ssc_codegen 0.36.2__tar.gz → 0.37.0__tar.gz

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: ssc_codegen
-Version: 0.36.2
+Version: 0.37.0
 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-0.36.2 → ssc_codegen-0.37.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ssc_codegen"
-version = "0.36.2"
+version = "0.37.0"
 description = "Python-dsl code converter to html parser for web scraping "
 readme = "README.md"
 requires-python = ">=3.10"

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/ast/extract.py

@@ -27,8 +27,12 @@ class Raw(Node):
     Extracts raw HTML string from element(s).
     DOCUMENT → STRING, LIST_DOCUMENT → LIST_STRING.
     accept/ret set by builder from cursor type.
+
+    mode="outer" (default) — outer HTML (element tag included).
+    mode="inner" — inner HTML (children only).
     """
 
+    mode: str = "outer"
     accept_type_info: TypeInfo = field(
         default_factory=lambda: TypeInfo(base=VariableType.DOCUMENT)
     )

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/core/expressions.py

@@ -717,11 +717,31 @@ def _expr_raw(
     else:
         accept_ti = TypeInfo(base=VariableType.DOCUMENT)
         is_arr = False
+    mode = "outer"
+    if len(node.args) > 1:
+        lint.error(
+            node,
+            message="'raw' accepts at most 1 argument",
+            code="E001",
+            hint="example: raw  or  raw outer  or  raw inner",
+        )
+    elif node.args:
+        val = str(node.args[0].value)
+        if val not in ("outer", "inner"):
+            lint.error(
+                node,
+                message=f"invalid 'raw' mode {val!r} — expected 'outer' or 'inner'",
+                code="E001",
+                hint="example: raw  or  raw outer  or  raw inner",
+            )
+        else:
+            mode = val
     return Raw(
         parent=parent,
         accept_type_info=accept_ti,
         ret_type_info=TypeInfo(base=VariableType.STRING, is_array=is_arr),
         is_array=is_arr,
+        mode=mode,
     )
 
 

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/core/linter.py

@@ -87,7 +87,6 @@ _DEFINE_NAME_RE = _re.compile(r"^[A-Z_][A-Z0-9_-]*\Z")
 _NO_ARGS_OPS: frozenset[str] = frozenset(
     {
         "text",
-        "raw",
         "normalize-space",
         "lower",
         "upper",
@@ -1157,6 +1156,26 @@ def lint_pipeline_op(node: KdlNode, lint: LintContext) -> None:
     if name == "attr":
         lint_require_args(node, lint, min_count=1, example='attr "href"')
 
+    elif name == "raw":
+        args = _node_args(node)
+        if len(args) > 1:
+            lint.error(
+                node,
+                message="'raw' accepts at most 1 argument",
+                code="E001",
+                hint="example: raw  or  raw outer  or  raw inner",
+            )
+        elif args and args[0] not in ("outer", "inner"):
+            lint.error(
+                node,
+                message=(
+                    f"invalid 'raw' mode {args[0]!r}"
+                    " — expected 'outer' or 'inner'"
+                ),
+                code="E001",
+                hint="example: raw  or  raw outer  or  raw inner",
+            )
+
     elif name in _TRIM_OPS:
         args = _node_args(node)
         if len(args) > 1:

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/targets/golang/visitor.py

@@ -230,15 +230,17 @@ def _gofmt(source: str) -> str:
     if not gofmt:
         return source
     try:
+        # Pass UTF-8 explicitly — Windows text-mode defaults to the OEM
+        # codepage (cp1251/cp437), corrupting non-ASCII bytes in @doc
+        # comments and source strings.
         proc = subprocess.run(
             [gofmt],
-            input=source,
+            input=source.encode("utf-8"),
             capture_output=True,
-            text=True,
             timeout=10,
         )
         if proc.returncode == 0 and proc.stdout:
-            return proc.stdout
+            return proc.stdout.decode("utf-8")
     except (subprocess.SubprocessError, OSError):
         pass
     return source
@@ -1156,14 +1158,28 @@ class GoVisitor(BaseWalker):
         ]
 
     def visit_raw(self, node: Raw, ctx: WalkContext) -> list[str]:
+        # goquery: Html() method = inner HTML (jQuery semantics);
+        # goquery.OuterHtml(sel) package func = outer HTML.
+        if node.mode == "inner":
+            if not node.is_array:
+                return [
+                    f"{ctx.indent}_html, _ := {ctx.prv}.Html()",
+                    f"{ctx.indent}{ctx.nxt} := _html",
+                ]
+            return [
+                f"{ctx.indent}{ctx.nxt} := {ctx.prv}.Map(func(_ int, s *goquery.Selection) string {{",
+                f"{ctx.indent}\t_h, _ := s.Html()",
+                f"{ctx.indent}\treturn _h",
+                f"{ctx.indent}}})",
+            ]
         if not node.is_array:
             return [
-                f"{ctx.indent}_html, _ := {ctx.prv}.Html()",
+                f"{ctx.indent}_html, _ := goquery.OuterHtml({ctx.prv})",
                 f"{ctx.indent}{ctx.nxt} := _html",
             ]
         return [
             f"{ctx.indent}{ctx.nxt} := {ctx.prv}.Map(func(_ int, s *goquery.Selection) string {{",
-            f"{ctx.indent}\t_h, _ := s.Html()",
+            f"{ctx.indent}\t_h, _ := goquery.OuterHtml(s)",
             f"{ctx.indent}\treturn _h",
             f"{ctx.indent}}})",
         ]
@@ -1357,6 +1373,14 @@ class GoVisitor(BaseWalker):
         if at and at.base == VT.STRING and not at.is_array:
             idx = f"len({ctx.prv})-{abs(i)}" if i < 0 else str(i)
             return [f"{ctx.indent}{ctx.nxt} := string({ctx.prv}[{idx}])"]
+        if at and at.base == VT.DOCUMENT:
+            # *goquery.Selection is not indexable in Go — use .Eq().
+            # Negative index → Length() + i (i is negative, e.g. -1 → last).
+            if i < 0:
+                idx = f"{ctx.prv}.Length() + {i}"
+            else:
+                idx = str(i)
+            return [f"{ctx.indent}{ctx.nxt} := {ctx.prv}.Eq({idx})"]
         if i < 0:
             return [f"{ctx.indent}{ctx.nxt} := {ctx.prv}[len({ctx.prv})+{i}]"]
         return [f"{ctx.indent}{ctx.nxt} := {ctx.prv}[{i}]"]

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/targets/javascript/visitor.py

@@ -827,11 +827,10 @@ class JsVisitor(BaseWalker):
         ]
 
     def visit_raw(self, node: Raw, ctx: WalkContext) -> list[str]:
+        prop = "innerHTML" if node.mode == "inner" else "outerHTML"
         if not node.is_array:
-            return [f"{ctx.indent}let {ctx.nxt} = {ctx.prv}.outerHTML;"]
-        return [
-            f"{ctx.indent}let {ctx.nxt} = {ctx.prv}.map(el => el.outerHTML);"
-        ]
+            return [f"{ctx.indent}let {ctx.nxt} = {ctx.prv}.{prop};"]
+        return [f"{ctx.indent}let {ctx.nxt} = {ctx.prv}.map(el => el.{prop});"]
 
     def visit_attr(self, node: Attr, ctx: WalkContext) -> list[str]:
         keys = node.keys

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/targets/python/html_libs/bs4.py

@@ -116,6 +116,12 @@ class Bs4DomSpelling(DomSpelling):
         return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.text"]
 
     def raw(self, ctx: ConverterContext, node: Raw) -> list[str]:
+        if node.mode == "inner":
+            if node.accept_type_info.is_array:
+                return [
+                    f"{ctx.indent}{ctx.nxt} = [i.decode_contents() for i in {ctx.prv}]"
+                ]
+            return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.decode_contents()"]
         if node.accept_type_info.is_array:
             return [f"{ctx.indent}{ctx.nxt} = [str(i) for i in {ctx.prv}]"]
         return [f"{ctx.indent}{ctx.nxt} = str({ctx.prv})"]

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/targets/python/html_libs/lxml.py

@@ -165,6 +165,22 @@ class LxmlDomSpelling(DomSpelling):
         return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.text_content()"]
 
     def raw(self, ctx: ConverterContext, node: Raw) -> list[str]:
+        if node.mode == "inner":
+            self._builder.require_std(
+                "std_inner_html",
+                code="""
+                    def std_inner_html(el):
+                        parts = [el.text or ""]
+                        for child in el.iterchildren():
+                            parts.append(html.tostring(child, encoding="unicode"))
+                        return "".join(parts)
+                """,
+            )
+            if node.accept_type_info.is_array:
+                return [
+                    f"{ctx.indent}{ctx.nxt} = [std_inner_html(i) for i in {ctx.prv}]"
+                ]
+            return [f"{ctx.indent}{ctx.nxt} = std_inner_html({ctx.prv})"]
         if node.accept_type_info.is_array:
             return [
                 f"{ctx.indent}{ctx.nxt} = [html.tostring(i, encoding='unicode') for i in {ctx.prv}]"

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/targets/python/html_libs/parsel.py

@@ -159,6 +159,14 @@ class ParselDomSpelling(DomSpelling):
         ]
 
     def raw(self, ctx: ConverterContext, node: Raw) -> list[str]:
+        if node.mode == "inner":
+            if node.accept_type_info.is_array:
+                return [
+                    f"{ctx.indent}{ctx.nxt} = [''.join(i.xpath('node()').getall()) for i in {ctx.prv}]"
+                ]
+            return [
+                f"{ctx.indent}{ctx.nxt} = ''.join({ctx.prv}.xpath('node()').getall())"
+            ]
         if node.accept_type_info.is_array:
             return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.getall()"]
         return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.get()"]

{ssc_codegen-0.36.2 → ssc_codegen-0.37.0}/ssc_codegen/targets/python/html_libs/slax.py

@@ -119,6 +119,24 @@ class SlaxDomSpelling(DomSpelling):
         return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.text()"]
 
     def raw(self, ctx: ConverterContext, node: Raw) -> list[str]:
+        if node.mode == "inner":
+            self._builder.require_std(
+                "std_inner_html",
+                code="""
+                    def std_inner_html(node):
+                        parts = []
+                        child = node.child
+                        while child:
+                            parts.append(child.html)
+                            child = child.next
+                        return "".join(parts)
+                """,
+            )
+            if node.accept_type_info.is_array:
+                return [
+                    f"{ctx.indent}{ctx.nxt} = [std_inner_html(i) for i in {ctx.prv}]"
+                ]
+            return [f"{ctx.indent}{ctx.nxt} = std_inner_html({ctx.prv})"]
         if node.accept_type_info.is_array:
             return [f"{ctx.indent}{ctx.nxt} = [i.html for i in {ctx.prv}]"]
         return [f"{ctx.indent}{ctx.nxt} = {ctx.prv}.html"]