Added support of character entities.

diff --git a/include/parser.h b/include/parser.h
index e8ad380..83ce5d4 100644 (file)
--- a/include/parser.h
+++ b/include/parser.h
@@ -52,6 +52,8 @@
 deck_t *markdown_load(FILE *input);
 int markdown_analyse(cstring_t *text, int prev);
 void markdown_debug(deck_t *deck, int debug);
+void expand_character_entities(line_t *line);
+void setup_character_entities(void);
 void adjust_line_length(line_t *line);
 int next_nonblank(cstring_t *text, int i);
 int prev_blank(cstring_t *text, int i);

diff --git a/src/main.c b/src/main.c
index 39b6cd5..19e7687 100644 (file)
--- a/src/main.c
+++ b/src/main.c
@@ -98,6 +98,9 @@ int main(int argc, char *argv[]) {
     // setup list string
     setup_list_strings();
 
+    // setup character entities
+    setup_character_entities();
+
     // open file or set input to STDIN
     char *file = NULL;
     FILE *input;

diff --git a/src/parser.c b/src/parser.c
index eb50016..f084667 100644 (file)
--- a/src/parser.c
+++ b/src/parser.c
@@ -121,6 +121,10 @@ deck_t *markdown_load(FILE *input) {
                 // calc offset
                 line->offset = next_nonblank(text, 0);
 
+                // expand character entities if enabled
+                if(line->text->value)
+                    expand_character_entities(line);
+
                 // adjust line length dynamicaly - excluding markup
                 if(line->text->value)
                     adjust_line_length(line);
@@ -603,6 +607,158 @@ void markdown_debug(deck_t *deck, int debug) {
     }
 }
 
+static int enable_character_entities = 0;
+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 },
+};
+
+/*
+export MDP_ENABLE_CHARENT=1
+*/
+void setup_character_entities(void)
+{
+    char *str = getenv("MDP_ENABLE_CHARENT");
+    if (str == NULL)
+        enable_character_entities = 0;
+    else if (str[0] == '\0')
+        enable_character_entities = 1;
+    else
+        enable_character_entities = atoi(str);
+}
+
+void expand_character_entities(line_t *line)
+{
+    wchar_t *ampersand;
+    wchar_t *prev, *curr;
+
+    if (!enable_character_entities)
+        return;
+
+    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;
+            goto clean;
+        }
+        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?
+                goto clean;
+            if (ampersand[1] == L'#') { // &#nnnn; or &#xhhhh;
+                if (ampersand + 2 >= curr)
+                    goto clean;
+                if (ampersand[2] != L'x') { // &#nnnn;
+                    cnt = wcsspn(&ampersand[2], L"0123456789");
+                    if (ampersand + 2 + cnt != curr)
+                        goto clean;
+                    ucs = wcstoul(&ampersand[2], NULL, 10);
+                } else { // &#xhhhh;
+                    if (ampersand + 3 >= curr)
+                        goto clean;
+                    cnt = wcsspn(&ampersand[3], L"0123456789abcdefABCDEF");
+                    if (ampersand + 3 + cnt != curr)
+                        goto clean;
+                    ucs = wcstoul(&ampersand[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, &ampersand[1], curr - ampersand - 1))
+                        continue;
+                    ucs = named_character_entities[cnt].ucs;
+                    break;
+                }
+                if (ucs == L'\0')
+                    goto clean;
+            }
+            *ampersand = ucs;
+            cstring_strip(line->text, ampersand + 1 - &line->text->value[0], curr - ampersand);
+            curr = ampersand;
+            continue;
+        }
+clean:
+        ampersand = NULL;
+    }
+}
+
 void adjust_line_length(line_t *line) {
     int l = 0;
     const static wchar_t *special = L"\\*_`"; // list of interpreted chars