further changes to LaTeX ebnf-transformation

DHParser/ebnf.py

@@ -317,9 +317,9 @@ class EBNFCompiler(Compiler):
         self.current_symbols = []   # type: List[Node]
         self.symbols = {}           # type: Dict[str, Node]
         self.variables = set()      # type: Set[str]
-        self.definitions = []  # type: List[Tuple[str, str]]
+        # self.definitions = []     # type: List[Tuple[str, str]]
         self.recursive = set()      # type: Set[str]
-        self.deferred_tasks = []  # type: List[Callable]
+        self.deferred_tasks = []    # type: List[Callable]
         self.root = ""              # type: str
         self.directives = {'whitespace': self.WHITESPACE['horizontal'],
                            'comment': '',

DHParser/testing.py

@@ -98,6 +98,10 @@ def get_report(test_unit):
     """
     Returns a text-report of the results of a grammar unit test.
     """
+    def indent(txt):
+        lines = txt.split('\n')
+        lines[0] = '    ' + lines[0]
+        return "\n    ".join(lines)
     report = []
     for parser_name, tests in test_unit.items():
         heading = 'Test of parser: "%s"' % parser_name
@@ -106,7 +110,7 @@ def get_report(test_unit):
             heading = 'Match-test "%s"' % test_name
             report.append('\n%s\n%s\n' % (heading, '-' * len(heading)))
             report.append('### Test-code:')
-            report.append(test_code)
+            report.append(indent(test_code))
             error = tests.get('__err__', {}).get(test_name, "")
             if error:
                 report.append('\n### Error:')
@@ -115,10 +119,10 @@ def get_report(test_unit):
             cst = tests.get('__cst__', {}).get(test_name, None)
             if cst and (not ast or cst == ast):
                 report.append('\n### CST')
-                report.append(cst.as_sxpr())
+                report.append(indent(cst.as_sxpr()))
             elif ast:
                 report.append('\n### AST')
-                report.append(ast.as_sxpr())
+                report.append(indent(ast.as_sxpr()))
     return '\n'.join(report)
 
 
@@ -199,7 +203,7 @@ def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, ve
         report_dir = "REPORT"
         if not os.path.exists(report_dir):
             os.mkdir(report_dir)
-        with open(os.path.join(report_dir, unit_name + '.report'), 'w') as f:
+        with open(os.path.join(report_dir, unit_name + '.md'), 'w') as f:
             f.write(get_report(test_unit))
 
     return errata

DHParser/toolkit.py

@@ -314,7 +314,7 @@ def expand_table(compact_table):
         value = compact_table[key]
         for k in smart_list(key):
             if k in expanded_table:
-                raise KeyError("Key %s used more than once in compact table!" % key)
+                raise KeyError('Key "%s" used more than once in compact table!' % key)
             expanded_table[k] = value
     return expanded_table
 

examples/LaTeX/LaTeX.ebnf

@@ -102,7 +102,7 @@ known_command       = footnote | includegraphics | caption
 generic_command     = !no_command CMDNAME [[ //~ config ] //~ block ]
 
 footnote            = "\footnote" block_of_paragraphs
-includegraphics     = "\includegraphics" config block
+includegraphics     = "\includegraphics" [ config ] block
 caption             = "\caption" block
 
 

examples/LaTeX/LaTeXCompiler.py

@@ -346,7 +346,7 @@ LaTeX_AST_transformation_table = {
     "block_environment": replace_by_single_child,
     "known_environment": replace_by_single_child,
     "generic_block": [],
-    "begin_generic_block, end_generic_block": reduce_single_child,
+    "begin_generic_block, end_generic_block": replace_by_single_child,
     "itemize, enumerate": [remove_brackets, flatten],
     "item": [remove_first],
     "figure": [],
@@ -361,7 +361,7 @@ LaTeX_AST_transformation_table = {
     "inline_environment": replace_by_single_child,
     "known_inline_env": replace_by_single_child,
     "generic_inline_env": [],
-    "begin_inline_env, end_inline_env": [reduce_single_child],
+    "begin_inline_env, end_inline_env": [replace_by_single_child],
     "begin_environment, end_environment": [remove_brackets, reduce_single_child],
     "inline_math": [remove_brackets, reduce_single_child],
     "command": [],
@@ -372,11 +372,9 @@ LaTeX_AST_transformation_table = {
     "caption": [],
     "config": [remove_brackets],
     "block": [remove_brackets, reduce_single_child(is_anonymous)],
-    "text": [reduce_single_child, merge_children('word_sequence', ':Whitespace', 'TEXTCHUNK')],
-    "cfgtext": [flatten, reduce_single_child, replace_parser('text')],
-    "word_sequence": collapse,  # [flatten, merge_children('TEXTCHUNK', ':Whitespace'), reduce_single_child],
-    "no_command": [],
-    "blockcmd": [],
+    "text": collapse,
+    "cfgtext, word_sequence": [],
+    "no_command, blockcmd": [],
     "structural": [],
     "CMDNAME": [remove_whitespace, reduce_single_child(is_anonymous)],
     "NAME": [reduce_single_child, remove_whitespace, reduce_single_child],

examples/LaTeX/grammar_tests/test_paragraph.ini

@@ -26,6 +26,7 @@
 8 : Unknwon \xy commands within paragraphs may be simple
     or \xy{complex}.
 
+
 [fail:paragraph]
 1 : \begin{enumerate}
 2 : \item

examples/LaTeX/grammar_tests/test_text.ini

+[match:text]
+1: Some plain text
+
+2: Text containing [ brackets ] is distinguished from cfgtext
+
+3: Text can also "escaped" characters like \&, \% or \_ .
+
+4: Text can contain both brackets [] and "esacped" characters \& .

examples/LaTeX/tst_LaTeX_grammar.py

@@ -35,7 +35,7 @@ if not DHParser.dsl.recompile_grammar('LaTeX.ebnf', force=True):  # recompiles G
 from DHParser import toolkit
 from LaTeXCompiler import get_grammar, get_transformer
 
-with toolkit.logging(True):
+with toolkit.logging(False):
     error_report = testing.grammar_suite('grammar_tests', get_grammar,
                                          get_transformer, report=True, verbose=True)
 if error_report: