reflow_tokenize_internal.h

Go to the documentation of this file.

#pragma once
 
#include <stddef.h>
#include <stdint.h>
 
#include "ra8_attributes.h"
#include "ra8_err.h"
#include "reflow.h"

typedef enum : uint32_t {
  k_priv_max_depth     = 64U,       
  k_priv_tag_name_cap  = 16U,       
  k_priv_entity_window = 12U,       
  k_priv_entity_min    = 4U,        
  k_priv_uc_2byte      = 0x80U,     
  k_priv_uc_3byte      = 0x800U,    
  k_priv_uc_4byte      = 0x10000U,  
  k_priv_uc_max        = 0x10FFFFU, 
  k_priv_utf8_lead2    = 0xC0U,     
  k_priv_utf8_lead3    = 0xE0U,     
  k_priv_utf8_lead4    = 0xF0U,     
  k_priv_utf8_cont     = 0x80U,     
  k_priv_utf8_mask     = 0x3FU,     
  k_priv_utf8_sh6      = 6U,        
  k_priv_utf8_sh12     = 12U,       
  k_priv_utf8_sh18     = 18U,       
  k_priv_base_dec      = 10U,       
  k_priv_base_hex      = 16U,       
  k_priv_hex_offset    = 10U,       
  k_priv_style_mask    = ((uint32_t)k_reflow_style_bold | (uint32_t)k_reflow_style_italic |
                          (uint32_t)k_reflow_style_underline), 
} priv_tok_consts_t;

RA8_PRIV bool priv_reflow_tok_is_xml_whitespace(char c);

RA8_PRIV reflow_html_tag_t priv_reflow_tok_classify(const char* name, size_t len);

RA8_PRIV bool
priv_reflow_tok_decode_entity(const char* src, size_t avail, uint32_t* out_cp, size_t* out_used);

RA8_PRIV size_t priv_reflow_tok_utf8_encode(uint32_t cp, uint8_t* dst);
 
/* ===========================================================================
 * Cross-TU helpers shared between reflow_tokenize.c and its companions
 * (reflow_tokenize_lex.c / reflow_tokenize_attr.c). Promoted from
 * file-local `static` so the markup engine can call the lexical + attribute
 * primitives. Production callers stay on the public API; only the defining
 * companion TU + the markup engine consume these.
 * ===========================================================================
 */

RA8_PRIV bool priv_reflow_tok_is_block(reflow_html_tag_t tag);

RA8_PRIV uint8_t priv_reflow_tok_style_for(reflow_html_tag_t tag);

RA8_PRIV bool priv_reflow_tok_emit(reflow_t*           engine,
                                   reflow_token_kind_t kind,
                                   reflow_html_tag_t   tag,
                                   uint8_t             style,
                                   uint32_t            off,
                                   uint32_t            len);

RA8_PRIV bool priv_reflow_tok_feed(reflow_t* engine, char ch, bool* last_ws);

RA8_PRIV bool priv_reflow_tok_stash_run(reflow_t*      engine,
                                        const uint8_t* buf,
                                        size_t         start,
                                        size_t         end,
                                        uint32_t*      out_off,
                                        uint32_t*      out_len);

RA8_PRIV bool
priv_reflow_tok_starts_with(const uint8_t* buf, size_t i, size_t len, const char* lit);

RA8_PRIV size_t priv_reflow_tok_skip_past(const uint8_t* buf,
                                          size_t         i,
                                          size_t         len,
                                          const char*    lit);

RA8_PRIV size_t priv_reflow_tok_find_lit(const uint8_t* buf, size_t i, size_t len, const char* lit);

RA8_PRIV bool priv_reflow_tok_find_attr(const uint8_t* tag,
                                        size_t         tag_len,
                                        const char*    name,
                                        size_t         name_len,
                                        size_t*        out_voff,
                                        size_t*        out_vlen);

RA8_PRIV void priv_reflow_tok_capture_attr(reflow_t*      engine,
                                           const uint8_t* tag,
                                           size_t         tag_len,
                                           const char*    name,
                                           size_t         name_len,
                                           uint32_t*      out_off,
                                           uint32_t*      out_len);

RA8_PRIV ra8_css_element_t priv_reflow_tok_css_element(reflow_html_tag_t kind,
                                                       const uint8_t*    tag,
                                                       size_t            span);

RA8_PRIV ra8_css_style_t priv_reflow_tok_css_inline(const uint8_t* tag, size_t span);

RA8_PRIV uint8_t priv_reflow_tok_intern_link(reflow_t* engine,
                                             uint32_t  href_off,
                                             uint32_t  href_len);

RA8_PRIV bool priv_reflow_tok_rel_is_stylesheet(const uint8_t* rel, size_t len);

RA8_PRIV uint8_t priv_reflow_tok_resolve_face_slot(const reflow_t*        engine,
                                                   const ra8_css_style_t* comp);

RA8_PRIV ra8_err_t priv_reflow_xml_walk(reflow_t*      engine,
                                        const uint8_t* xhtml_buf,
                                        size_t         xhtml_len);