You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@daffodil.apache.org by "tuxji (via GitHub)" <gi...@apache.org> on 2023/04/01 18:39:24 UTC

[GitHub] [daffodil] tuxji opened a new pull request, #998: Add left over data check to generated C parsers

tuxji opened a new pull request, #998:
URL: https://github.com/apache/daffodil/pull/998

   Fix generated C parsers not checking for left over data after parse calls and not clearing infoset between parse calls.  Rename some infoset C functions more consistently (C snake style).
   
   Also enhance C code generator to support DFDL schemas containing simple type root elements, allowing generated C parsers to parse data and infosets containing only a single element.  Add examples and new tests for simple root elements as well.
   
   Also replace any non-alphabetical / non-numerical characters with underscores when converting XML identifiers to C identifiers, allowing XML element names like "simple-boolean" to become "simple_boolean" in generated C code.
   
   DAFFODIL-2807
   
   Main.scala: Avoid second exception if TDMLRunner throws a NullPointerException and Main tries to print its message (NPEs don't have a message).
   
   testNonCompatibleImplementation.tdml: Both ibm and daffodilC can use same schema with single element now.  Merge s1 into s2 and remove s2.
   
   TestCLItdml.scala: Remove unnecessary "-iii" options from CLI test.
   
   daffodil_main.c: Call `get_infoset(CLEAR_INFOSET)` instead of `rootElement()`, call `parse_data(infoset, &pstate)` instead of `root->erd->parseSelf(root, &pstate)`, call `walk_infoset` instead of `walkInfoset`, and call `unparse_infoset(infoset, &ustate)` instead of `root->erd->unparseSelf(root, &ustate)`, showing how you can clear the infoset between `parse_data` calls if you call the C parser in a loop.
   
   xml_reader.h: Remove unused `root` field from XMLReader struct.
   
   errors.c: Add error message for new error `ERR_LEFTOVER_DATA`.
   
   errors.h: Define new error `ERR_LEFTOVER_DATA` and remove UNUSED macro (define it in infoset.h instead).
   
   infoset.c: Rename some infoset C functions more consistently (C snake style).  Define new functions `parse_data` and `unparse_infoset` and make them call `check_pstate` and `flush_ustate` so user will only need to call `get_infoset(CLEAR_INFOSET)` and `parse_data` to get correct C parser behavior.  Also rename ERD field `offsets` to `childrenOffsets`.  Remove flushUstate function (define it as flush_ustate in unparsers.c instead).
   
   infoset.h: Rename ERD field `offsets` to `childrenOffsets`.  Rename function `rootElement(void)` to `get_infoset(clear_infoset)`.  Declare new functions `parse_data` and `unparse_infoset`.  Rename walkInfoset to walk_infoset (snake style).  Define UNUSED macro here to let extras.c use it.
   
   parsers.[ch]: Define function `check_pstate` to check for leftover data (called by `parse_data`).
   
   unparsers.[ch]: Define function `flush_ustate` to flush unwritten fractional bits (called by `unparse_infoset`).
   
   bits.c: Call `flush_ustate` instead of `flushUstate`.
   
   extras.c: Implement `get_infoset` instead of `rootElement`.
   
   DaffodilCExamplesGenerator.scala: Generate a C example with a simple root element too.
   
   BinaryBooleanCodeGenerator.scala: Convert XML name to C name.
   
   BinaryValueCodeGenerator.scala: Convert XML name to C name.
   
   CodeGeneratorState.scala: Detect the case when a root element is a simple type and translate it as a hybrid complex and simple type (that is, push state for a complex element on the stack and when popping that state, generate the hybrid offset computations and ERD declarations needed to parse the simple type root element successfully).  Define new `cName` method to convert XML names to C names.  Generate C function `get_infoset(clear_infoset)` instead of `rootElement(void)` to clear infoset between parses.  Use renamed ERD field `childrenOffsets` instead of `offsets`.
   
   HexBinaryCodeGenerator.scala: Convert XML name to C name.
   
   examples/**: Regenerate C examples due to above changes and add a C example for a simple root element schema.
   
   data/simple*.dat: Add new data files to test simple root elements.
   
   infosets/simple*.dat.xml: Add simple root element infosets.
   
   simple.dfdl.xsd: Add new DFDL schema with choice of simple root elements to test (one root element for each primitive type).
   
   simple.tdml: Add test cases for all simple root elements listed in simple.dfdl.xsd.
   
   simple-errors.tdml: Add error test cases for simple root elements listed in simple.dfdl.xsd, although don't need to test every integer type.
   
   TestSimple.scala: Add unit tests for all test cases listed in simple.tdml.
   
   TestSimpleErrors.scala: Add unit tests for all test cases listed in simple-errors.tdml.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@daffodil.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org