void loop(t_list **list, t_token **cur_tok, char *input)
{
int index;
char cur;
char prev;
ft_init(&index, &cur, &prev, input);
while (cur)
{
if ((*cur_tok)->token && is_operator(prev) && !is_newline(cur))
rule_1(*cur_tok, list, input, &index);
else if (is_quoting(cur))
handle_quoting(*cur_tok, list, input, &index);
else if (is_start_of_expansion(cur) || is_operator(cur))
rule_34(*cur_tok, list, input, &index);
else if (is_newline(cur) || is_whitespace(cur))
rule_56(*cur_tok, list, input, &index);
else if ((*cur_tok)->token && is_word(prev))
build_token(*cur_tok, cur);
else if (cur == '#')
discard_comment(input, &index);
else
build_token(*cur_tok, cur);
prev = cur;
cur = input[++index];
}
}
static bool config_parse_string(struct lexer *lex, struct strref *ref,
char end)
{
bool success = end != 0;
struct base_token token;
base_token_clear(&token);
while (lexer_getbasetoken(lex, &token, false)) {
if (end) {
if (*token.text.array == end) {
success = true;
break;
} else if (is_newline(*token.text.array)) {
success = false;
break;
}
} else {
if (is_newline(*token.text.array)) {
success = true;
break;
}
}
strref_add(ref, &token.text);
}
remove_ref_whitespace(ref);
return success;
}
static
bool token_line_comment(SLCONFIG_STRING* str, TOKEN* token)
{
if(str->start < str->end && *str->start == '/')
{
str->start++;
if(str->start < str->end && *str->start == '/')
{
str->start++;
token->str.start = str->start;
token->type = TOKEN_COMMENT;
while(str->start < str->end)
{
if(is_newline(*str->start))
break;
str->start++;
}
token->str.end = str->start;
return true;
}
else
{
str->start--;
}
}
return false;
}
bool _slc_is_naked_string_character(char c)
{
return !is_whitespace(c) && !is_newline(c) &&
c != ':' && c != '$' && c != '{' && c != '~' &&
c != '}' && c != ';' && c != '=' && c != '"' &&
c != '#';
}
static LTSV *parse_ltsv(const char **string) {
LTSV *ltsv = ltsv_init();
LTSV_Record *record;
if (!ltsv) {
fprintf(stderr, "couldn't allocate LTSV!\n");
return NULL;
}
while (**string != '\0') {
record = parse_record(string);
if (!record) {
if (!is_newline(**string)) {
goto bail;
}
fprintf(stderr, "string: %c\n", **string);
return ltsv;
}
ltsv_add(ltsv, record);
if (**string == 0x0d) {
skip_char(string);
}
if (**string == 0x0a) {
skip_char(string);
}
}
return ltsv;
bail:
ltsv_free(ltsv);
return NULL;
}
static bool cf_lexer_is_include(struct cf_lexer *lex)
{
bool found_include_import = false;
bool found_preprocessor = false;
size_t i;
for (i = lex->tokens.num; i > 0; i--) {
struct cf_token *token = lex->tokens.array+(i-1);
if (is_space_or_tab(*token->str.array))
continue;
if (!found_include_import) {
if (strref_cmp(&token->str, "include") != 0 &&
strref_cmp(&token->str, "import") != 0)
break;
found_include_import = true;
} else if (!found_preprocessor) {
if (*token->str.array != '#')
break;
found_preprocessor = true;
} else {
return is_newline(*token->str.array);
}
}
/* if starting line */
return found_preprocessor && found_include_import;
}
/* Raw parsing */
static uint16_t hb_parse_character( hb_csv_file_t * file )
{
int byte;
uint16_t c = 0;
int need_char = 1;
if( file == NULL )
{
return CSV_CHAR_ERROR;
}
while( need_char )
{
byte = fgetc( file->fileref );
if( feof( file->fileref ) )
{
return CSV_CHAR_EOF;
}
if( ferror( file->fileref ) )
{
return CSV_CHAR_ERROR;
}
if( file->parse_state == CSV_PARSE_SEEK && is_white(byte) )
{
continue;
}
else if( file->parse_state != CSV_PARSE_ESC && is_esc(byte) )
{
file->parse_state = CSV_PARSE_ESC;
continue;
}
else if( file->parse_state != CSV_PARSE_ESC && is_sep(byte) )
{
file->parse_state = CSV_PARSE_SEEK;
need_char = 0;
c = CSV_CHAR_COLSEP;
}
else if( file->parse_state == CSV_PARSE_ESC )
{
file->parse_state = CSV_PARSE_NORMAL;
need_char = 0;
c = (uint16_t)byte;
}
else if( is_newline(byte) )
{
file->parse_state = CSV_PARSE_SEEK;
need_char = 0;
c = CSV_CHAR_ROWSEP;
}
else
{
file->parse_state = CSV_PARSE_NORMAL;
need_char = 0;
c = (uint16_t)byte;
}
}
return c;
}
void CL_SpanLayout_Impl::layout_text(CL_GraphicContext & gc, std::vector<TextBlock> blocks, std::vector<TextBlock>::size_type block_index, CurrentLine ¤t_line, int max_width)
{
TextSizeResult text_size_result = find_text_size(gc, blocks[block_index], current_line.object_index);
current_line.object_index += text_size_result.objects_traversed;
current_line.cur_line.width = current_line.x_position;
if (is_newline(blocks[block_index]))
{
current_line.cur_line.height = cl_max(current_line.cur_line.height, text_size_result.height);
current_line.cur_line.ascender = cl_max(current_line.cur_line.ascender, text_size_result.ascender);
next_line(current_line);
}
else
{
if (!fits_on_line(current_line.x_position, text_size_result, max_width) && !is_whitespace(blocks[block_index]))
{
if (larger_than_line(text_size_result, max_width))
{
// force line breaks to make it fit
force_place_line_segments(current_line, text_size_result, max_width);
}
else
{
next_line(current_line);
place_line_segments(current_line, text_size_result);
}
}
else
{
place_line_segments(current_line, text_size_result);
}
}
}
// Scan newline characters.
//
// TODO: Scan for newline sequences ('\r\n').
std::size_t
newline(Lexer& t) {
if (t.head != t.tail and is_newline(*t.head))
return 1;
else
return 0;
}
static int config_parse(struct darray *sections, const char *file,
bool always_open)
{
char *file_data;
struct lexer lex;
struct base_token token;
struct strref section_name;
FILE *f;
f = os_fopen(file, "rb");
if (always_open && !f)
f = os_fopen(file, "w+");
if (!f)
return CONFIG_FILENOTFOUND;
os_fread_utf8(f, &file_data);
fclose(f);
if (!file_data)
return CONFIG_SUCCESS;
lexer_init(&lex);
lexer_start_move(&lex, file_data);
base_token_clear(&token);
while (lexer_getbasetoken(&lex, &token, PARSE_WHITESPACE)) {
struct config_section *section;
while (token.type == BASETOKEN_WHITESPACE) {
if (!lexer_getbasetoken(&lex, &token, PARSE_WHITESPACE))
goto complete;
}
if (*token.text.array != '[') {
while (!is_newline(*token.text.array)) {
if (!lexer_getbasetoken(&lex, &token,
PARSE_WHITESPACE))
goto complete;
}
continue;
}
strref_clear(§ion_name);
config_parse_string(&lex, §ion_name, ']');
if (!section_name.len)
break;
section = darray_push_back_new(sizeof(struct config_section),
sections);
section->name = bstrdup_n(section_name.array,
section_name.len);
config_parse_section(section, &lex);
}
complete:
lexer_free(&lex);
return CONFIG_SUCCESS;
}
pointer parse_expr(parser* parse)
{
eat_whitespace(parse);
pointer ret_car;
pointer ret_cdr;
switch(*parse->curr)
{
case '(':
parse->curr++;
ret_car = parse_expr(parse);
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
case '"':
ret_car = parse_string(parse);
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
case '\'':
parse->curr++;
ret_car = parse_quote(parse);
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
case ')':
parse->curr++;
return NIL;
case '+': case '-': case 'b':
ret_car = parse_number_or_symbol(parse);
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
case '.':
return parse_number_or_pair(parse);
case '\\':
parse->curr++;
ret_car = create_char(*(parse->curr++));
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
case ';':
while(!is_newline(*parse->curr) && *parse->curr != '\0')
parse->curr++;
return parse_expr(parse);
case 0:
return NIL;
default:
if(is_number_char(*parse->curr))
{
ret_car = parse_number(parse);
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
}
else if(is_symbol_char(*parse->curr))
{
ret_car = parse_symbol(parse);
ret_cdr = parse_expr(parse);
return create_pair(ret_car, ret_cdr);
}
else
return parser_error(parse, "Unexpected char in expression.");
}
parse->curr++;
}
void prewind(parser *p, char c) {
sseek(p->f, -1, SEEK_CUR);
if (is_newline(c)) {
p->current_location.line--;
} else {
p->current_location.column--;
}
}
void consume_comment(TokenizeContext& context)
{
int lookahead = 0;
while (context.withinRange(lookahead) && !is_newline(context.next(lookahead)))
lookahead++;
context.consume(tok_Comment, lookahead);
}
static void content(void)
{
char current;
/*
* We should get one character at a time.
*/
assert(strlen(yytext) == 1);
current = yytext[0];
if (current == EOF)
return;
if (is_newline(current)) {
newline();
return;
}
buffer_push_char(current_buf, current);
/*
* Forcing newline changes 'text' so lets do it after we've pushed
* it to the buffer.
*/
if (wrap_long_lines && need_wrap(current_buf)) {
struct buffer tmp;
buffer_init(&tmp, buffer_size(current_buf));
/*
* Find last character that was not whitespace
*/
for (;;) {
int c;
if (buffer_size(current_buf) == 0)
break;
c = buffer_pop_char(current_buf);
if (is_whitespace(c)) {
/*
* Do not push whitespace because it would appear
* after the newline.
*/
break;
}
/*
* Characters are put in tmp buffer in reverse order.
*/
buffer_push_char(&tmp, c);
}
force_newline_for_wrap(current_buf);
/*
* Restore non-wrapped text into buffer.
*/
while (buffer_size(&tmp) > 0) {
buffer_push_char(current_buf, buffer_pop_char(&tmp));
}
buffer_release(&tmp);
}
}
static LTSV_Record *parse_record(const char **string) {
LTSV_Record *output_record = ltsv_record_init();
const char *label = NULL, *value;
if (!output_record) {
fprintf(stderr, "couldn't allocate LTSV_Record!\n");
return NULL;
}
while (**string != '\0') {
label = parse_string(string, &label_char_test_func);
if (!label) {
fprintf(stderr, "could not parse label!\n");
goto bail;
}
if (strlen(label) == 0) {
if (is_newline(**string)) {
free((void*)label);
return output_record;
}
fprintf(stderr, "invalid character for label: %c\n", **string);
goto bail;
}
if (**string != ':') {
fprintf(stderr, "':' not found: '%c' found instead!\n", **string);
goto bail;
}
skip_char(string);
value = parse_string(string, &field_char_test_func);
if (ltsv_record_get_value(output_record, label)) {
fprintf(stderr, "dupliate entry for record: %s\n", label);
goto bail;
}
ltsv_record_add_entry(output_record, label, value);
if (**string == '\t') {
skip_char(string);
}
free((void*)label);
}
return output_record;
bail:
if (label) {
free((void*)label);
}
ltsv_record_free(output_record);
return NULL;
}
char pgetc(parser *p) {
char c = sgetc(p->f);
if (is_newline(c)) {
p->current_location.column = 1;
p->current_location.line++;
} else {
p->current_location.column++;
}
return c;
}
/* Force newline after tag. Use lexer to get current character. */
static void force_newline_after_tag(struct buffer * buffer)
{
int current = input();
do_unput(current);
if (!is_newline(current)) {
do_newline(buffer, "\n");
eat_whitespace();
}
}
// Consume horizontal and vertical whitespace.
//
// TODO: Manage the input location while lexing.
void
Lexer::space()
{
while (true) {
char c = peek();
if (is_space(c))
get();
else if (is_newline(c))
get();
else
break;
}
}
UniChar Reader::next()
{
mCurrentIndex++;
mOffset.column++;
UniChar current = this->current();
if(is_newline(current)) {
mOffset.line++;
mOffset.column = 0;
}
return current;
}
static void internal_putchar(int b, int c) {
serial_lock();
/* Add to buffer */
int index = output_buffers_used[b];
uint8_t *buffer = output_buffers[b];
buffer[index] = (uint8_t)c;
output_buffers_used[b]++;
if (index + 1 == GUEST_OUTPUT_BUFFER_SIZE || (index >= 1 && is_newline(buffer + index - 1)) || last_out == b) {
flush_buffer(b);
}
has_data = 1;
serial_unlock();
}
void add_newline(t_token *cur_tok, t_list **l, char *input, \
int *index)
{
char cur;
cur = input[*index];
if (is_newline(cur))
{
if (cur_tok->token)
append_and_reset(l, cur_tok);
discard_whitespace(input, index);
list_append(*l, init_token(NEWLINE, "\n"));
}
}
static void parse_buffer( vdb_info_bam_hdr * bam_hdr, char * buffer, size_t len )
{
char * line;
size_t idx, line_len, state = 0;
for ( idx = 0; idx < len; ++idx )
{
switch( state )
{
case 0 : if ( is_newline( buffer[ idx ] ) ) /* init */
state = 2;
else
{
line = &( buffer[ idx ] );
line_len = 1;
state = 1;
}
break;
case 1 : if ( is_newline( buffer[ idx ] ) ) /* content */
{
inspect_line( bam_hdr, line, line_len );
state = 2;
}
else
line_len++;
break;
case 2 : if ( !is_newline( buffer[ idx ] ) ) /* newline */
{
line = &( buffer[ idx ] );
line_len = 1;
state = 1;
}
break;
}
}
}
char ArffScanner::next() {
if(eof()) {
return (char)-1;
}
if(is_newline(m_char)) {
++m_line;
m_col = 0;
}
++m_col;
m_prev_char = m_char;
if(fread(&m_char, sizeof(m_char), 1, m_fp) != sizeof(m_char)) {
m_char = (char)-1; // you would have reached end-of-file?
}
return m_char;
}
/* Force newline for wrapping line. Use lexer to get current character and
do not eat whitespace from next line. */
static void force_newline_for_wrap(struct buffer * buffer)
{
int current = input();
/*
* Flush all pending stuff before doing the newline.
*/
if (!using_primary_buf()) {
merge_buffers();
}
do_newline(current_buf, "\n");
if (!is_newline(current))
do_unput(current);
}
void
Lexer::comment()
{
get();
while (true) {
char c = peek();
if (!c || is_newline(c))
break;
get();
}
// TODO: Do something interesting with comments
// instead of just discarding them.
build_.clear();
}
int readLine(FILE* file, char** outStr)
{
char buf[BUFSIZ];
int i = 0;
buf[i] = fgetc(file);
if (is_eof(buf[i])) {
return -1;
}
while (!is_eof(buf[i]) && !is_newline(buf[i])) {
i++;
buf[i] = fgetc(file);
}
buf[i] = '\0';
COPY_STR((*outStr), buf);
return i;
}
/* Force newline before tag. Use buffer for getting current character. */
static void force_newline_before_tag(struct buffer * buffer)
{
int current;
if (buffer_size(buffer) == 0) {
/*
* We just did a newline, no need to force it.
*/
return;
}
current = buffer_pop_char(buffer);
buffer_push_char(buffer, current);
if (!is_newline(current)) {
do_newline(buffer, "\n");
eat_whitespace();
}
}
static bool cf_lexer_process_comment(struct cf_lexer *lex,
struct cf_token *out_token)
{
const char *offset;
if (!cf_is_comment(out_token->unmerged_str.array))
return false;
offset = lex->base_lexer.offset;
cf_pass_any_splices(&offset);
strcpy(lex->write_offset++, " ");
out_token->str.len = 1;
if (*offset == '/') {
while (*++offset && !is_newline(*offset))
cf_pass_any_splices(&offset);
} else if (*offset == '*') {
bool was_star = false;
lex->unexpected_eof = true;
while (*++offset) {
cf_pass_any_splices(&offset);
if (was_star && *offset == '/') {
offset++;
lex->unexpected_eof = false;
break;
} else {
was_star = (*offset == '*');
}
}
}
out_token->unmerged_str.len +=
(size_t)(offset - out_token->unmerged_str.array);
out_token->type = CFTOKEN_SPACETAB;
lex->base_lexer.offset = offset;
return true;
}
static void config_parse_section(struct config_section *section,
struct lexer *lex)
{
struct base_token token;
while (lexer_getbasetoken(lex, &token, false)) {
struct strref name, value;
while (token.type == BASETOKEN_WHITESPACE) {
if (!lexer_getbasetoken(lex, &token, false))
return;
}
if (token.type == BASETOKEN_OTHER) {
if (*token.text.array == '#') {
do {
if (!lexer_getbasetoken(lex, &token,
false))
return;
} while (!is_newline(*token.text.array));
continue;
} else if (*token.text.array == '[') {
lex->offset--;
return;
}
}
strref_copy(&name, &token.text);
if (!config_parse_string(lex, &name, '='))
continue;
strref_clear(&value);
config_parse_string(lex, &value, 0);
config_add_item(§ion->items, &name, &value);
}
}
static inline enum cf_token_type cf_get_token_type(const struct cf_token *token,
const struct base_token *start_token)
{
switch (start_token->type) {
case BASETOKEN_ALPHA:
return CFTOKEN_NAME;
case BASETOKEN_DIGIT:
return CFTOKEN_NUM;
case BASETOKEN_WHITESPACE:
if (is_newline(*token->str.array))
return CFTOKEN_NEWLINE;
else
return CFTOKEN_SPACETAB;
case BASETOKEN_NONE:
case BASETOKEN_OTHER:
break;
}
return CFTOKEN_OTHER;
}
