ssc_codegen 0.37.0__tar.gz → 0.37.2__tar.gz

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: ssc_codegen
-Version: 0.37.0
+Version: 0.37.2
 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
@@ -86,6 +86,9 @@ ssc-gen generate js examples/ -o ./output
 # Go (goquery + net/http)
 ssc-gen generate go schema.kdl -o ./output
 
+# Go package defaults to main; override explicitly when generating a library
+ssc-gen generate go schema.kdl -o ./output --package scraper
+
 # with custom package name
 ssc-gen generate python schema.kdl -L bs4 -o ./parsers --package scraper
 
@@ -116,12 +119,14 @@ ssc-gen check examples/
 
 ```bash
 # from file
-ssc-gen run examples/booksToScrape.kdl:MainCatalogue -l python -L bs4 -i page.html
+ssc-gen run examples/booksToScrape.kdl:MainCatalogue -L bs4 -i page.html
 
 # from stdin
-curl https://books.toscrape.com/ | ssc-gen run examples/booksToScrape.kdl:MainCatalogue -l python -L bs4
+curl https://books.toscrape.com/ | ssc-gen run examples/booksToScrape.kdl:MainCatalogue -L bs4
 ```
 
+`ssc-gen run` executes generated Python in-process. Run only trusted schema files.
+
 ### Health check (verify selectors match elements)
 
 ```bash

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/README.md

@@ -51,8 +51,11 @@ ssc-gen generate python schema.kdl -L bs4 -o ./output
 # all .kdl files in a directory (JavaScript)
 ssc-gen generate js examples/ -o ./output
 
-# Go (goquery + net/http)
-ssc-gen generate go schema.kdl -o ./output
+# Go (goquery + net/http)
+ssc-gen generate go schema.kdl -o ./output
+
+# Go package defaults to main; override explicitly when generating a library
+ssc-gen generate go schema.kdl -o ./output --package scraper
 
 # with custom package name
 ssc-gen generate python schema.kdl -L bs4 -o ./parsers --package scraper
@@ -84,11 +87,13 @@ ssc-gen check examples/
 
 ```bash
 # from file
-ssc-gen run examples/booksToScrape.kdl:MainCatalogue -l python -L bs4 -i page.html
+ssc-gen run examples/booksToScrape.kdl:MainCatalogue -L bs4 -i page.html
 
 # from stdin
-curl https://books.toscrape.com/ | ssc-gen run examples/booksToScrape.kdl:MainCatalogue -l python -L bs4
-```
+curl https://books.toscrape.com/ | ssc-gen run examples/booksToScrape.kdl:MainCatalogue -L bs4
+```
+
+`ssc-gen run` executes generated Python in-process. Run only trusted schema files.
 
 ### Health check (verify selectors match elements)
 

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ssc_codegen"
-version = "0.37.0"
+version = "0.37.2"
 description = "Python-dsl code converter to html parser for web scraping "
 readme = "README.md"
 requires-python = ">=3.10"
@@ -78,10 +78,6 @@ exclude = ["__init__.py"]
 [tool.mypy]
 python_version = "3.10"
 show_error_codes = true
-exclude = [
-    "ssc_codegen/ast"  # ignore built-in ast and project ast
-
-]
 [[tool.mypy.overrides]]
 module="tests.*"
 disallow_untyped_defs = false
@@ -91,12 +87,6 @@ disallow_untyped_defs = false
 module = ["colorama", "lxml", "lxml.*"]
 ignore_missing_imports = true
 
-[[tool.mypy.overrides]]
-# Large visitor classes with match statements may need relaxed return checks
-module = ["ssc_codegen.targets.python.visitor", "ssc_codegen.targets.javascript.visitor"]
-disable_error_code = ["return", "return-value"]
-
-
 [tool.coverage.report]
 exclude_also = [
     'def __repr__',

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/ssc_codegen/ast/struct.py

@@ -136,10 +136,16 @@ class PlaceholderSpec:
 def parse_placeholder(match: "_re.Match[str]") -> PlaceholderSpec:
     return PlaceholderSpec(
         name=match.group(1),
-        type_name=match.group(2) or "str",
+        type_name=cast(
+            Literal["str", "int", "float", "bool"],
+            match.group(2) or "str",
+        ),
         is_array=bool(match.group(3)),
         is_optional=bool(match.group(4)),
-        style=match.group(5) or None,
+        style=cast(
+            Literal["repeat", "csv", "bracket", "pipe", "space"] | None,
+            match.group(5) or None,
+        ),
     )
 
 
@@ -260,7 +266,7 @@ class RequestHttp(Node):
 
     Created once at parse time from ``parse_to_http(raw_payload)``;
     converters read fields directly instead of re-parsing raw_payload.
-    String fields (url, headers, cookies, params, body) are ``Template``
+    String fields (url, headers, cookies, params, payload) are ``Template``
     instances: tokenized literal/placeholder parts.
     """
 
@@ -270,7 +276,7 @@ class RequestHttp(Node):
     cookies: dict[str, PlaceholderTemplate] = field(default_factory=dict)
     params: dict[str, PlaceholderTemplate] = field(default_factory=dict)
     body_kind: str = "empty"  # "empty" | "json" | "form" | "raw"
-    body: PlaceholderTemplate | dict[str, PlaceholderTemplate] | None = None
+    payload: PlaceholderTemplate | dict[str, PlaceholderTemplate] | None = None
 
     @property
     def placeholders(self) -> list[PlaceholderSpec]:
@@ -289,10 +295,10 @@ class RequestHttp(Node):
         yield self.url
         for d in (self.headers, self.cookies, self.params):
             yield from d.values()
-        if isinstance(self.body, PlaceholderTemplate):
-            yield self.body
-        elif isinstance(self.body, dict):
-            yield from self.body.values()
+        if isinstance(self.payload, PlaceholderTemplate):
+            yield self.payload
+        elif isinstance(self.payload, dict):
+            yield from self.payload.values()
 
     def with_renamed_placeholders(
         self, transform: Callable[[str], str]
@@ -312,13 +318,13 @@ class RequestHttp(Node):
         ) -> dict[str, PlaceholderTemplate]:
             return {k: v.renamed(mapping) for k, v in d.items()}
 
-        new_body: (
+        new_payload: (
             PlaceholderTemplate | dict[str, PlaceholderTemplate] | None
-        ) = self.body
-        if isinstance(self.body, PlaceholderTemplate):
-            new_body = self.body.renamed(mapping)
-        elif isinstance(self.body, dict):
-            new_body = _rename_dict(self.body)
+        ) = self.payload
+        if isinstance(self.payload, PlaceholderTemplate):
+            new_payload = self.payload.renamed(mapping)
+        elif isinstance(self.payload, dict):
+            new_payload = _rename_dict(self.payload)
 
         return RequestHttp(
             method=self.method,
@@ -327,7 +333,7 @@ class RequestHttp(Node):
             cookies=_rename_dict(self.cookies),
             params=_rename_dict(self.params),
             body_kind=self.body_kind,
-            body=new_body,
+            payload=new_payload,
         )
 
 

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/ssc_codegen/ast/typedef.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 from dataclasses import dataclass
+from typing import cast
 
 from .base import Node
 from .types import StructType
@@ -29,4 +30,4 @@ class TypeDef(Node):
 
     @property
     def fields(self) -> list[TypeDefField]:
-        return self.body
+        return cast(list[TypeDefField], self.body)

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/ssc_codegen/core/contexts.py

@@ -50,6 +50,7 @@ class ParseContext:
     structs: dict[str, StructBase] = field(default_factory=dict)
     json_defs: dict[str, JsonDef] = field(default_factory=dict)
     source_path: Path | None = None
+    node_source_paths: dict[int, Path] = field(default_factory=dict)
 
     def all_names(self) -> set[str]:
         return (
@@ -77,6 +78,7 @@ class LintContext:
     )
     _predicate_depth: int = field(default=0)
     _predicate_context: str = ""
+    node_source_paths: dict[int, Path] = field(default_factory=dict)
 
     def error(
         self,
@@ -88,18 +90,22 @@ class LintContext:
         label: str | None = None,
         notes: list[str] | None = None,
     ) -> None:
-        self.diagnostics.append(
-            ReadDiagnostic(
-                message=message,
-                severity=Severity.ERROR,
-                span=node.span,
-                path=self.path,
-                hint=hint,
-                code=code,
-                label=label,
-                notes=tuple(notes) if notes else (),
-            )
+        scope = self.path
+        source_path = self.node_source_paths.get(id(node))
+        diagnostic_notes = tuple(notes) if notes else ()
+        if source_path is not None and scope:
+            diagnostic_notes = (*diagnostic_notes, f"scope: {scope}")
+        diagnostic = ReadDiagnostic(
+            message=message,
+            severity=Severity.ERROR,
+            span=node.span,
+            path=str(source_path) if source_path is not None else scope,
+            hint=hint,
+            code=code,
+            label=label,
+            notes=diagnostic_notes,
         )
+        self.diagnostics.append(diagnostic)
 
     def warning(
         self,
@@ -111,18 +117,22 @@ class LintContext:
         label: str | None = None,
         notes: list[str] | None = None,
     ) -> None:
-        self.diagnostics.append(
-            ReadDiagnostic(
-                message=message,
-                severity=Severity.WARNING,
-                span=node.span,
-                path=self.path,
-                hint=hint,
-                code=code,
-                label=label,
-                notes=tuple(notes) if notes else (),
-            )
+        scope = self.path
+        source_path = self.node_source_paths.get(id(node))
+        diagnostic_notes = tuple(notes) if notes else ()
+        if source_path is not None and scope:
+            diagnostic_notes = (*diagnostic_notes, f"scope: {scope}")
+        diagnostic = ReadDiagnostic(
+            message=message,
+            severity=Severity.WARNING,
+            span=node.span,
+            path=str(source_path) if source_path is not None else scope,
+            hint=hint,
+            code=code,
+            label=label,
+            notes=diagnostic_notes,
         )
+        self.diagnostics.append(diagnostic)
 
     @property
     def path(self) -> str:

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/ssc_codegen/core/expressions.py

@@ -606,6 +606,9 @@ def _reg_expr(name: str):
 def _expr_css(
     node: KdlNode, parent: FieldLikeNode, ctx: ParseContext, lint: LintContext
 ):
+    if not node.children and not node.args:
+        lint.error(node, message="'css' requires a selector", code="E001")
+        return None
     if node.children:
         queries = [
             resolve_selector_child_name(c.name, ctx) for c in node.children
@@ -622,6 +625,9 @@ def _expr_css(
 def _expr_css_all(
     node: KdlNode, parent: FieldLikeNode, ctx: ParseContext, lint: LintContext
 ):
+    if not node.children and not node.args:
+        lint.error(node, message="'css-all' requires a selector", code="E001")
+        return None
     if node.children:
         queries = [
             resolve_selector_child_name(c.name, ctx) for c in node.children
@@ -638,6 +644,9 @@ def _expr_css_all(
 def _expr_xpath(
     node: KdlNode, parent: FieldLikeNode, ctx: ParseContext, lint: LintContext
 ):
+    if not node.children and not node.args:
+        lint.error(node, message="'xpath' requires an expression", code="E001")
+        return None
     if node.children:
         queries = [
             resolve_selector_child_name(c.name, ctx) for c in node.children
@@ -654,6 +663,11 @@ def _expr_xpath(
 def _expr_xpath_all(
     node: KdlNode, parent: FieldLikeNode, ctx: ParseContext, lint: LintContext
 ):
+    if not node.children and not node.args:
+        lint.error(
+            node, message="'xpath-all' requires an expression", code="E001"
+        )
+        return None
     if node.children:
         queries = [
             resolve_selector_child_name(c.name, ctx) for c in node.children

{ssc_codegen-0.37.0 → ssc_codegen-0.37.2}/ssc_codegen/core/format.py

@@ -41,6 +41,7 @@ def format_diagnostic(
     context_lines: int = 1,
 ) -> str:
     """Format a single diagnostic as a human-readable Rust-style string."""
+    filepath = _diagnostic_filepath(d, filepath)
     src = _resolve_src(src, filepath)
     renderer = _Renderer(use_color=_resolve_color(use_color))
     return renderer.render(
@@ -66,12 +67,14 @@ def format_diagnostics(
             [diagnostic_to_dict(d) for d in diagnostics], indent=2
         )
 
-    src = _resolve_src(src, filepath)
     renderer = _Renderer(use_color=_resolve_color(use_color))
 
     sections = [
         renderer.render(
-            d, src=src, filepath=filepath, context_lines=context_lines
+            d,
+            src=_resolve_src(src, _diagnostic_filepath(d, filepath)),
+            filepath=_diagnostic_filepath(d, filepath),
+            context_lines=context_lines,
         )
         for d in diagnostics
     ]
@@ -97,6 +100,19 @@ def diagnostic_to_dict(d: ReadDiagnostic) -> dict[str, Any]:
     }
 
 
+def _diagnostic_filepath(
+    diagnostic: ReadDiagnostic, fallback: str | Path | None
+) -> str | Path | None:
+    if diagnostic.path:
+        try:
+            path = Path(diagnostic.path)
+            if path.is_file():
+                return path
+        except OSError:
+            pass
+    return fallback
+
+
 # ── Renderer ──────────────────────────────────────────────────────────────────
 
 
@@ -271,7 +287,7 @@ class _Renderer:
         parts: list[str] = []
         eq = self._styled("=", color=self._dim)
 
-        if d.path:
+        if d.path and _diagnostic_filepath(d, None) is None:
             scope = self._styled("scope:", color=self._cyan, bold=True)
             parts.append(f"{pad} {eq} {scope} {d.path}")
 
@@ -280,8 +296,16 @@ class _Renderer:
             parts.append(f"{pad} {eq} {help_} {d.hint}")
 
         for note in d.notes:
-            note_label = self._styled("note:", color=self._magenta, bold=True)
-            parts.append(f"{pad} {eq} {note_label} {note}")
+            if note.startswith("scope: "):
+                scope_label = self._styled(
+                    "scope:", color=self._cyan, bold=True
+                )
+                parts.append(f"{pad} {eq} {scope_label} {note[7:]}")
+            else:
+                note_label = self._styled(
+                    "note:", color=self._magenta, bold=True
+                )
+                parts.append(f"{pad} {eq} {note_label} {note}")
 
         return parts