Feature/improvements to standards to code tool

tools/standardTextToCode/DummyReaderAndBuildFiles/CMakeLists.txt

@@ -0,0 +1,12 @@
+cmake_minimum_required(VERSION 3.10)
+project(ParsingTestApp)
+
+set(CMAKE_CXX_STANDARD 20)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+
+# Collect all .cpp files in the current directory
+file(GLOB SOURCES "*.cpp")
+
+# Change to create a library instead of executable
+# Remove the add_executable line and replace with:
+add_library(ParsingTestApp ${SOURCES})

tools/standardTextToCode/DummyReaderAndBuildFiles/SubByteReaderDummy.h

@@ -0,0 +1,24 @@
+
+#pragma once
+
+#include <string>
+
+namespace parser
+{
+
+class SubByteReaderDummy
+{
+public:
+  SubByteReaderDummy() = default;
+
+  bool byte_aligned() const { return false; }
+
+  bool        readFlag(const std::string &name) { return false; }
+  int         readBits(const std::string &name, const int numBits) { return 0; }
+  int         readBitsSigned(const std::string &name, const int numBits) { return 0; }
+  unsigned    readUEV(const std::string &name) { return 0; }
+  int         readSEV(const std::string &name) { return 0; }
+  std::string readString(const std::string &name) { return ""; }
+};
+
+} // namespace parser

tools/standardTextToCode/DummyReaderAndBuildFiles/Units.h

@@ -0,0 +1,22 @@
+#pragma once
+
+namespace parser
+{
+
+class NalUnit
+{
+public:
+  NalUnit() = default;
+
+  // This could contain file position info, nal unit type info, etc...
+};
+
+class SEI
+{
+public:
+  SEI() = default;
+
+  // This could contain a type as an enum etc...
+};
+
+} // namespace parser

tools/standardTextToCode/README.md

@@ -11,3 +11,7 @@ Read the `docx` document and parse all the tables (e.g `VPS`, `SPS`, ...). Put t
 ### writeTablesC++.py
 
 Take the pickled file and write CPP classes from it. This can be used as the starting point to create the actual parsing code. Or it can be used to add new parsing code which was added in a later version of a standard.
+
+### Prerequirements
+
+Install python-docx (https://github.com/python-openxml/python-docx).

tools/standardTextToCode/codingType.py

@@ -6,7 +6,11 @@ class Coding(Enum):
     UNSIGNED_FIXED = auto()    # u(x) - unsigned int with a fixed number of bits
     UNSIGNED_VARIABLE = auto() # u(v) - unsigned int with a variable number of bits
     UNSIGNED_EXP = auto()      # ue(v) - unsigned int wit exp golomb coding
+    SIGNED_FIXED = auto()      # i(x) - signed int with a fixed number of bits using twos complement
     SIGNED_EXP = auto()        # se(v) - signed int with exp golomb coding
+    BYTE = auto()              # b(8) - a byte (8 bits)
+    STRING = auto()            # st(x) - null terminated string in ISO/IEC 10646 UCS
+    UNKNOWN = auto()
 
 def isCodingType(descriptor : str):
     try:
@@ -20,30 +24,48 @@ def __init__(self, descriptor : str):
         self.length = 0
         self.codingType = None
         self.parseCodingType(descriptor)
+
     def parseCodingType(self, descriptor : str):
-        if (descriptor.startswith("f(")):
+        if descriptor.startswith("f("):
             self.codingType = Coding.FIXED_CODE
             self.length = int(descriptor[2:-1])
-        elif (descriptor.startswith("u(v)")):
+        elif descriptor.startswith("u(v)"):
             self.codingType = Coding.UNSIGNED_VARIABLE
-        elif (descriptor.startswith("u(")):
+        elif descriptor.startswith("u("):
             self.codingType = Coding.UNSIGNED_FIXED
             self.length = int(descriptor[2:-1])
-        elif (descriptor.startswith("ue(v)")):
+        elif descriptor.startswith("ue(v)"):
             self.codingType = Coding.UNSIGNED_EXP
-        elif (descriptor.startswith("se(v)")):
+        elif descriptor.startswith("se(v)"):
             self.codingType = Coding.SIGNED_EXP
+        elif descriptor.startswith("b(8)"):
+            # Used in e.g. VSEI standard
+            self.codingType = Coding.BYTE
+        elif descriptor.startswith("i("):
+            self.codingType = Coding.SIGNED_FIXED
+            self.length = int(descriptor[2:-1])
+        elif descriptor.startswith("st(v)"):
+            self.codingType = Coding.STRING
         else:
-            raise SyntaxError("Unknown descriptor type " + descriptor)
+            self.codingType = Coding.UNKNOWN
+
     def __str__(self):
-        if (self.codingType == Coding.FIXED_CODE):
+        if self.codingType == Coding.FIXED_CODE:
             return f"f({self.length})"
-        if (self.codingType == Coding.UNSIGNED_VARIABLE):
+        if self.codingType == Coding.UNSIGNED_VARIABLE:
             return "u(v)"
-        if (self.codingType == Coding.UNSIGNED_FIXED):
+        if self.codingType == Coding.UNSIGNED_FIXED:
             return f"u({self.length})"
-        if (self.codingType == Coding.UNSIGNED_EXP):
+        if self.codingType == Coding.UNSIGNED_EXP:
             return "ue(v)"
-        if (self.codingType == Coding.SIGNED_EXP):
+        if self.codingType == Coding.SIGNED_EXP:
             return "se(v)"
+        if self.codingType == Coding.BYTE:
+            return "b(8)"
+        if self.codingType == Coding.SIGNED_FIXED:
+            return f"i({self.length})"
+        if self.codingType == Coding.STRING:
+            return "st(v)"
+        if self.codingType == Coding.UNKNOWN:
+            return "unknown"
         return "Err"