Use stylesheet transformations to output HTML, for example
This commit is contained in:
David Baer
208
src/format.c
Normal file
208
src/format.c
Normal file
@@ -0,0 +1,208 @@
|
||||
#include <string.h>
|
||||
#include "queue.h"
|
||||
#include "stack.h"
|
||||
#include "utf8.h"
|
||||
#include "format.h"
|
||||
|
||||
DEFINE_QUEUE(FormatElement, FormatElementQueue);
|
||||
|
||||
#define EM_DASH_UTF8 "\xe2\x80\x94"
|
||||
|
||||
typedef utf8iterator* Tokenizer;
|
||||
typedef enum {
|
||||
TOK_TEXT,
|
||||
TOK_GREEK,
|
||||
TOK_UNICODE,
|
||||
TOK_STAR,
|
||||
TOK_REF,
|
||||
/*
|
||||
TOK_DASH,
|
||||
TOK_OPEN_DOUBLE_QUOTE,
|
||||
TOK_CLOSE_DOUBLE_QUOTE,
|
||||
TOK_OPEN_SINGLE_QUOTE,
|
||||
TOK_CLOSE_SINGLE_QUOTE,
|
||||
*/
|
||||
TOK_EOF
|
||||
} TokenType;
|
||||
typedef struct {
|
||||
TokenType toktype;
|
||||
char* toktext;
|
||||
} Token;
|
||||
|
||||
static Tokenizer
|
||||
initializeTokenizer(const char* txt) {
|
||||
return utf8NewIterator(txt);
|
||||
}
|
||||
|
||||
static void
|
||||
freeTokenizer(utf8iterator* iter) {
|
||||
utf8FreeIterator(iter);
|
||||
}
|
||||
|
||||
inline int
|
||||
greekChar(uint32_t ch) {
|
||||
return (((0x370 <= ch) && (ch <= 0x3ff)) ||
|
||||
((0x1f00 <= ch) && (ch <= 0x1fff)));
|
||||
}
|
||||
|
||||
inline int
|
||||
extendedPunctuation(uint32_t ch) {
|
||||
return ((0x2000 <= ch) && (ch <= 0x206f));
|
||||
}
|
||||
|
||||
inline int
|
||||
latinChar(uint32_t ch) {
|
||||
return (ch <= 0xff) || extendedPunctuation(ch);
|
||||
}
|
||||
|
||||
static Token
|
||||
nextToken(Tokenizer tokenizer) {
|
||||
int startIndex = tokenizer->byteIndex;
|
||||
uint32_t ch = utf8CharAt(tokenizer);
|
||||
Token result;
|
||||
memset(&result, 0, sizeof(result));
|
||||
if (ch == '\0') {
|
||||
result.toktype = TOK_EOF;
|
||||
result.toktext = NULL;
|
||||
return result;
|
||||
} else if (ch == '*') {
|
||||
utf8Advance(tokenizer);
|
||||
result.toktype = TOK_STAR;
|
||||
result.toktext = NULL;
|
||||
return result;
|
||||
} else if (greekChar(ch)) {
|
||||
while ((ch != 0) &&
|
||||
(greekChar(ch) || (ch == ' ') || (ch == ',') || (ch == '.'))) {
|
||||
utf8Advance(tokenizer);
|
||||
ch = utf8CharAt(tokenizer);
|
||||
}
|
||||
result.toktype = TOK_GREEK;
|
||||
result.toktext = strndup(tokenizer->txt + startIndex, tokenizer->byteIndex - startIndex);
|
||||
return result;
|
||||
} else if ((ch == '^') && (tokenizer->txt[tokenizer->byteIndex + 1] == '{')) {
|
||||
int idStart = 0, idEnd = 0;
|
||||
utf8Advance(tokenizer); /* to { */
|
||||
utf8Advance(tokenizer); /* to id */
|
||||
ch = utf8CharAt(tokenizer);
|
||||
idStart = tokenizer->byteIndex;
|
||||
while ((ch != 0) && (ch != '}')) {
|
||||
utf8Advance(tokenizer);
|
||||
ch = utf8CharAt(tokenizer);
|
||||
}
|
||||
idEnd = tokenizer->byteIndex;
|
||||
/* reached end-of-string or } */
|
||||
if (ch == '}')
|
||||
utf8Advance(tokenizer);
|
||||
result.toktype = TOK_REF;
|
||||
result.toktext = strndup(tokenizer->txt + idStart, idEnd - idStart);
|
||||
return result;
|
||||
} else if (latinChar(ch)) {
|
||||
while ((ch != 0) && latinChar(ch) && (ch != '*')) {
|
||||
utf8Advance(tokenizer);
|
||||
ch = utf8CharAt(tokenizer);
|
||||
if (ch == '^') {
|
||||
if (tokenizer->txt[tokenizer->byteIndex + 1] == '{') break;
|
||||
}
|
||||
}
|
||||
result.toktype = TOK_TEXT;
|
||||
result.toktext = strndup(tokenizer->txt + startIndex, tokenizer->byteIndex - startIndex);
|
||||
return result;
|
||||
} else {
|
||||
while ((ch != 0) && (!latinChar(ch) || (ch == ' ') || (ch == ',') || (ch == '.') || extendedPunctuation(ch))) {
|
||||
utf8Advance(tokenizer);
|
||||
ch = utf8CharAt(tokenizer);
|
||||
}
|
||||
result.toktype = TOK_UNICODE;
|
||||
result.toktext = strndup(tokenizer->txt + startIndex, tokenizer->byteIndex - startIndex);
|
||||
return result;
|
||||
}
|
||||
}
|
||||
|
||||
DEFINE_STACK(FormatElementQueue, FEQueueStack);
|
||||
|
||||
int formatText(const char* txt, FormatElement** dst, CitationRecordQueue* citationQPtr) {
|
||||
Tokenizer tokenizer = initializeTokenizer(txt);
|
||||
Token tok;
|
||||
int listLength = 0, em = 0;
|
||||
NEW_QUEUE(FormatElementQueue, formatElementQ);
|
||||
NEW_STACK(FEQueueStack, st);
|
||||
while ((tok = nextToken(tokenizer)).toktype != TOK_EOF) {
|
||||
if (tok.toktype == TOK_STAR) {
|
||||
if (em) { /* end emphasis */
|
||||
FormatElement* content = NULL;
|
||||
FormatElement elt = { .elementType = FORMAT_EM };
|
||||
QUEUE_TO_ARRAY(FormatElementQueue, formatElementQ, FormatElement, content);
|
||||
elt.elementContentLength = QUEUE_LENGTH(formatElementQ);
|
||||
elt.elementContent.nestedContent = content;
|
||||
formatElementQ = STACK_HEAD(st);
|
||||
APPEND_QUEUE(FormatElementQueue, formatElementQ, elt);
|
||||
POP_STACK(FEQueueStack, st);
|
||||
em = 0;
|
||||
} else { /* begin emphasis */
|
||||
PUSH_STACK(FEQueueStack, st, formatElementQ);
|
||||
REINIT_QUEUE(formatElementQ);
|
||||
em = 1;
|
||||
}
|
||||
} else {
|
||||
FormatElementType t;
|
||||
FormatElement elt = { .elementContent = { .textContent = tok.toktext } } ;
|
||||
if (tok.toktype == TOK_TEXT) {
|
||||
t = FORMAT_TEXT;
|
||||
} else if (tok.toktype == TOK_GREEK) {
|
||||
t = FORMAT_GREEK;
|
||||
} else if (tok.toktype == TOK_UNICODE) {
|
||||
t = FORMAT_UNICODE;
|
||||
} else if (tok.toktype == TOK_REF) {
|
||||
t = FORMAT_CITATION;
|
||||
if (citationQPtr && !lookupCitation(*citationQPtr, tok.toktext)) {
|
||||
addCitation(citationQPtr, tok.toktext);
|
||||
}
|
||||
}
|
||||
elt.elementType = t;
|
||||
APPEND_QUEUE(FormatElementQueue, formatElementQ, elt);
|
||||
}
|
||||
}
|
||||
if (em) {
|
||||
/* unmatched star -- close the emphasis here */
|
||||
FormatElement* content = NULL;
|
||||
FormatElement elt = { .elementType = FORMAT_EM };
|
||||
QUEUE_TO_ARRAY(FormatElementQueue, formatElementQ, FormatElement, content);
|
||||
elt.elementContentLength = QUEUE_LENGTH(formatElementQ);
|
||||
elt.elementContent.nestedContent = content;
|
||||
formatElementQ = STACK_HEAD(st);
|
||||
APPEND_QUEUE(FormatElementQueue, formatElementQ, elt);
|
||||
POP_STACK(FEQueueStack, st);
|
||||
}
|
||||
|
||||
QUEUE_TO_ARRAY(FormatElementQueue, formatElementQ, FormatElement, *dst);
|
||||
listLength = QUEUE_LENGTH(formatElementQ);
|
||||
DESTROY_QUEUE(FormatElementQueue, formatElementQ);
|
||||
freeTokenizer(tokenizer);
|
||||
return listLength;
|
||||
}
|
||||
|
||||
void freeFormatElementArray(FormatElement* a, int length) {
|
||||
int i = 0;
|
||||
for (i = 0; i < length; i++) {
|
||||
if (a[i].elementType == FORMAT_EM) {
|
||||
freeFormatElementArray(a[i].elementContent.nestedContent,
|
||||
a[i].elementContentLength);
|
||||
} else {
|
||||
free(a[i].elementContent.textContent);
|
||||
}
|
||||
}
|
||||
free(a);
|
||||
}
|
||||
|
||||
#ifdef FORMATTER_TEST
|
||||
#include <stdio.h>
|
||||
|
||||
const char* str = "My name in Chinese is \xe7\x86\x8a\xe5\xa4\xa7\xe8\xa1\x9b, or *xiao da wei*. My favorite Greek passage is \xe1\xbc\x90\xce\xbd \xe1\xbc\x80\xcf\x81\xcf\x87\xe1\xbf\x87 \xe1\xbc\xa6\xce\xbd \xe1\xbd\x81 \xce\xbb\xe1\xbd\xb9\xce\xb3\xce\xbf\xcf\x82.^{cite}";
|
||||
|
||||
int
|
||||
main() {
|
||||
FormatElement* lst;
|
||||
int l = formatText(str, &lst);
|
||||
return 0;
|
||||
}
|
||||
#endif
|
||||
Reference in New Issue
Block a user