llparse 0.1.1__tar.gz → 0.1.3__tar.gz

{llparse-0.1.1 → llparse-0.1.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: llparse
-Version: 0.1.1
+Version: 0.1.3
 Summary: A Parody of llparse written for writing C Parsers with Python
 Author-email: Vizonex <VizonexBusiness@gmail.com>
 Requires-Python: >=3.9
@@ -15,7 +15,7 @@ Dynamic: license-file
 
 A python parody of the typescript library llparse.
 
-I take no credit for the orginal work done by indutny and I was originally very nervous about making 
+I take no credit for the orginal work done by indutny and the other node-js contributors involved and I was originally very nervous about making 
 this python library that I made public... 
 
 Links to the original library 
@@ -49,7 +49,7 @@ do it at your own risk. They maybe incompleted or not throughly stress-tested.
 
 ## New Features
 - Throw me an issue if typescript llparse introduces something new that you want for me or another contributor to try and implement 
-just seeing llparse add new features is nothing but exciting to me.
+just seeing llparse add new features is exciting to me.
 
 - If you want a feature that typescript llparse doesn't have, be sure to try making a pull request over there as well and not just here, 
 there's a good chance they will appericate you for helping over there too and your helping make llhttp better by doing so. :)
@@ -62,7 +62,7 @@ there's a good chance they will appericate you for helping over there too and yo
 - Make it easy for me or someone else to find a problem and solve it in typescript after testing it in python
 - Typescript takes 2 commands to run a script with node-js it while python only takes one cutting the time required tremendously...
 - The orginal project was MIT licensed.
-- I wanted to write my own C Parsering tool with llhttp styled callbacks of my own using a language I was the most comfortable with using.
+- I wanted to write my own C Parser tool with llhttp styled callbacks of my own using a language I was the most comfortable with using.
 - I didn't like __Lemon Parser__ or __Yacc__ all that much and a good ide for handling them in Visual Studio Code with error checking to my knowlegde does not exist.
 - The closest thing I got to what I wanted was a project named __NMFU__ shorthand for no memory for you and even I had problems with writing things using that library...
 
@@ -131,3 +131,6 @@ print(c.c)
 open("http_parser.c", "w").write(c.c)
 open("http_parser.h", "w").write(c.header)
 ```
+
+## Video Showcasing this library
+- https://youtu.be/YQOzJ2BghQw

{llparse-0.1.1 → llparse-0.1.3}/README.md

@@ -5,7 +5,7 @@
 
 A python parody of the typescript library llparse.
 
-I take no credit for the orginal work done by indutny and I was originally very nervous about making 
+I take no credit for the orginal work done by indutny and the other node-js contributors involved and I was originally very nervous about making 
 this python library that I made public... 
 
 Links to the original library 
@@ -39,7 +39,7 @@ do it at your own risk. They maybe incompleted or not throughly stress-tested.
 
 ## New Features
 - Throw me an issue if typescript llparse introduces something new that you want for me or another contributor to try and implement 
-just seeing llparse add new features is nothing but exciting to me.
+just seeing llparse add new features is exciting to me.
 
 - If you want a feature that typescript llparse doesn't have, be sure to try making a pull request over there as well and not just here, 
 there's a good chance they will appericate you for helping over there too and your helping make llhttp better by doing so. :)
@@ -52,7 +52,7 @@ there's a good chance they will appericate you for helping over there too and yo
 - Make it easy for me or someone else to find a problem and solve it in typescript after testing it in python
 - Typescript takes 2 commands to run a script with node-js it while python only takes one cutting the time required tremendously...
 - The orginal project was MIT licensed.
-- I wanted to write my own C Parsering tool with llhttp styled callbacks of my own using a language I was the most comfortable with using.
+- I wanted to write my own C Parser tool with llhttp styled callbacks of my own using a language I was the most comfortable with using.
 - I didn't like __Lemon Parser__ or __Yacc__ all that much and a good ide for handling them in Visual Studio Code with error checking to my knowlegde does not exist.
 - The closest thing I got to what I wanted was a project named __NMFU__ shorthand for no memory for you and even I had problems with writing things using that library...
 
@@ -121,3 +121,6 @@ print(c.c)
 open("http_parser.c", "w").write(c.c)
 open("http_parser.h", "w").write(c.header)
 ```
+
+## Video Showcasing this library
+- https://youtu.be/YQOzJ2BghQw

{llparse-0.1.1 → llparse-0.1.3}/llparse/C_compiler.py

@@ -29,32 +29,32 @@ class CCompiler:
         out.append("#include <string.h>")
         out.append("")
         # Seems LLParse was updated from /* UNREACHABLE */ abort(); to a Macro, Intresting...
-        out.append('#ifdef __SSE4_2__')
-        out.append(' #ifdef _MSC_VER')
-        out.append('  #include <nmmintrin.h>')
-        out.append(' #else  /* !_MSC_VER */')
-        out.append('  #include <x86intrin.h>')
-        out.append(' #endif  /* _MSC_VER */')
-        out.append('#endif  /* __SSE4_2__ */')
-        out.append('')
-
-        out.append('#ifdef __ARM_NEON__')
-        out.append(' #include <arm_neon.h>')
-        out.append('#endif  /* __ARM_NEON__ */')
-        out.append('')
-
-        out.append('#ifdef __wasm__')
-        out.append(' #include <wasm_simd128.h>')
-        out.append('#endif  /* __wasm__ */')
-        out.append('')
-
-        out.append('#ifdef _MSC_VER')
-        out.append(' #define ALIGN(n) _declspec(align(n))')
-        out.append(' #define UNREACHABLE __assume(0)')
-        out.append('#else  /* !_MSC_VER */')
-        out.append(' #define ALIGN(n) __attribute__((aligned(n)))')
-        out.append(' #define UNREACHABLE __builtin_unreachable()')
-        out.append('#endif  /* _MSC_VER */')
+        out.append("#ifdef __SSE4_2__")
+        out.append(" #ifdef _MSC_VER")
+        out.append("  #include <nmmintrin.h>")
+        out.append(" #else  /* !_MSC_VER */")
+        out.append("  #include <x86intrin.h>")
+        out.append(" #endif  /* _MSC_VER */")
+        out.append("#endif  /* __SSE4_2__ */")
+        out.append("")
+
+        out.append("#ifdef __ARM_NEON__")
+        out.append(" #include <arm_neon.h>")
+        out.append("#endif  /* __ARM_NEON__ */")
+        out.append("")
+
+        out.append("#ifdef __wasm__")
+        out.append(" #include <wasm_simd128.h>")
+        out.append("#endif  /* __wasm__ */")
+        out.append("")
+
+        out.append("#ifdef _MSC_VER")
+        out.append(" #define ALIGN(n) _declspec(align(n))")
+        out.append(" #define UNREACHABLE __assume(0)")
+        out.append("#else  /* !_MSC_VER */")
+        out.append(" #define ALIGN(n) __attribute__((aligned(n)))")
+        out.append(" #define UNREACHABLE __builtin_unreachable()")
+        out.append("#endif  /* _MSC_VER */")
 
         out.append("")
         out.append(
@@ -83,7 +83,7 @@ class CCompiler:
         out.append("}")
         out.append("")
 
-        # TODO (Vizonex) Make llparse_state_t's Name Optional and alterable incase mixed with 
+        # TODO (Vizonex) Make llparse_state_t's Name Optional and alterable incase mixed with
         # llhttp or another parser
         out.append(f"static llparse_state_t {info.prefix}__run(")
         out.append(f"    {info.prefix}_t* {ARG_STATE},")
@@ -182,9 +182,7 @@ class CCompiler:
             else:
                 # TODO (Vizonex) Merge lines 139 & 140 together in a future update
                 callback = (
-                    f"(({info.prefix}__span_cb)"
-                    + ctx.spanCbField(span.index)
-                    + f")"
+                    f"(({info.prefix}__span_cb)" + ctx.spanCbField(span.index) + f")"
                 )
 
             args = [ctx.stateArg(), posField, f"(const char*) {ctx.endPosArg()}"]

{llparse-0.1.1 → llparse-0.1.3}/llparse/compilator.py

@@ -170,7 +170,7 @@ class Load(Field):
     def doBuild(self, ctx: "Compilation", out: list[str]):
         out.append(f"return {self.field(ctx)};")
 
-    
+
 # BIG ONE
 
 
@@ -374,7 +374,6 @@ class Consume(Node):
         index = ctx.stateField(self.ref.field)
         ty = ctx.getFieldType(self.ref.field)
 
-
         if ty == "i64":
             pass
         elif ty == "i32":
@@ -412,7 +411,6 @@ class Empty(Node):
     def __init__(self, ref: _frontend.node.Empty) -> None:
         self.ref = ref
         super().__init__(ref)
-    
 
     def doBuild(self, out: list[str]):
         assert self.ref.otherwise
@@ -574,7 +572,7 @@ class Single(Node):
             else:
                 ch = f"'{chr(e.key)}'"
 
-            out.append(f"  case {ch}:" + "{")
+            out.append(f"  case {ch}: " + "{")
             tmp: list[str] = []
 
             # For now debug everything....
@@ -884,7 +882,7 @@ class Compilation:
     def buildStateEnum(self, out: list[str]):
         # TODO (Vizonex) Give out other names that you could pass as an enum statename
         # this is incase multiple llparse_state_e states are given to compile
-        # example would be mixing llhttp with some other source... 
+        # example would be mixing llhttp with some other source...
         out.append("enum llparse_state_e {")
         out.append(f"  {STATE_ERROR},")
         for stateName in self.stateDict.keys():
@@ -992,8 +990,7 @@ class Compilation:
             out.append(f"{LABEL_PREFIX}{name} : " + "{")
             for line in lines:
                 out.append(f"  {line}")
-            out.append("  /* UNREACHABLE */;")
-            out.append("  abort();")
+            out.append("  UNREACHABLE;")
             out.append("}")
 
     def buildInternalStates(self, out: list[str]):
@@ -1004,8 +1001,7 @@ class Compilation:
             out.append(f"{LABEL_PREFIX}{name}: " + "{")
             for line in lines:
                 out.append(f"  {line}")
-            out.append("  /* UNREACHABLE */;")
-            out.append("  abort();")
+            out.append("  UNREACHABLE;")
             out.append("}")
 
     def addState(self, state: str, lines: list[str]):

{llparse-0.1.1 → llparse-0.1.3}/llparse/cython_builder.py

@@ -2,6 +2,7 @@
 
 from contextlib import contextmanager
 from typing import Optional
+
 from .frontend import IFrontendResult
 from .pyfront.front import Match
 from .pyfront.nodes import Invoke

{llparse-0.1.1 → llparse-0.1.3}/llparse/debug.py

@@ -9,9 +9,6 @@ class Debugger:
 
         while queue:
             node = queue.pop()
-            print(node.name)
-            if node.name == "nmethods":
-                print(node.getEdges())
             if edges := node.getEdges():
                 for slot in edges:
                     if slot.node in nodes:

{llparse-0.1.1 → llparse-0.1.3}/llparse/dot.py

@@ -208,6 +208,3 @@ class Dot:
 
     def escape(self, value: str):
         return "'" + value.replace("\\", "\\$1").replace('"', "\\$1") + "'"
-
-
-# TODO FIX ALL BUFFERS BACK TO STRINGS!

llparse-0.1.3/llparse/errors.py

@@ -0,0 +1,2 @@
+class Error(Exception):
+    pass

{llparse-0.1.1 → llparse-0.1.3}/llparse/frontend.py

@@ -4,6 +4,7 @@ from typing import Literal, Optional, Union
 from .enumerator import Enumerator
 from .pybuilder import LoopChecker
 from .pybuilder import builder as source
+
 # from pyfront.namespace import code, node , transform
 from .pyfront import namespace as _frontend
 from .pyfront.front import Identifier, IWrap, SpanField
@@ -11,19 +12,20 @@ from .pyfront.implementation import IImplementation
 from .pyfront.nodes import ITableEdge
 from .pyfront.peephole import Peephole
 from .spanalloc import SpanAllocator
-from .trie import Trie, TrieEmpty, TrieNode, TrieSequence, TrieSingle
+from .trie import Trie, TrieEmpty, TrieNode, TrieSequence, TrieSingle, ITrieSingleChild
 
 DEFAULT_MIN_TABLE_SIZE = 32
 DEFAULT_MAX_TABLE_WIDTH = 4
 
+from logging import getLogger
+
+log = getLogger("llparse.frontend")
 
 
 WrappedNode = IWrap[_frontend.node.Node]
 WrappedCode = IWrap[_frontend.code.Code]
 
 
-
-
 @dataclass
 class ITableLookupTarget:
     trie: TrieEmpty
@@ -168,7 +170,7 @@ class Frontend:
         trieNode = trie.build(list(node))
 
         if not trieNode:
-            # print("[DEBUG]", "TrieNode was nonexistant")
+            log.debug("TrieNode was nonexistant")
             return self.implementation.node.Empty(
                 _frontend.node.Empty(self.Id.id(node.name))
             )
@@ -180,7 +182,7 @@ class Frontend:
 
         return children
 
-    def registerNode(self, node: WrappedNode):
+    def registerNode(self, node: WrappedNode) -> None:
         # NOTE NO Implementations required here since this is python!
         if isinstance(
             node.ref,
@@ -289,7 +291,6 @@ class Frontend:
 
             if isinstance(single.ref, _frontend.node.Invoke):
                 for edge in node:
-                    # print(edge.key)
                     single.ref.addEdge(
                         ord(edge.key) if isinstance(edge.key, str) else edge.key,
                         self.translate(edge.node),
@@ -306,52 +307,48 @@ class Frontend:
             return None
 
         targets: dict[source.code.Node, ITableLookupTarget] = {}
-        bailout = False
-        for child in trie.children:
-            if isinstance(child.node, TrieEmpty):
-                # print(
-                #     'non-leaf trie child of "%s" prevents table allocation' % node.name
-                # )
-                bailout = False
-                continue
-
-            empty: TrieEmpty = child.node
-            if getattr(empty, "value", None) is None:
-                # print(
-                #     'value passing trie leaf of "%s" prevents table allocation'
-                #     % node.name
-                # )
-                bailout = False
-                continue
+
+        def check_child(child: ITrieSingleChild):
+            nonlocal targets
+            if not isinstance(child.node, TrieEmpty):
+                log.debug(
+                    'non-leaf trie child of "%s" prevents table allocation' % node.name
+                )
+                return False
+            empty = child.node
+            if empty.value is not None:
+                log.debug(
+                    'value passing trie leaf of "%s" prevents table allocation'
+                    % node.name
+                )
+                return False
 
             target = empty.node
-            if not targets.get(target):
+            if target not in targets:
                 targets[target] = ITableLookupTarget(
                     keys=[child.key], noAdvance=child.noAdvance, trie=empty
                 )
-                bailout = True
-                break
+                return True
 
             existing = targets[target]
-
             if existing.noAdvance != child.noAdvance:
-                # print('noAdvance mismatch in a trie leaf of "%s" prevents table allocation' % node.name)
-                bailout = False
-                break
-
+                log.debug(
+                    f'noAdvance mismatch in a trie leaf of "{node.name}" prevents '
+                    "table allocation"
+                )
+                return False
             existing.keys.append(child.key)
+            return True
 
-            # TODO: see if breaking or continue block after out is breakout has been determined is good ot not...
-            bailout = True
-            break
-
-        # assert len(trie.children) == len(targets), "Something went wrong"
-        if bailout:
+        if not all([check_child(child) for child in trie.children]):
             return
 
         # Weave width limit for optimization...
-        if len(targets.keys()) >= (1 << self.options["maxTableElemWidth"]):
-            # print('too many different trie targets of "%s" for a table allocation' % node.name)
+        if len(targets) >= (1 << self.options["maxTableElemWidth"]):
+            log.debug(
+                'too many different trie targets of "%s" for a table allocation'
+                % node.name
+            )
             return
 
         table = self.implementation.node.TableLookup(
@@ -360,11 +357,11 @@ class Frontend:
         children.append(table)
 
         # Break Loop
-        if self.Map.get(node):
+        if not self.Map.get(node):
             self.Map[node] = table
 
         for target in targets.values():
-            _next = self.translateTrie(node, target, children)
+            _next = self.translateTrie(node, target.trie, children)
             table.ref.addEdge(
                 ITableEdge(keys=target.keys, noAdvance=target.noAdvance, node=_next)
             )
@@ -410,9 +407,7 @@ class Frontend:
         self, node: source.code.Match, trie: TrieSingle, children: MatchChildren
     ):
         # Check if Tablelookup could be a valid option to Optimze our code up...
-        maybeTable = self.maybeTableLookup(node, trie, children)
-
-        if maybeTable:
+        if maybeTable := self.maybeTableLookup(node, trie, children):
             return maybeTable
 
         single = self.implementation.node.Single(
@@ -434,8 +429,7 @@ class Frontend:
                 value=child.node.value if isinstance(child.node, TrieEmpty) else None,
             )
 
-        otherwise = trie.otherwise
-        if otherwise:
+        if otherwise := trie.otherwise:
             single.ref.setOtherwise(
                 self.translateTrie(node, otherwise, children), True, otherwise.value
             )
@@ -444,8 +438,9 @@ class Frontend:
     def translateSpanCode(self, code: source.code._Span):
         return self.translateCode(code)
 
-    # TODO Vizonex Maybe better typehining can be used in this function alone....
-    def translateCode(self, code: source.code.Code):
+    def translateCode(
+        self, code: source.code.Code
+    ):
         """Translates Builder Classes to Frontend Classes..."""
 
         prefixed = self.codeId.id(code.name).name
@@ -501,9 +496,8 @@ class Frontend:
         else:
             raise Exception(f'UnSupported code:"{code.name}" type: "{type(code)}"')
 
-        if self.codeCache.get(res.ref.cacheKey):
-            return self.codeCache[res.ref.cacheKey]
-
+        if _res := self.codeCache.get(res.ref.cacheKey):
+            return _res
         self.codeCache[res.ref.cacheKey] = res
         return res
 

{llparse-0.1.1 → llparse-0.1.3}/llparse/llparse.py

@@ -4,8 +4,13 @@ from dataclasses import dataclass
 from typing import Optional
 
 from .C_compiler import CCompiler
-from .frontend import (DEFAULT_MAX_TABLE_WIDTH, DEFAULT_MIN_TABLE_SIZE,
-                       Frontend, IImplementation, source)
+from .frontend import (
+    DEFAULT_MAX_TABLE_WIDTH,
+    DEFAULT_MIN_TABLE_SIZE,
+    Frontend,
+    IImplementation,
+    source,
+)
 from .header import HeaderBuilder
 
 
@@ -62,12 +67,20 @@ class Compiler:
         properties: list[source.Property],
         header_name: Optional[str] = None,
         Impl: Optional[IImplementation] = IImplementation(),
+        override_llparse_name: bool = False
     ):
         """Creates the C and header file..."""
         info = self.to_frontend(root, properties, Impl)
         hb = HeaderBuilder(self.prefix, self.headerGuard, properties, info.spans)
+        cdata =  CCompiler(header_name, self.debug).compile(info)
+        if override_llparse_name:
+            # sometimes users want to combine parsers together when compiling with C
+            # to make up for conflicts with other parsers example: llhttp
+            # there should be a fair way of compiling everything.
+            cdata = cdata.replace('llparse', self.prefix)
+
         return CompilerResult(
-            CCompiler(header_name, self.debug).compile(info), hb.build()
+           cdata , hb.build()
         )
 
 
@@ -117,6 +130,7 @@ class LLParse(source.Builder):
         maxTableElemWidth: Optional[int] = None,
         minTableSize: Optional[int] = None,
         header_name: Optional[str] = None,
+        override_llparse_name:bool = False
     ):
         """Builds Graph and then compiles the data into C code , returns with the header and C file inside of a Dataclass"""
 
@@ -128,7 +142,8 @@ class LLParse(source.Builder):
             minTableSize if minTableSize else DEFAULT_MIN_TABLE_SIZE,
         )
 
-        return compiler.compile(root, self.properties(), header_name=header_name)
+        return compiler.compile(root, self.properties(), header_name=header_name, override_llparse_name=override_llparse_name)
+    
 
     def to_frontend(
         self,