first working version with minimal syntax (check test.nhtml)

Makefile

@@ -1,8 +1,10 @@
-OBJ:= main.o
+OBJ:= html.o main.o
 OUTPUT:=nhtmlc
 
-debug: CXXFLAGS = -g 
+all: debug
-debug: CFLAGS = -g 
+
+debug: CXXFLAGS = -g -DDEBUG
+debug: CFLAGS = -g -DDEBUG
 debug: $(OBJ)
 	gcc $(OBJ) -g -o $(OUTPUT)
 

main.c

@@ -2,6 +2,12 @@
 #include <stdlib.h> // needs: realloc, malloc, free
 #include <string.h> // needs: strcmp, memcpy
 #include <errno.h> 
+#include <ctype.h> // needs: isspace
+
+#include <getopt.h> // needs: getopt_long
+
+#include "html.h" // needs: html_escaped
+
 typedef struct {
 	char * c_str;
 	size_t len;
@@ -23,6 +29,7 @@ int string_append(string_t *str, char c) {
 	str->c_str[str->len-1] = c;
 	str->c_str[str->len] = 0; // make sure
 	str->len++;
+	return 0;
 }
 
 void string_destroy(string_t s) {
@@ -62,6 +69,8 @@ attr_t * attr_set_find(attr_set_t *set, const char * name) {
 
 // Append new Attribute to set
 int attr_set_append(attr_set_t * set, attr_t *new_entry) {
+	if (new_entry->name.c_str== NULL) return -1; // reject empty entries
+
 	// search first
 	attr_t * new_ptr = attr_set_find(set, new_entry->name.c_str);
 	if (new_ptr != NULL) {
@@ -84,15 +93,90 @@ int attr_set_append(attr_set_t * set, attr_t *new_entry) {
 	return 0;
 }
 
+void attr_reset(attr_t * attr) {
+	memset(attr, 0, sizeof(attr_t));
+}
+
+int strip(FILE * stream) {
+	int current = 0;
+	while((current = fgetc(stream)) != EOF) {
+		if (!isspace(current))
+			break;
+	}
+
+	return current;
+}
+
 int parse_attr(FILE * stream, attr_set_t * output) {
-	fprintf(stderr, "stub: parse_attr\n");
+	attr_t current_attr = {};
-	return 0;
+#ifdef DEBUG
+	printf("parse_attr\n");
+#endif
+	int buffer = 0;
+	unsigned char isKey = 1;
+
+	while((buffer = fgetc(stream)) != EOF) {
+		// parse key=value pairs
+		// check for delim
+		//if (buffer == ' ' || buffer == '\t' || buffer == '\n'){
+		if (isspace(buffer)) {
+			attr_set_append(output, &current_attr);
+#ifdef DEBUG
+			printf("parsed attr: %s=%s\n", current_attr.name.c_str, current_attr.value);
+#endif
+			// reset attribute
+			attr_reset(&current_attr);
+			isKey = 1;
+			continue;
+		}
+		if (buffer == '=') {
+			isKey = 0;
+			continue;
+		} else
+		if (buffer == ']') {
+			break;
+		}
+
+		if (isKey) {
+			string_append(&current_attr.name, buffer);
+		} else {
+			string_append(&current_attr.value, buffer);
+		}
+	}
+
+	// append last attribute
+	attr_set_append(output, &current_attr);
+
+	//memset(output, 0, sizeof(attr_set_t));
+	return strip(stream);
 }
 
 int parse_text(FILE * stream, FILE * output) {
-	// STUB
+#ifdef DEBUG
-	fprintf(stderr, "stub: parse_text\n");
+	printf("parse text\n");
-	return 0;
+#endif
+	int buffer = 0;
+	char escaped = 0;
+
+	while((buffer = fgetc(stream)) != EOF) {
+		if (!escaped && buffer == '\\') {
+			escaped = 1;
+			continue;
+		}
+		if (!escaped && buffer == '"') {
+			break;
+		}
+		escaped = 0;
+
+		if (html_escape(buffer, output)) {
+			continue;
+		}
+
+		fputc(buffer, output);
+	}
+
+	//fprintf(stderr, "stub: parse_text\n");
+	return strip(stream);
 }
 
 typedef struct node {
@@ -102,7 +186,10 @@ typedef struct node {
 
 
 // write out node
-void print_node(node_t * node, FILE * output) {
+void open_node(node_t * node, FILE * output) {
+	// check for empty node (text mostly)
+	if (!node->name.c_str)  return;
+
 	// start by writing tag
 	fputc('<', output);
 
@@ -113,43 +200,99 @@ void print_node(node_t * node, FILE * output) {
 	size_t i = 0;
 	attr_t * current = node->attributes.arr;
 	for (; i < node->attributes.len; i++,current++) {
-		fprintf(output, " %s=\"%s\"", current->name.c_str, current->value.c_str);
+		if (current->value.c_str)
+			fprintf(output, " %s=\"%s\"", current->name.c_str, current->value.c_str);
+		else {
+			fputc(' ', output);
+			fputs(current->name.c_str, output);
+		}
 	}
 
 	// close tag
 	fputc('>', output);
 }
+void close_node(node_t * node, FILE * output) {
+	if (node->name.c_str) {
+		fprintf(output, "</%s>", node->name.c_str);
+	}
+}
 
-int print_tag(FILE * stream, FILE * output) {
+int readName(FILE *stream, node_t *node) {
-	node_t new_node;
+	int current = strip(stream);
-	int buffer = 0;
+	if (current == EOF) {
-
+		return current;
-	while ((buffer = fgetc(stream)) != EOF) {
-		switch(buffer) {
-		case '[': // parse attributes
-			if (parse_attr(stream, &new_node.attributes) != -1) {
-				fprintf(stderr, "could not parse attributes: %d", -1);
-				return -1;
-			}
-			break;
-		case '"':
-			// raw text
-			parse_text(stream, output);
-			break;
-		default:
-			if (buffer != ' ' && buffer != '\t' && buffer != '\n') {
-				string_append(&new_node.name, buffer);
-			}
-		}
 	}
 
+	do {
+		if (current == '{' || current == '[') {
+			break;
+		}
 
-	return 0;
+		if (current == ' ' || current == '\t' || current == '\n') {
+			current = strip(stream);
+			break;
+		}
+
+		string_append(&node->name, current);
+	} while((current = fgetc(stream)) != EOF);
+
+	return current;
+}
+
+
+enum node_type {
+	NODE_SELFCLOSING,
+	NODE_TEXT,
+	NODE_TAG
+};
+
+int parse_node(int current, FILE * stream, FILE * output) {
+	if (current == '"') {
+		return parse_text(stream, output);
+	}
+
+	// normal node
+	node_t current_node = {};
+	string_append(&current_node.name, current);
+	current = readName(stream, &current_node);
+	if (current == EOF) {
+		goto done;
+	}
+#ifdef DEBUG
+	printf("parse_node: %s\n", current_node.name.c_str);
+#endif
+	if (current == '[') {
+		current = parse_attr(stream, &current_node.attributes);
+	}
+
+	if (current != '{') {
+		// tag is selfclosing
+		open_node(&current_node, output);
+		return current;
+	}
+	current = strip(stream);
+
+	open_node(&current_node, output);
+	while(current != '}' && current != EOF) {
+		current = parse_node(current, stream, output);
+	}
+
+	close_node(&current_node, output);
+done:
+	return strip(stream);
 }
 
 int main(int argc, char ** args) {
+	//getopt_long(argc, args, NULL, NULL, NULL);
+
 	char * filename= NULL;
-	for(;argc != 1; --argc) {
+	FILE * output = fopen("a.html", "w");
+	if (output == NULL) {
+		fprintf(stderr, "could not create output file\n");
+		return -1;
+	}
+
+	for(;argc > 1; --argc) {
 		filename = args[argc-1];
 		printf("starting conversion of %s\n", filename);
 		
@@ -160,16 +303,19 @@ int main(int argc, char ** args) {
 			continue;
 		}
 		
-		while(!feof(handle)) {
+		int current = strip(handle);
-			if (print_tag(handle, stdout) != 0) {
+
-				fprintf(stderr, "error during parsing of node\n");
+		while((current = parse_node(current, handle, output)) != EOF) {
-				break;
+//			if () {
-			}
+//				fprintf(stderr, "error during parsing of node\n");
+//				break;
+//			}
 		}	
 
 		fclose(handle);
 	}
 
+	fclose(output);
 	printf("done compiling\n");
 	return 0;
 }

test.nhtml

@@ -1,12 +1,28 @@
-div {
+!DOCTYPE[html]
-	"hi i'm raw text"
+head {
-	"I'm raw text too :)" 
+	title { "was geht" }
-	"Including \"escaping\""
+	meta[charset=utf-8]
-}
-h1 {
-div {
-	"Another, but bigger font"
-	"Adding a 
-	new line adds a <br/> :)" 
 }
+body {
+	h1 { "Alle Meine Endchen schwimmen auf dem See" }
+	p {
+		"Alle meine Entchen schwimmen auf dem See,
+		schwimmen auf dem See,
+		Köpfchen in das Wasser, Schwänzchen in die Höh'." 
+	}
+	p {
+		"Alle meine Täubchen gurren auf dem Dach,
+		gurren auf dem Dach,
+		eins fliegt in die Lüfte, fliegen alle nach." 
+	}
+	p {
+		"Alle meine Hühner scharren in dem Stroh,
+		scharren in dem Stroh,
+		finden sie ein Körnchen, sind sie alle froh."
+	}
+	p {
+		"Alle meine Gänschen watscheln durch den Grund,
+		watscheln durch den Grund,
+		suchen in dem Tümpel, werden kugelrund."
+	}
 }