view release on metacpan or  search on metacpan

modules/CommonMark/src/blocks.c  view on Meta::CPAN

  finalize(parser, parser->root);

  // Limit total size of extra content created from reference links to
  // document size to avoid superlinear growth. Always allow 100KB.
  if (parser->total_size > 100000)
    parser->refmap->max_ref_size = parser->total_size;
  else
    parser->refmap->max_ref_size = 100000;

  process_inlines(parser->mem, parser->root, parser->refmap, parser->options);

  cmark_strbuf_free(&parser->content);

  return parser->root;
}

cmark_node *cmark_parse_file(FILE *f, int options) {
  unsigned char buffer[4096];
  cmark_parser *parser = cmark_parser_new(options);
  size_t bytes;
  cmark_node *document;

  while ((bytes = fread(buffer, 1, sizeof(buffer), f)) > 0) {
    bool eof = bytes < sizeof(buffer);
    S_parser_feed(parser, buffer, bytes, eof);
    if (eof) {
      break;
    }
  }

  document = cmark_parser_finish(parser);
  cmark_parser_free(parser);
  return document;
}

cmark_node *cmark_parse_document(const char *buffer, size_t len, int options) {
  cmark_parser *parser = cmark_parser_new(options);
  cmark_node *document;

  S_parser_feed(parser, (const unsigned char *)buffer, len, true);

  document = cmark_parser_finish(parser);
  cmark_parser_free(parser);
  return document;
}

void cmark_parser_feed(cmark_parser *parser, const char *buffer, size_t len) {
  S_parser_feed(parser, (const unsigned char *)buffer, len, false);
}

static void S_parser_feed(cmark_parser *parser, const unsigned char *buffer,
                          size_t len, bool eof) {
  const unsigned char *end = buffer + len;
  static const uint8_t repl[] = {239, 191, 189};

  if (len > UINT_MAX - parser->total_size)
    parser->total_size = UINT_MAX;
  else
    parser->total_size += (int)len;

  // Skip UTF-8 BOM if present; see #334
  if (parser->line_number == 0 && parser->column == 0 && len >= 3 &&
      *buffer == 0xEF && *(buffer + 1) == 0xBB &&
      *(buffer + 2) == 0xBF) {
    buffer += 3;
  } else if (parser->last_buffer_ended_with_cr && *buffer == '\n') {
    // skip NL if last buffer ended with CR ; see #117
    buffer++;
  }

  parser->last_buffer_ended_with_cr = false;
  while (buffer < end) {
    const unsigned char *eol;
    bufsize_t chunk_len;
    bool process = false;
    for (eol = buffer; eol < end; ++eol) {
      if (S_is_line_end_char(*eol)) {
        process = true;
        break;
      }
      if (*eol == '\0' && eol < end) {
        break;
      }
    }
    if (eol >= end && eof) {
      process = true;
    }

    chunk_len = (eol - buffer);
    if (process) {
      if (parser->linebuf.size > 0) {
        cmark_strbuf_put(&parser->linebuf, buffer, chunk_len);
        S_process_line(parser, parser->linebuf.ptr, parser->linebuf.size);
        cmark_strbuf_clear(&parser->linebuf);
      } else {
        S_process_line(parser, buffer, chunk_len);
      }
    } else {
      if (eol < end && *eol == '\0') {
        // omit NULL byte
        cmark_strbuf_put(&parser->linebuf, buffer, chunk_len);
        // add replacement character
        cmark_strbuf_put(&parser->linebuf, repl, 3);
      } else {
        cmark_strbuf_put(&parser->linebuf, buffer, chunk_len);
      }
    }

    buffer += chunk_len;
    if (buffer < end) {
      if (*buffer == '\0') {
        // skip over NULL
        buffer++;
      } else {
        // skip over line ending characters
        if (*buffer == '\r') {
          buffer++;
          if (buffer == end)
            parser->last_buffer_ended_with_cr = true;
        }
        if (buffer < end && *buffer == '\n')