* Functions necessary to parse a file and transform its content into
* a deck of slides containing lines. All based on markdown formating
* rules.
- * Copyright (C) 2014 Michael Goehler
+ * Copyright (C) 2018 Michael Goehler
*
* This file is part of mdp.
*
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
+#include <wchar.h>
+#include <wctype.h>
#include <string.h>
#include "parser.h"
-deck_t *markdown_load(FILE *input) {
-
- int c = 0; // char
+// char entry translation table
+static struct named_character_entity {
+ wchar_t ucs;
+ const wchar_t *name;
+} named_character_entities[] = {
+ { L'\x0022', L"quot" },
+ { L'\x0026', L"amp" },
+ { L'\x0027', L"apos" },
+ { L'\x003C', L"lt" },
+ { L'\x003E', L"gt" },
+ { L'\x00A2', L"cent" },
+ { L'\x00A3', L"pound" },
+ { L'\x00A5', L"yen" },
+ { L'\x00A7', L"sect" },
+ { L'\x00A9', L"copy" },
+ { L'\x00AA', L"laquo" },
+ { L'\x00AE', L"reg" },
+ { L'\x00B0', L"deg" },
+ { L'\x00B1', L"plusmn" },
+ { L'\x00B2', L"sup2" },
+ { L'\x00B3', L"sup3" },
+ { L'\x00B6', L"para" },
+ { L'\x00B9', L"sup1" },
+ { L'\x00BB', L"raquo" },
+ { L'\x00BC', L"frac14" },
+ { L'\x00BD', L"frac12" },
+ { L'\x00BE', L"frac34" },
+ { L'\x00D7', L"times" },
+ { L'\x00F7', L"divide" },
+ { L'\x2018', L"lsquo" },
+ { L'\x2019', L"rsquo" },
+ { L'\x201C', L"ldquo" },
+ { L'\x201D', L"rdquo" },
+ { L'\x2020', L"dagger" },
+ { L'\x2021', L"Dagger" },
+ { L'\x2022', L"bull" },
+ { L'\x2026', L"hellip" },
+ { L'\x2030', L"permil" },
+ { L'\x2032', L"prime" },
+ { L'\x2033', L"Prime" },
+ { L'\x2039', L"lsaquo" },
+ { L'\x203A', L"rsaquo" },
+ { L'\x20AC', L"euro" },
+ { L'\x2122', L"trade" },
+ { L'\x2190', L"larr" },
+ { L'\x2191', L"uarr" },
+ { L'\x2192', L"rarr" },
+ { L'\x2193', L"darr" },
+ { L'\x2194', L"harr" },
+ { L'\x21B5', L"crarr" },
+ { L'\x21D0', L"lArr" },
+ { L'\x21D1', L"uArr" },
+ { L'\x21D2', L"rArr" },
+ { L'\x21D3', L"dArr" },
+ { L'\x21D4', L"hArr" },
+ { L'\x221E', L"infin" },
+ { L'\x2261', L"equiv" },
+ { L'\x2308', L"lceil" },
+ { L'\x2309', L"rceil" },
+ { L'\x230A', L"lfloor" },
+ { L'\x230B', L"rfloor" },
+ { L'\x25CA', L"loz" },
+ { L'\x2660', L"spades" },
+ { L'\x2663', L"clubs" },
+ { L'\x2665', L"hearts" },
+ { L'\x2666', L"diams" },
+ { L'\0', NULL },
+};
+
+deck_t *markdown_load(FILE *input, int noexpand) {
+
+ wchar_t c = L'\0'; // char
int i = 0; // increment
int hc = 0; // header count
int lc = 0; // line count
// initialize bits as empty line
SET_BIT(bits, IS_EMPTY);
- while ((c = fgetc(input)) != EOF) {
+ while ((c = fgetwc(input)) != WEOF) {
if (ferror(input)) {
fprintf(stderr, "markdown_load() failed to read input: %s\n", strerror(errno));
exit(EXIT_FAILURE);
}
- if(c == '\n') {
+ if(c == L'\n') {
// markdown analyse
prev = bits;
// clear text
(text->reset)(text);
+ } else if(line && CHECK_BIT(bits, IS_STOP)) {
+
+ // set stop bit on last line
+ SET_BIT(line->bits, IS_STOP);
+
+ // clear text
+ (text->reset)(text);
+
// if text is markdown hr
} else if(CHECK_BIT(bits, IS_HR) &&
CHECK_BIT(line->bits, IS_EMPTY)) {
slide = next_slide(slide);
sc++;
+ } else if((CHECK_BIT(bits, IS_TILDE_CODE) ||
+ CHECK_BIT(bits, IS_GFM_CODE)) &&
+ CHECK_BIT(bits, IS_EMPTY)) {
+ // remove tilde code markers
+ (text->reset)(text);
+
} else {
// if slide ! has line
- if(!slide->line) {
+ if(!slide->line || !line) {
// create new line
line = new_line();
// calc offset
line->offset = next_nonblank(text, 0);
+ // expand character entities if enabled
+ if(line->text->value &&
+ !noexpand &&
+ !CHECK_BIT(line->bits, IS_CODE))
+ expand_character_entities(line);
+
// adjust line length dynamicaly - excluding markup
- if(line->text->text)
+ if(line->text->value)
adjust_line_length(line);
// new text
text = cstring_init();
}
- } else if(c == '\t') {
+ } else if(c == L'\t') {
// expand tab to spaces
for (i = 0; i < EXPAND_TABS; i++) {
- (text->expand)(text, ' ');
+ (text->expand)(text, L' ');
}
- } else if(c == '\\') {
+ } else if(c == L'\\') {
// add char to line
(text->expand)(text, c);
// and do not increase line count
if(next_nonblank(text, 0) < CODE_INDENT) {
- c = fgetc(input);
+ c = fgetwc(input);
(text->expand)(text, c);
-
- if(is_utf8(c)) {
-
- // if utf-8 char > 1 byte add remaing to line
- for(i = 0; i < length_utf8(c) - 1; i++) {
- c = fgetc(input);
- (text->expand)(text, c);
- }
- }
-
}
- } else if(isprint(c) || isspace((unsigned char) c)) {
-
- // add char to line
- (text->expand)(text, c);
-
- } else if(is_utf8(c)) {
+ } else if(iswprint(c) || iswspace(c)) {
// add char to line
(text->expand)(text, c);
-
- // if utf-8 char > 1 byte add remaing to line
- for(i = 0; i < length_utf8(c) - 1; i++) {
- c = fgetc(input);
- (text->expand)(text, c);
- }
}
}
(text->delete)(text);
// detect header
line = deck->slide->line;
- if(line && line->text->size > 0 && line->text->text[0] == '%') {
+ if(line && line->text->size > 0 && line->text->value[0] == L'%') {
// assign header to deck
deck->header = line;
// find first non-header line
- while(line && line->text->size > 0 && line->text->text[0] == '%') {
+ while(line && line->text->size > 0 && line->text->value[0] == L'%') {
hc++;
line = line->next;
}
// adjust counts
deck->headers += hc;
deck->slide->lines -= hc;
+ } else {
+
+ // remove header from deck
+ deck->header = NULL;
}
}
slide = deck->slide;
while(slide) {
line = slide->line;
+
+ // ignore mdpress format attributes
+ if(line &&
+ slide->lines > 1 &&
+ !CHECK_BIT(line->bits, IS_EMPTY) &&
+ line->text->value[line->offset] == L'=' &&
+ line->text->value[line->offset + 1] == L' ') {
+
+ // remove line from linked list
+ slide->line = line->next;
+ line->next->prev = NULL;
+
+ // maintain loop condition
+ tmp = line;
+ line = line->next;
+
+ // adjust line count
+ slide->lines -= 1;
+
+ // delete line
+ (tmp->text->delete)(tmp->text);
+ free(tmp);
+ }
+
while(line) {
// combine underlined H1/H2 in single line
if((CHECK_BIT(line->bits, IS_H1) ||
// the program remembers their value on every function calls
static int unordered_list_level = 0;
static int unordered_list_level_offset[] = {-1, -1, -1, -1};
+ static int num_tilde_characters = 0;
+ static int num_backticks = 0;
int i = 0; // increment
int bits = 0; // markdown bits
const int unordered_list_offset = unordered_list_level_offset[unordered_list_level];
// return IS_EMPTY on null pointers
- if(!text || !text->text) {
+ if(!text || !text->value) {
SET_BIT(bits, IS_EMPTY);
+
+ // continue fenced code blocks across empty lines
+ if(num_tilde_characters > 0)
+ SET_BIT(bits, IS_CODE);
+
return bits;
}
// count leading spaces
offset = next_nonblank(text, 0);
+ // IS_TILDE_CODE
+ if (wcsncmp(text->value, L"~~~", 3) == 0) {
+ int tildes_in_line = next_nontilde(text, 0);
+ if (tildes_in_line >= num_tilde_characters) {
+ if (num_tilde_characters > 0) {
+ num_tilde_characters = 0;
+ } else {
+ num_tilde_characters = tildes_in_line;
+ }
+ SET_BIT(bits, IS_EMPTY);
+ SET_BIT(bits, IS_TILDE_CODE);
+ return bits;
+ }
+ }
+
+ if (num_tilde_characters > 0) {
+ SET_BIT(bits, IS_CODE);
+ SET_BIT(bits, IS_TILDE_CODE);
+ return bits;
+ }
+
+ // IS_GFM_CODE
+ if (wcsncmp(text->value, L"```", 3) == 0) {
+ int backticks_in_line = next_nonbacktick(text, 0);
+ if (backticks_in_line >= num_backticks) {
+ if (num_backticks > 0) {
+ num_backticks = 0;
+ } else {
+ num_backticks = backticks_in_line;
+ }
+ SET_BIT(bits, IS_EMPTY);
+ SET_BIT(bits, IS_GFM_CODE);
+ return bits;
+ }
+ }
+
+ if (num_backticks > 0) {
+ SET_BIT(bits, IS_CODE);
+ SET_BIT(bits, IS_GFM_CODE);
+ return bits;
+ }
+
+ // IS_STOP
+ if((offset < CODE_INDENT || !CHECK_BIT(prev, IS_CODE)) &&
+ (!wcsncmp(&text->value[offset], L"<br>", 4) ||
+ !wcsncmp(&text->value[offset], L"<BR>", 4) ||
+ !wcsncmp(&text->value[offset], L"^", 1))) {
+ SET_BIT(bits, IS_STOP);
+ return bits;
+ }
+
// strip trailing spaces
- for(eol = text->size; eol > offset && isspace((unsigned char) text->text[eol - 1]); eol--);
+ for(eol = text->size; eol > offset && iswspace(text->value[eol - 1]); eol--);
+ text->size = eol;
// IS_UNORDERED_LIST_#
if(text->size >= offset + 2 &&
- (text->text[offset] == '*' || text->text[offset] == '-') &&
- text->text[offset + 1] == ' ') {
+ (text->value[offset] == L'*' || text->value[offset] == L'-') &&
+ iswspace(text->value[offset + 1])) {
// if different from last lines offset
if(offset != unordered_list_offset) {
// IS_CODE
if(offset >= CODE_INDENT &&
(CHECK_BIT(prev, IS_EMPTY) ||
- CHECK_BIT(prev, IS_CODE))) {
+ CHECK_BIT(prev, IS_CODE) ||
+ CHECK_BIT(prev, IS_STOP))) {
SET_BIT(bits, IS_CODE);
} else {
// IS_QUOTE
- if(text->text[offset] == '>') {
+ if(text->value[offset] == L'>') {
SET_BIT(bits, IS_QUOTE);
}
// IS_CENTER
if(text->size >= offset + 3 &&
- text->text[offset] == '-' &&
- text->text[offset + 1] == '>' &&
- text->text[offset + 2] == ' ') {
+ text->value[offset] == L'-' &&
+ text->value[offset + 1] == L'>' &&
+ iswspace(text->value[offset + 2])) {
SET_BIT(bits, IS_CENTER);
// remove start tag
eol -= 3;
if(text->size >= offset + 3 &&
- text->text[eol - 1] == '-' &&
- text->text[eol - 2] == '<' &&
- text->text[eol - 3] == ' ') {
+ text->value[eol - 1] == L'-' &&
+ text->value[eol - 2] == L'<' &&
+ iswspace(text->value[eol - 3])) {
// remove end tags
(text->strip)(text, eol - 3, 3);
// adjust end of line
- for(eol = text->size; eol > offset && isspace((unsigned char) text->text[eol - 1]); eol--);
+ for(eol = text->size; eol > offset && iswspace(text->value[eol - 1]); eol--);
}
}
for(i = offset; i < eol; i++) {
- if(text->text[i] == ' ') {
+ if(iswspace(text->value[i])) {
spaces++;
} else {
- switch(text->text[i]) {
- case '=': equals++; break;
- case '#': hashes++; break;
- case '*': stars++; break;
- case '-': minus++; break;
- case '\\': other++; i++; break;
+ switch(text->value[i]) {
+ case L'=': equals++; break;
+ case L'#': hashes++; break;
+ case L'*': stars++; break;
+ case L'-': minus++; break;
+ case L'\\': other++; i++; break;
default: other++; break;
}
}
// IS_H1
if(equals > 0 &&
- hashes + stars + minus + spaces + other == 0) {
+ hashes + stars + minus + spaces + other == 0) {
SET_BIT(bits, IS_H1);
}
- if(text->text[offset] == '#' &&
- text->text[offset+1] == ' ') {
+ if(text->value[offset] == L'#' &&
+ iswspace(text->value[offset+1])) {
SET_BIT(bits, IS_H1);
SET_BIT(bits, IS_H1_ATX);
}
// IS_H2
if(minus > 0 &&
- equals + hashes + stars + spaces + other == 0) {
+ equals + hashes + stars + spaces + other == 0) {
SET_BIT(bits, IS_H2);
}
- if(text->text[offset] == '#' &&
- text->text[offset+1] == '#' &&
- text->text[offset+2] == ' ') {
+ if(text->value[offset] == L'#' &&
+ text->value[offset+1] == L'#' &&
+ iswspace(text->value[offset+2])) {
SET_BIT(bits, IS_H2);
SET_BIT(bits, IS_H2_ATX);
}
line_t *header;
if(debug == 1) {
- fprintf(stderr, "headers: %i\nslides: %i\n", deck->headers, deck->slides);
+ fwprintf(stderr, L"headers: %i\nslides: %i\n", deck->headers, deck->slides);
} else if(debug > 1) {
header = deck->header;
while(header &&
header->length > 0 &&
- header->text->text[0] == '%') {
+ header->text->value[0] == L'%') {
// skip descriptor word (e.g. %title:)
offset = next_blank(header->text, 0) + 1;
- fprintf(stderr, "header: %s\n", &header->text->text[offset]);
+ fwprintf(stderr, L"header: %S\n", &header->text->value[offset]);
header = header->next;
}
}
sc++;
if(debug == 1) {
- fprintf(stderr, " slide %i: %i lines\n", sc, slide->lines);
+ fwprintf(stderr, L" slide %i: %i lines\n", sc, slide->lines);
} else if(debug > 1) {
// also print bits and line length
- fprintf(stderr, " slide %i:\n", sc);
+ fwprintf(stderr, L" slide %i:\n", sc);
line = slide->line;
lc = 0;
while(line) {
lc++;
- fprintf(stderr, " line %i: bits = %i, length = %i\n", lc, line->bits, line->length);
+ fwprintf(stderr, L" line %i: bits = %i, length = %i\n", lc, line->bits, line->length);
line = line->next;
}
}
}
}
+void expand_character_entities(line_t *line)
+{
+ wchar_t *ampersand;
+ wchar_t *prev, *curr;
+
+ ampersand = NULL;
+ curr = &line->text->value[0];
+
+ // for each char in line
+ for(prev = NULL; *curr; prev = curr++) {
+ if (*curr == L'&' && (prev == NULL || *prev != L'\\')) {
+ ampersand = curr;
+ continue;
+ }
+ if (ampersand == NULL) {
+ continue;
+ }
+ if (*curr == L'#') {
+ if (prev == ampersand)
+ continue;
+ ampersand = NULL;
+ continue;
+ }
+ if (iswalpha(*curr) || iswxdigit(*curr)) {
+ continue;
+ }
+ if (*curr == L';') {
+ int cnt;
+ wchar_t ucs = L'\0';
+ if (ampersand + 1 >= curr || ampersand + 16 < curr) { // what is a good limit?
+ ampersand = NULL;
+ continue;
+ }
+ if (ampersand[1] == L'#') { // &#nnnn; or &#xhhhh;
+ if (ampersand + 2 >= curr) {
+ ampersand = NULL;
+ continue;
+ }
+ if (ampersand[2] != L'x') { // &#nnnn;
+ cnt = wcsspn(&ersand[2], L"0123456789");
+ if (ampersand + 2 + cnt != curr) {
+ ampersand = NULL;
+ continue;
+ }
+ ucs = wcstoul(&ersand[2], NULL, 10);
+ } else { // &#xhhhh;
+ if (ampersand + 3 >= curr) {
+ ampersand = NULL;
+ continue;
+ }
+ cnt = wcsspn(&ersand[3], L"0123456789abcdefABCDEF");
+ if (ampersand + 3 + cnt != curr) {
+ ampersand = NULL;
+ continue;
+ }
+ ucs = wcstoul(&ersand[3], NULL, 16);
+ }
+ } else { // &name;
+ for (cnt = 0; cnt < sizeof(named_character_entities)/sizeof(named_character_entities[0]); ++cnt) {
+ if (wcsncmp(named_character_entities[cnt].name, &ersand[1], curr - ampersand - 1))
+ continue;
+ ucs = named_character_entities[cnt].ucs;
+ break;
+ }
+ if (ucs == L'\0') {
+ ampersand = NULL;
+ continue;
+ }
+ }
+ *ampersand = ucs;
+ cstring_strip(line->text, ampersand + 1 - &line->text->value[0], curr - ampersand);
+ curr = ampersand;
+ ampersand = NULL;
+ }
+ }
+}
+
void adjust_line_length(line_t *line) {
int l = 0;
- const static char *special = "\\*_`"; // list of interpreted chars
- const char *c = &line->text->text[line->offset];
+ const static wchar_t *special = L"\\*_`"; // list of interpreted chars
+ const wchar_t *c = &line->text->value[0];
cstack_t *stack = cstack_init();
// for each char in line
for(; *c; c++) {
// if char is in special char list
- if(strchr(special, *c)) {
+ if(wcschr(special, *c)) {
// closing special char (or second backslash)
if((stack->top)(stack, *c)) {
- if(*c == '\\') l++;
+ if(*c == L'\\') l++;
(stack->pop)(stack);
// treat special as regular char
- } else if((stack->top)(stack, '\\')) {
+ } else if((stack->top)(stack, L'\\')) {
l++;
(stack->pop)(stack);
} else {
// remove backslash from stack
- if((stack->top)(stack, '\\'))
+ if((stack->top)(stack, L'\\'))
(stack->pop)(stack);
l++;
}
(stack->delete)(stack);
}
-bool is_utf8(char ch) {
- return (ch & 0x80) != 0x00;
-}
-
-int length_utf8(char ch) {
-
- int i = 0; // increment
-
- while(is_utf8(ch)) {
- i++;
- ch <<= 1;
- }
-
- return i;
-}
-
int next_nonblank(cstring_t *text, int i) {
- while ((i < text->size) && isspace((unsigned char) (text->text)[i]))
+ while ((i < text->size) && iswspace((text->value)[i]))
i++;
return i;
}
int prev_blank(cstring_t *text, int i) {
- while ((i > 0) && !isspace((unsigned char) (text->text)[i]))
+ while ((i > 0) && !iswspace((text->value)[i]))
i--;
return i;
}
int next_blank(cstring_t *text, int i) {
- while ((i < text->size) && !isspace((unsigned char) (text->text)[i]))
+ while ((i < text->size) && !iswspace((text->value)[i]))
i++;
return i;
int next_word(cstring_t *text, int i) {
return next_nonblank(text, next_blank(text, i));
}
+
+int next_nontilde(cstring_t *text, int i) {
+ while ((i < text->size) && text->value[i] == L'~')
+ i++;
+
+ return i;
+}
+
+int next_nonbacktick(cstring_t *text, int i) {
+ while ((i < text->size) && text->value[i] == L'`')
+ i++;
+
+ return i;
+}
+
