X-Git-Url: https://git.danieliu.xyz/?a=blobdiff_plain;f=src%2Fparser.c;h=5ca7ea26a211c2fc223a23a7706bb59e6a31303d;hb=680d0478fc66d6d25b1f258be0114d6c915ec0a0;hp=c4fb88089d23999e34d3b00ae7a1b4a76508b0b6;hpb=5ccf244654d7b94b8c88cd6158549f42453fa7b3;p=smdp.git

diff --git a/src/parser.c b/src/parser.c
index c4fb880..5ca7ea2 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -25,18 +25,21 @@
 #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
+    wchar_t c = L'\0';    // char
     int i = 0;    // increment
     int hc = 0;   // header count
     int lc = 0;   // line count
     int sc = 1;   // slide count
     int bits = 0; // markdown bits
+    int prev = 0; // markdown bits of previous line
 
     deck_t *deck = new_deck();
     slide_t *slide = deck->slide;
@@ -44,16 +47,20 @@ deck_t *markdown_load(FILE *input) {
     line_t *tmp = NULL;
     cstring_t *text = cstring_init();
 
-    while ((c = fgetc(input)) != EOF) {
+    // initialize bits as empty line
+    SET_BIT(bits, IS_EMPTY);
+
+    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
-            bits = markdown_analyse(text);
+            prev = bits;
+            bits = markdown_analyse(text, prev);
 
             // if first line in file is markdown hr
             if(!line && CHECK_BIT(bits, IS_HR)) {
@@ -61,6 +68,14 @@ deck_t *markdown_load(FILE *input) {
                 // 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)) {
@@ -77,7 +92,7 @@ deck_t *markdown_load(FILE *input) {
             } else {
 
                 // if slide ! has line
-                if(!slide->line) {
+                if(!slide->line || !line) {
 
                     // create new line
                     line = new_line();
@@ -102,21 +117,21 @@ deck_t *markdown_load(FILE *input) {
                 line->offset = next_nonblank(text, 0);
 
                 // 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);
@@ -125,76 +140,65 @@ deck_t *markdown_load(FILE *input) {
             // 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);
 
     slide->lines = lc;
     deck->slides = sc;
 
     // 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->text->size > 0 && line->text->text[0] == '%') {
+        while(line && line->text->size > 0 && line->text->value[0] == L'%') {
             hc++;
             line = line->next;
         }
 
-        // split linked list
-        line->prev->next = NULL;
-        line->prev = NULL;
+        // only split header if any non-header line is found
+        if(line) {
 
-        // remove header lines from slide
-        deck->slide->line = line;
+            // split linked list
+            line->prev->next = NULL;
+            line->prev = NULL;
+
+            // remove header lines from slide
+            deck->slide->line = line;
+
+            // adjust counts
+            deck->headers += hc;
+            deck->slide->lines -= hc;
+        } else {
 
-        // adjust counts
-        deck->headers += hc;
-        deck->slide->lines -= hc;
+            // remove header from deck
+            deck->header = NULL;
+        }
     }
 
     slide = deck->slide;
     while(slide) {
         line = slide->line;
         while(line) {
+            // combine underlined H1/H2 in single line
             if((CHECK_BIT(line->bits, IS_H1) ||
                 CHECK_BIT(line->bits, IS_H2)) &&
                CHECK_BIT(line->bits, IS_EMPTY) &&
                line->prev &&
                !CHECK_BIT(line->prev->bits, IS_EMPTY)) {
-                // combine underlined H1/H2 in single line
+
 
                 // remove line from linked list
                 line->prev->next = line->next;
@@ -219,6 +223,8 @@ deck_t *markdown_load(FILE *input) {
                 (tmp->text->delete)(tmp->text);
                 free(tmp);
 
+            // pass enclosing flag IS_UNORDERED_LIST_3
+            // to nested levels for unordered lists
             } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_3)) {
                 tmp = line->next;
                 line_t *list_last_level_3 = line;
@@ -235,6 +241,8 @@ deck_t *markdown_load(FILE *input) {
                     SET_BIT(tmp->bits, IS_UNORDERED_LIST_3);
                 }
 
+            // pass enclosing flag IS_UNORDERED_LIST_2
+            // to nested levels for unordered lists
             } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_2)) {
                 tmp = line->next;
                 line_t *list_last_level_2 = line;
@@ -252,6 +260,8 @@ deck_t *markdown_load(FILE *input) {
                     SET_BIT(tmp->bits, IS_UNORDERED_LIST_2);
                 }
 
+            // pass enclosing flag IS_UNORDERED_LIST_1
+            // to nested levels for unordered lists
             } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_1)) {
                 tmp = line->next;
                 line_t *list_last_level_1 = line;
@@ -279,8 +289,10 @@ deck_t *markdown_load(FILE *input) {
     return deck;
 }
 
-int markdown_analyse(cstring_t *text) {
+int markdown_analyse(cstring_t *text, int prev) {
 
+    // static variables can not be redeclaired, but changed outside of a declaration
+    // 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};
 
@@ -296,7 +308,7 @@ int markdown_analyse(cstring_t *text) {
     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);
         return bits;
     }
@@ -304,47 +316,52 @@ int markdown_analyse(cstring_t *text) {
     // count leading spaces
     offset = next_nonblank(text, 0);
 
+    // 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--);
 
     // IS_UNORDERED_LIST_#
     if(text->size >= offset + 2 &&
-       (text->text[offset] == '*' || text->text[offset] == '-') &&
-       text->text[offset + 1] == ' ') {
-
-        for(i = offset; i<eol; i++) {
-            if(text->text[i] != '*' &&
-               text->text[i] != '-' &&
-               text->text[i] != ' ') {
-                if(offset > unordered_list_offset + CODE_INDENT) {
-                    SET_BIT(bits, IS_CODE);
-                } else if(offset != unordered_list_offset) {
-                    for(i = unordered_list_level; i >= 0; i--) {
-                        if(unordered_list_level_offset[i] == offset) {
-                            unordered_list_level = i;
-                            break;
-                        }
-                    }
-                    if(i != unordered_list_level) {
-                        unordered_list_level = MIN(unordered_list_level + 1, UNORDERED_LIST_MAX_LEVEL);
-                        unordered_list_level_offset[unordered_list_level] = offset;
-                    }
-                }
+       (text->value[offset] == L'*' || text->value[offset] == L'-') &&
+       iswspace(text->value[offset + 1])) {
 
-                if(unordered_list_level == 0) {
-                    unordered_list_level = 1;
-                    unordered_list_level_offset[1] = offset;
-                }
+        // if different from last lines offset
+        if(offset != unordered_list_offset) {
 
-                switch(unordered_list_level) {
-                    case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
-                    case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
-                    case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
-                    default: break;
+            // test if offset matches a lower indent level
+            for(i = unordered_list_level; i >= 0; i--) {
+                if(unordered_list_level_offset[i] == offset) {
+                    unordered_list_level = i;
+                    break;
                 }
-
-                break;
             }
+            // if offset doesn't match any previously stored indent level
+            if(i != unordered_list_level) {
+                unordered_list_level = MIN(unordered_list_level + 1, UNORDERED_LIST_MAX_LEVEL);
+                // memorize the offset as next bigger indent level
+                unordered_list_level_offset[unordered_list_level] = offset;
+            }
+        }
+
+        // if no previous indent level matches, this must be the first line of the list
+        if(unordered_list_level == 0) {
+            unordered_list_level = 1;
+            unordered_list_level_offset[1] = offset;
+        }
+
+        switch(unordered_list_level) {
+            case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
+            case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
+            case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
+            default: break;
         }
     }
 
@@ -352,24 +369,51 @@ int markdown_analyse(cstring_t *text) {
        !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
        !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {
 
-        unordered_list_level = 0;
+        // continue list if indent level is still the same as in previous line
+        if ((CHECK_BIT(prev, IS_UNORDERED_LIST_1) ||
+             CHECK_BIT(prev, IS_UNORDERED_LIST_2) ||
+             CHECK_BIT(prev, IS_UNORDERED_LIST_3)) &&
+            offset >= unordered_list_offset) {
+
+            switch(unordered_list_level) {
+                case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
+                case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
+                case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
+                default: break;
+            }
+
+            // this line extends the previous list item
+            SET_BIT(bits, IS_UNORDERED_LIST_EXT);
+
+        // or reset indent level
+        } else {
+            unordered_list_level = 0;
+        }
+    }
+
+    if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
+       !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
+       !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {
 
         // IS_CODE
-        if(offset >= CODE_INDENT) {
+        if(offset >= CODE_INDENT &&
+           (CHECK_BIT(prev, IS_EMPTY) ||
+            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
@@ -377,31 +421,31 @@ int markdown_analyse(cstring_t *text) {
                 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;
                     }
                 }
@@ -409,23 +453,23 @@ int markdown_analyse(cstring_t *text) {
 
             // 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);
             }
@@ -456,7 +500,7 @@ void markdown_debug(deck_t *deck, int debug) {
     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) {
 
@@ -465,12 +509,12 @@ void markdown_debug(deck_t *deck, int debug) {
             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;
             }
         }
@@ -484,17 +528,17 @@ void markdown_debug(deck_t *deck, int debug) {
         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;
             }
         }
@@ -505,22 +549,22 @@ void markdown_debug(deck_t *deck, int debug) {
 
 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[line->offset];
     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);
 
@@ -531,7 +575,7 @@ void adjust_line_length(line_t *line) {
 
         } else {
             // remove backslash from stack
-            if((stack->top)(stack, '\\'))
+            if((stack->top)(stack, L'\\'))
                 (stack->pop)(stack);
             l++;
         }
@@ -547,38 +591,22 @@ void adjust_line_length(line_t *line) {
     (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;