selectolax 0.3.29__tar.gz → 0.3.31__tar.gz

{selectolax-0.3.29 → selectolax-0.3.31}/CHANGES.rst

@@ -1,6 +1,25 @@
 selectolax Changelog
 ====================
 
+Version 0.3.31
+-------------
+
+Released
+
+- Improve type hints, add docstrings to type hints
+- Prevent decomposing of the root node
+- Unpin Cython version and make it Optional
+- Allow empty attribute values. Fixes #165.
+
+Version 0.3.30
+-------------
+
+Released
+
+- Update lexbor
+- Expose `SelectolaxError` exception in lexbor.pyi
+
+
 Version 0.3.29
 -------------
 

{selectolax-0.3.29/selectolax.egg-info → selectolax-0.3.31}/PKG-INFO

@@ -1,42 +1,31 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.1
 Name: selectolax
-Version: 0.3.29
+Version: 0.3.31
 Summary: Fast HTML5 parser with CSS selectors.
 Home-page: https://github.com/rushter/selectolax
 Author: Artem Golubin
-Author-email: me@rushter.com
-License: MIT license
-Project-URL: Source code, https://github.com/rushter/selectolax
-Keywords: selectolax
+Author-email: Artem Golubin <me@rushter.com>
+License: MIT
+Project-URL: Repository, https://github.com/rushter/selectolax
+Project-URL: Documentation, https://selectolax.readthedocs.io/en/latest/parser.html
+Project-URL: Changelog, https://github.com/rushter/selectolax/blob/main/CHANGES.rst
+Keywords: selectolax,html,parser,css,fast
 Classifier: Development Status :: 5 - Production/Stable
 Classifier: Topic :: Text Processing :: Markup :: HTML
 Classifier: Topic :: Internet
 Classifier: Topic :: Internet :: WWW/HTTP
 Classifier: Intended Audience :: Developers
-Classifier: License :: OSI Approved :: MIT License
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.7
-Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
-License-File: LICENSE
+Requires-Python: >=3.9
+Description-Content-Type: text/x-rst
 Provides-Extra: cython
-Requires-Dist: Cython==3.0.11; extra == "cython"
-Dynamic: author
-Dynamic: author-email
-Dynamic: classifier
-Dynamic: description
-Dynamic: home-page
-Dynamic: keywords
-Dynamic: license
-Dynamic: license-file
-Dynamic: project-url
-Dynamic: provides-extra
-Dynamic: summary
+License-File: LICENSE
 
 .. image:: docs/logo.png
   :alt: selectolax logo

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/README.md

@@ -232,7 +232,7 @@ Alexander Borisov <borisov@lexbor.com>
 
    Lexbor.
 
-   Copyright 2018-2024 Alexander Borisov
+   Copyright 2018-2025 Alexander Borisov
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/core/conv.c

@@ -305,13 +305,18 @@ done:
 }
 
 size_t
-lexbor_conv_dec_to_hex(uint32_t number, lxb_char_t *out, size_t length)
+lexbor_conv_dec_to_hex(uint32_t number, lxb_char_t *out, size_t length,
+                       bool upper)
 {
     lxb_char_t c;
     size_t len;
     uint32_t tmp;
+    const lxb_char_t *map_str;
 
-    static const lxb_char_t map_str[] = "0123456789abcdef";
+    static const lxb_char_t map_str_l[] = "0123456789abcdef";
+    static const lxb_char_t map_str_u[] = "0123456789ABCDEF";
+
+    map_str = (upper) ? map_str_u : map_str_l;
 
     if(number != 0) {
         tmp = number;

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/core/conv.h

@@ -37,7 +37,8 @@ LXB_API unsigned
 lexbor_conv_data_to_uint(const lxb_char_t **data, size_t length);
 
 LXB_API size_t
-lexbor_conv_dec_to_hex(uint32_t number, lxb_char_t *out, size_t length);
+lexbor_conv_dec_to_hex(uint32_t number, lxb_char_t *out, size_t length,
+                       bool upper);
 
 lxb_inline long
 lexbor_conv_double_to_long(double number)

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/base.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2019-2024 Alexander Borisov
+ * Copyright (C) 2019-2025 Alexander Borisov
  *
  * Author: Alexander Borisov <borisov@lexbor.com>
  */
@@ -18,7 +18,7 @@ extern "C" {
 
 
 #define LXB_CSS_VERSION_MAJOR 1
-#define LXB_CSS_VERSION_MINOR 2
+#define LXB_CSS_VERSION_MINOR 3
 #define LXB_CSS_VERSION_PATCH 0
 
 #define LXB_CSS_VERSION_STRING                                                 \

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/declaration.c

@@ -70,12 +70,6 @@ lxb_css_declaration_list_prepare(lxb_css_parser_t *parser,
     parser->tkz->with_comment = false;
     parser->stage = LXB_CSS_PARSER_RUN;
 
-    parser->chunk_cb = parser->tkz->chunk_cb;
-    parser->chunk_ctx = parser->tkz->chunk_ctx;
-
-    parser->tkz->chunk_cb = lxb_css_syntax_parser_tkz_cb;
-    parser->tkz->chunk_ctx = parser;
-
     return LXB_STATUS_OK;
 }
 
@@ -118,9 +112,6 @@ failed:
 void
 lxb_css_declaration_list_finish(lxb_css_parser_t *parser)
 {
-    parser->tkz->chunk_cb = parser->chunk_cb;
-    parser->tkz->chunk_ctx = parser->chunk_ctx;
-
     parser->memory = parser->old_memory;
     parser->stage = LXB_CSS_PARSER_END;
 }

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/parser.h

@@ -156,9 +156,6 @@ struct lxb_css_parser {
     lxb_css_syntax_token_type_t      *types_end;
     lxb_css_syntax_token_type_t      *types_pos;
 
-    lxb_css_syntax_tokenizer_chunk_f chunk_cb;
-    void                             *chunk_ctx;
-
     const lxb_char_t                 *pos;
     uintptr_t                        offset;
 

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/property/state.c

@@ -396,7 +396,7 @@ lxb_css_property_state_angle(lxb_css_parser_t *parser,
         return false;
     }
 
-    unit = lxb_css_unit_angel_by_name(lxb_css_syntax_token_dimension(token)->str.data,
+    unit = lxb_css_unit_angle_by_name(lxb_css_syntax_token_dimension(token)->str.data,
                                       lxb_css_syntax_token_dimension(token)->str.length);
     if (unit == NULL) {
         return false;
@@ -404,7 +404,7 @@ lxb_css_property_state_angle(lxb_css_parser_t *parser,
 
     angle->num = lxb_css_syntax_token_dimension(token)->num.num;
     angle->is_float = lxb_css_syntax_token_dimension(token)->num.is_float;
-    angle->unit = (lxb_css_unit_angel_t) unit->unique;
+    angle->unit = (lxb_css_unit_angle_t) unit->unique;
 
     lxb_css_syntax_parser_consume(parser);
 
@@ -450,7 +450,7 @@ lxb_css_property_state_hue(lxb_css_parser_t *parser,
 
     switch (token->type) {
         case LXB_CSS_SYNTAX_TOKEN_DIMENSION:
-            unit = lxb_css_unit_angel_by_name(lxb_css_syntax_token_dimension(token)->str.data,
+            unit = lxb_css_unit_angle_by_name(lxb_css_syntax_token_dimension(token)->str.data,
                                               lxb_css_syntax_token_dimension(token)->str.length);
             if (unit == NULL) {
                 return false;
@@ -459,7 +459,7 @@ lxb_css_property_state_hue(lxb_css_parser_t *parser,
             hue->type = LXB_CSS_VALUE__ANGLE;
             hue->u.angle.num = lxb_css_syntax_token_dimension(token)->num.num;
             hue->u.angle.is_float = lxb_css_syntax_token_dimension(token)->num.is_float;
-            hue->u.angle.unit = (lxb_css_unit_angel_t) unit->unique;
+            hue->u.angle.unit = (lxb_css_unit_angle_t) unit->unique;
             break;
 
         case LXB_CSS_SYNTAX_TOKEN_NUMBER:
@@ -3313,7 +3313,7 @@ lxb_css_property_state_font_style(lxb_css_parser_t *parser,
                 return lxb_css_parser_success(parser);
             }
             else {
-                fs->angle.unit = (lxb_css_unit_angel_t) LXB_CSS_UNIT__UNDEF;
+                fs->angle.unit = (lxb_css_unit_angle_t) LXB_CSS_UNIT__UNDEF;
             }
 
             break;

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/property.c

@@ -1941,7 +1941,7 @@ lxb_css_property_font_style_serialize(const void *style,
         return status;
     }
 
-    if (fs->angle.unit != (lxb_css_unit_angel_t) LXB_CSS_UNIT__UNDEF) {
+    if (fs->angle.unit != (lxb_css_unit_angle_t) LXB_CSS_UNIT__UNDEF) {
         lexbor_serialize_write(cb, str_ws.data, str_ws.length, ctx, status);
 
         status = lxb_css_value_angle_sr(&fs->angle, cb, ctx);

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/stylesheet.c

@@ -181,12 +181,6 @@ lxb_css_stylesheet_prepare(lxb_css_parser_t *parser, lxb_css_memory_t *memory,
     parser->tkz->with_comment = false;
     parser->stage = LXB_CSS_PARSER_RUN;
 
-    parser->chunk_cb = parser->tkz->chunk_cb;
-    parser->chunk_ctx = parser->tkz->chunk_ctx;
-
-    parser->tkz->chunk_cb = lxb_css_syntax_parser_tkz_cb;
-    parser->tkz->chunk_ctx = parser;
-
     return LXB_STATUS_OK;
 }
 
@@ -229,9 +223,6 @@ lxb_css_stylesheet_process(lxb_css_parser_t *parser,
 void
 lxb_css_stylesheet_finish(lxb_css_parser_t *parser)
 {
-    parser->tkz->chunk_cb = parser->chunk_cb;
-    parser->tkz->chunk_ctx = parser->chunk_ctx;
-
     parser->memory = parser->old_memory;
     parser->selectors = parser->old_selectors;
     parser->stage = LXB_CSS_PARSER_END;

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/syntax/base.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2018-2024 Alexander Borisov
+ * Copyright (C) 2018-2025 Alexander Borisov
  *
  * Author: Alexander Borisov <borisov@lexbor.com>
  */
@@ -17,7 +17,7 @@ extern "C" {
 
 
 #define LXB_CSS_SYNTAX_VERSION_MAJOR 1
-#define LXB_CSS_SYNTAX_VERSION_MINOR 2
+#define LXB_CSS_SYNTAX_VERSION_MINOR 3
 #define LXB_CSS_SYNTAX_VERSION_PATCH 0
 
 #define LXB_CSS_SYNTAX_VERSION_STRING                                          \

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/syntax/parser.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2020-2022 Alexander Borisov
+ * Copyright (C) 2020-2025 Alexander Borisov
  *
  * Author: Alexander Borisov <borisov@lexbor.com>
  */
@@ -175,47 +175,6 @@ lxb_css_syntax_parser_consume(lxb_css_parser_t *parser)
     }
 }
 
-lxb_status_t
-lxb_css_syntax_parser_tkz_cb(lxb_css_syntax_tokenizer_t *tkz,
-                             const lxb_char_t **data, const lxb_char_t **end,
-                             void *ctx)
-{
-    size_t length, size;
-    lxb_char_t *new_data;
-    lxb_status_t status;
-    lxb_css_parser_t *parser = ctx;
-
-    if (parser->pos == NULL) {
-        return parser->chunk_cb(tkz, data, end, parser->chunk_ctx);
-    }
-
-    length = (size_t) (*end - parser->pos);
-
-    if (SIZE_MAX - parser->str.length < length) {
-        return LXB_STATUS_ERROR_OVERFLOW;
-    }
-
-    if (parser->str.length + length >= parser->str_size) {
-        size = parser->str.length + length + 1;
-
-        new_data = lexbor_realloc(parser->str.data, size);
-        if (new_data == NULL) {
-            return LXB_STATUS_ERROR_MEMORY_ALLOCATION;
-        }
-
-        parser->str.data = new_data;
-        parser->str_size = size;
-    }
-
-    memcpy(parser->str.data + parser->str.length, parser->pos, length);
-
-    status = parser->chunk_cb(tkz, data, end, parser->chunk_ctx);
-    parser->str.length += length;
-    parser->pos = *data;
-
-    return status;
-}
-
 lxb_css_syntax_rule_t *
 lxb_css_syntax_parser_list_rules_push(lxb_css_parser_t *parser,
                                       const lxb_css_syntax_token_t *token,

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/syntax/parser.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2020-2022 Alexander Borisov
+ * Copyright (C) 2020-2025 Alexander Borisov
  *
  * Author: Alexander Borisov <borisov@lexbor.com>
  */
@@ -84,11 +84,6 @@ lxb_css_syntax_parser_pipe_push(lxb_css_parser_t *parser,
                                 const lxb_css_syntax_cb_pipe_t *pipe,
                                 void *ctx, lxb_css_syntax_token_type_t stop);
 
-LXB_API lxb_status_t
-lxb_css_syntax_parser_tkz_cb(lxb_css_syntax_tokenizer_t *tkz,
-                             const lxb_char_t **data, const lxb_char_t **end,
-                             void *ctx);
-
 LXB_API const lxb_css_syntax_token_t *
 lxb_css_syntax_parser_start_block(lxb_css_parser_t *parser,
                                   const lxb_css_syntax_token_t *token,

{selectolax-0.3.29 → selectolax-0.3.31}/lexbor/source/lexbor/css/syntax/res.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2018-2019 Alexander Borisov
+ * Copyright (C) 2018-2025 Alexander Borisov
  *
  * Author: Alexander Borisov <borisov@lexbor.com>
  */
@@ -26,20 +26,20 @@ static const lxb_char_t lxb_css_syntax_res_name_map[256] =
     0x01, 0x00, 0x00, 0x00, 0x00, 0x01, 0x00, 0x01, 0x01, 0x01,
     0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
     0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+    0x01, 0x01, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+    0x00, 0x00, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
     0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+    0x01, 0x01, 0x01, 0x01, 0x01, 0x00, 0x01, 0x01, 0x01, 0x01,
     0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
     0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
-    0x01, 0x01, 0x01, 0x01, 0x01, 0x01
+    0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x00, 0x01, 0x01,
+    0x01, 0x01, 0x01, 0x01, 0x01
 };
 
 #endif /* LXB_CSS_SYNTAX_RES_NAME_MAP_ENABLED */