ssc_codegen 0.29.3__tar.gz → 0.29.5__tar.gz

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ssc_codegen
-Version: 0.29.3
+Version: 0.29.5
 Summary: Python-dsl code converter to html parser for web scraping 
 Project-URL: Documentation, https://github.com/vypivshiy/selector_schema_codegen#readme
 Project-URL: Issues, https://github.com/vypivshiy/selector_schema_codegen/issues

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ssc_codegen"
-version = "0.29.3"
+version = "0.29.5"
 description = "Python-dsl code converter to html parser for web scraping "
 readme = "README.md"
 requires-python = ">=3.10"

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/ast/struct.py

@@ -13,7 +13,7 @@ from .types import VariableType, StructType
 #   PRIM   = str | int | float | bool       (default: str)
 #   STYLE  = repeat | csv | bracket | pipe | space   (arrays only; default: repeat)
 # Legacy `{{name}}` remains valid (groups 2-5 = None → str, scalar, required).
-_PLACEHOLDER_RE = _re.compile(
+PLACEHOLDER_RE = _re.compile(
     r"\{\{"
     r"([A-Za-z][A-Za-z0-9_-]*)"
     r"(?::(str|int|float|bool))?"
@@ -25,7 +25,7 @@ _PLACEHOLDER_RE = _re.compile(
 
 # Widened pattern — any `{{…}}`-shaped token. Used by the linter to flag
 # malformed placeholders that the strict _PLACEHOLDER_RE would silently skip.
-_PLACEHOLDER_WIDE_RE = _re.compile(r"\{\{([^{}]*)\}\}")
+PLACEHOLDER_WIDE_RE = _re.compile(r"\{\{([^{}]*)\}\}")
 
 
 @dataclass
@@ -39,7 +39,7 @@ class PlaceholderSpec:
     style: str | None = None  # None == default "repeat" when is_array
 
 
-def _parse_placeholder(match: "_re.Match[str]") -> PlaceholderSpec:
+def parse_placeholder(match: "_re.Match[str]") -> PlaceholderSpec:
     return PlaceholderSpec(
         name=match.group(1),
         type_name=match.group(2) or "str",
@@ -383,8 +383,8 @@ class RequestConfig(Node):
         """Unique placeholders in declaration order (dedup by name)."""
         seen: set[str] = set()
         result: list[PlaceholderSpec] = []
-        for m in _PLACEHOLDER_RE.finditer(self.raw_payload):
-            spec = _parse_placeholder(m)
+        for m in PLACEHOLDER_RE.finditer(self.raw_payload):
+            spec = parse_placeholder(m)
             if spec.name not in seen:
                 seen.add(spec.name)
                 result.append(spec)

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/js_pure.py

@@ -33,7 +33,7 @@ from ssc_codegen.converters.request_spec import (
     normalize_placeholder_names,
     _PH,
 )
-from ssc_codegen.ast.struct import _parse_placeholder
+from ssc_codegen.ast.struct import parse_placeholder
 
 JS_CONVERTER = BaseConverter(indent=" " * 2)
 
@@ -1662,7 +1662,7 @@ def _js_array_join(ph: a.PlaceholderSpec) -> str:
 def _js_render_value(v: str) -> str:
     """Render a RequestSpec string value as a JS expression."""
     if m := _PH.fullmatch(v):
-        ph = _parse_placeholder(m)
+        ph = parse_placeholder(m)
         if ph.is_array and ph.style in ("csv", "pipe", "space"):
             return _js_array_join(ph)
         return ph.name
@@ -1685,7 +1685,7 @@ def _js_render_obj(d: dict[str, str]) -> str:
 
 def _js_dict_entry_placeholder(v: str) -> a.PlaceholderSpec | None:
     m = _PH.fullmatch(str(v))
-    return _parse_placeholder(m) if m else None
+    return parse_placeholder(m) if m else None
 
 
 def _js_dict_needs_builder(d: dict[str, str]) -> bool:

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/py_bs4.py

@@ -55,7 +55,7 @@ _BASE_UTILITY_LINES: list[str] = [
     "    pass",
     "",
     "def unescape_text(text: str) -> str:",
-    "    s = _html_unescape(text)",
+    "    s = ssc_html_unescape(text)",
     "    s = _RE_HEX_ENTITY.sub(lambda m: chr(int(m.group(1), 16)), s)",
     "    s = _RE_UNICODE_ENTITY.sub(lambda m: chr(int(m.group(1), 16)), s)",
     "    s = _RE_BYTES_ENTITY.sub(lambda m: chr(int(m.group(1), 16)), s)",
@@ -111,7 +111,7 @@ def runtime_module_content(node: a.Node) -> str:
         "import re",
         "import sys",
         "from typing import Dict",
-        "from html import unescape as _html_unescape",
+        "from html import unescape as ssc_html_unescape",
     ]
     has_rest = _module_has_rest(node)
     if has_rest:
@@ -252,7 +252,7 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         base_imports.extend(NOT_REQUIRED_IMPORT)
     else:
@@ -261,10 +261,12 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         if not module_is_rest_only(node):
-            base_imports.append("from html import unescape as _html_unescape")
+            base_imports.append(
+                "from html import unescape as ssc_html_unescape"
+            )
         base_imports.extend(NOT_REQUIRED_IMPORT)
         base_imports.extend(rest_imports(node))
 

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/py_lxml.py

@@ -36,7 +36,7 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
     else:
         base_imports = [
@@ -44,10 +44,12 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         if not py_bs4.module_is_rest_only(node):
-            base_imports.append("from html import unescape as _html_unescape")
+            base_imports.append(
+                "from html import unescape as ssc_html_unescape"
+            )
         base_imports.extend(py_bs4.rest_imports(node))
 
     base_imports.extend(py_bs4.NOT_REQUIRED_IMPORT)

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/py_parsel.py

@@ -28,7 +28,7 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         base_imports.extend(py_bs4.NOT_REQUIRED_IMPORT)
     else:
@@ -37,10 +37,12 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         if not py_bs4.module_is_rest_only(node):
-            base_imports.append("from html import unescape as _html_unescape")
+            base_imports.append(
+                "from html import unescape as ssc_html_unescape"
+            )
         base_imports.extend(py_bs4.NOT_REQUIRED_IMPORT)
         base_imports.extend(py_bs4.rest_imports(node))
 

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/py_render.py

@@ -7,7 +7,7 @@ fragments suitable for `requests(...)` / `httpx(...)` call arguments.
 import json
 import re
 
-from ssc_codegen.ast.struct import PlaceholderSpec, _parse_placeholder
+from ssc_codegen.ast.struct import PlaceholderSpec, parse_placeholder
 from ssc_codegen.converters.request_spec import _PH, RequestSpec
 
 __all__ = [
@@ -46,7 +46,7 @@ def render_value(v: str) -> str:
     ``Mozilla/5.0``        -> ``"Mozilla/5.0"``     (string literal)
     """
     if m := _PH.fullmatch(v):
-        ph = _parse_placeholder(m)
+        ph = parse_placeholder(m)
         if ph.is_array and ph.style in ("csv", "pipe", "space"):
             return _render_array_join(ph)
         return ph.name
@@ -94,7 +94,7 @@ def render_dict(d: dict[str, str], *, indent: str = "") -> str:
 def _dict_entry_placeholder(v: str) -> PlaceholderSpec | None:
     """Return PlaceholderSpec if *v* is a fullmatch placeholder, else None."""
     m = _PH.fullmatch(str(v))
-    return _parse_placeholder(m) if m else None
+    return parse_placeholder(m) if m else None
 
 
 def dict_needs_builder(d: dict[str, str]) -> bool:
@@ -160,7 +160,7 @@ def render_json_body(raw: str) -> str:
         if raw[i : i + 2] == "{{":
             m = _PH.match(raw, i)
             if m is not None:
-                name = _parse_placeholder(m).name
+                name = parse_placeholder(m).name
                 key = f"__SSC_PH_{len(sentinels)}__"
                 sentinels[key] = name
                 out.append(key if in_string else '"' + key + '"')

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/py_slax.py

@@ -27,7 +27,7 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         base_imports.extend(py_bs4.NOT_REQUIRED_IMPORT)
     else:
@@ -36,10 +36,12 @@ def pre_imports(node: a.Imports, ctx: ConverterContext):
             "import re",
             "import sys",
             "from dataclasses import dataclass",
-            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal",
+            "from typing import TypedDict, Optional, Any, List, Dict, Union, Literal, Mapping",
         ]
         if not py_bs4.module_is_rest_only(node):
-            base_imports.append("from html import unescape as _html_unescape")
+            base_imports.append(
+                "from html import unescape as ssc_html_unescape"
+            )
         base_imports.extend(py_bs4.NOT_REQUIRED_IMPORT)
         base_imports.extend(py_bs4.rest_imports(node))
 

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/converters/request_spec.py

@@ -12,7 +12,7 @@ from dataclasses import dataclass, field
 from typing import Callable
 from urllib.parse import urlparse, urlunparse
 
-from ssc_codegen.ast.struct import PlaceholderSpec, _parse_placeholder
+from ssc_codegen.ast.struct import PlaceholderSpec, parse_placeholder
 from ssc_codegen.parsers.curl import parse_curl_command
 from ssc_codegen.parsers.http import parse_http_request
 
@@ -48,7 +48,7 @@ class RequestSpec:
         result: list[PlaceholderSpec] = []
         for text in _iter_strings(self):
             for m in _PH.finditer(text):
-                spec = _parse_placeholder(m)
+                spec = parse_placeholder(m)
                 if spec.name not in seen:
                     seen.add(spec.name)
                     result.append(spec)

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/core/linting.py

@@ -6,6 +6,7 @@ import difflib as _difflib
 import re as _re
 
 from ssc_codegen.ast import JsonDefField, Module, VariableType
+from ssc_codegen.ast.struct import PLACEHOLDER_RE, PLACEHOLDER_WIDE_RE
 from ssc_codegen.kdl import KdlNode
 from ssc_codegen.core.contexts import (
     DefineKind,
@@ -799,6 +800,39 @@ def lint_reserved_field(
             )
 
 
+def lint_request_placeholders(
+    node: KdlNode, raw_payload: str, lint: LintContext
+) -> None:
+    """Validate placeholder names in an @request raw payload.
+
+    Convention: uppercase {{NAME}} is a define substitution (only resolved
+    within ``define`` values), lowercase {{name}} is a runtime fetch() param.
+    Define substitution does NOT run on @request payloads, so any remaining
+    uppercase tokens are bugs that silently become literal strings.
+    """
+    for m in PLACEHOLDER_WIDE_RE.finditer(raw_payload):
+        token = m.group(0)
+        strict = PLACEHOLDER_RE.match(token)
+        if strict is None:
+            lint.error(
+                node,
+                message=f"malformed placeholder {token!r} in @request",
+                code="E002",
+                hint="expected syntax: {{name}} or {{name:type}} (lowercase)",
+            )
+            continue
+        name = strict.group(1)
+        if name != name.lower():
+            lint.error(
+                node,
+                message=f"placeholder '{{{{{name}}}}}' in @request must be lowercase; "
+                f"uppercase names are define substitutions which don't resolve in @request",
+                code="E002",
+                hint=f"use lowercase for runtime params (e.g. {{{{{name.lower()}}}}}), "
+                f"or compose the URL in a define first",
+            )
+
+
 def lint_regular_field(
     node: KdlNode,
     field_name: str,

{ssc_codegen-0.29.3 → ssc_codegen-0.29.5}/ssc_codegen/core/struct_parser.py

@@ -32,6 +32,7 @@ from ssc_codegen.kdl import KdlArg, KdlNode
 
 from ssc_codegen.core.contexts import LintContext, ParseContext, WalkCtx
 from ssc_codegen.core.expressions import parse_expressions
+from ssc_codegen.core.linting import lint_request_placeholders
 from ssc_codegen.core.type_checking import check_pipeline_types
 
 
@@ -103,6 +104,7 @@ def parse_struct(
             )
             req = RequestConfig(parent=parent)
             req.raw_payload = raw_payload
+            lint_request_placeholders(node, raw_payload, lint)
             req.response_path = str(
                 node.properties.get(
                     "response-path",