dddd-dddd-dddd|dddX'
    symmetry_operation char '[-+0-9XxYyZ/ ]+,[-+0-9XxYyZ/ ]+,[-+0-9XxYyZ/ ]+' 'Allowed characters for use in symmetry operation such as 1/2-x,y,1/2-z'
-   sequence_dep        char '[a-zA-Z0-9\t \r\n\v\f\(\)]+$'  'Deposition specific one letter code'
+   sequence_dep        char '[a-zA-Z0-9\t \r\n()]+$'  'Deposition specific one letter code'
    date_dep            char '([1-9][0-9](([02468][048])|([13579][26]))-02-29)|[1-9][0-9][0-9][0-9]-((((0[1-9])|(1[0-2]))-((0[1-9])|(1[0-9])|(2[0-8])))|((((0[13578])|(1[02]))-31)|(((0[1,3-9])|(1[0-2]))-(29|30))))' 'Deposition specific date with better checking'
 
 #####################

tests/test_cif.py

@@ -391,12 +391,7 @@ def test_validation(self):
         msg_list = []
         ddl = cif.Ddl(logger=(lambda msg: msg_list.append(msg), 6))
         ddl.read_ddl(cif.read(full_path('mmcif_pdbx_v50_frag.dic')))
-        self.assertEqual(len(msg_list), 1,
-                         msg=f'Messages:\n  {"\n  ".join(msg_list)}')
-        for msg in msg_list:
-            expected = "Bad DDL2: can't parse regex for 'binary':"
-            self.assertTrue(msg.startswith(expected), msg=msg)
-        msg_list = []
+        self.assertEqual(msg_list, [])
         ddl.validate_cif(doc)
         self.assertEqual(msg_list,
                          ['[dummy_block] unknown tag _custom_tag',