snbt

1 files changed, 740 insertions, 0 deletions

diff --git a/src/snbt/read.c b/src/snbt/read.c
new file mode 100644
index 0000000..231404a
--- /dev/null
+++ b/src/snbt/read.c
@@ -0,0 +1,740 @@
+#include "snbt.h"
+#include "../lib.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+typedef enum {
+	TOK_RBRACE,
+	TOK_LBRACE,
+	TOK_RBRACK,
+	TOK_LBRACK,
+	TOK_COMMA,
+	TOK_COLON,
+	TOK_SEMI_COLON,
+	TOK_STRING,
+	TOK_IDENT,
+} tokentype_t;
+
+typedef union {
+	struct {
+		char *data;
+		uint16_t len;
+	} text;
+	double d;
+	int64_t l;
+} tokendata_t;
+
+typedef struct {
+	tokentype_t type;
+	tokendata_t data;
+} token_t;
+
+static char ret = '\0';
+
+static void snbt_free_token(token_t *token) {
+	if (token->type == TOK_IDENT || token->type == TOK_STRING) {
+		free(token->data.text.data);
+	}
+}
+
+static bool snbt_parse_string(token_t *token, const stream_t *stream, char delimiter) {
+	int len = 0;
+	int capacity = 8;
+	char *buf = xalloc(capacity * sizeof(char));
+
+	while (1) {
+		char c;
+		if (stream_read(stream, &c, 1) == false) {
+			free(buf);
+			return false;
+		}
+
+		if (c == delimiter)
+			break;
+
+		if (c == '\\') {
+			if (stream_read(stream, &c, 1) == false) {
+				free(buf);
+				return false;
+			}
+
+			if (c != '\\' && c != delimiter) {
+				free(buf);
+				return false;
+			}
+		}
+
+		if (len == capacity) {
+			capacity *= 2;
+			buf = xrealloc(buf, capacity);
+		}
+
+		buf[len++] = c;
+	}
+
+	token->data.text.data = xalloc(len * sizeof(char));
+	token->data.text.len = len;
+	memcpy(token->data.text.data, buf, len * sizeof(char));
+	free(buf);
+
+	return true;
+}
+
+bool snbt_allowed_ident(char c) {
+	if (c >= '0' && c <= '9')
+		return true;
+	if (c >= 'a' && c <= 'z')
+		return true;
+	if (c >= 'A' && c <= 'Z')
+		return true;
+	if (c == '_' || c == '-' || c == '.' || c == '+')
+		return true;
+	return false;
+}
+
+static bool snbt_parse_ident(token_t *token, const stream_t *stream, char first) {
+	int len = 0;
+	int capacity = 8;
+	char *buf = xalloc(capacity * sizeof(char));
+
+	buf[len++] = first;
+
+	while (1) {
+		char c;
+		if (stream_read(stream, &c, 1) == false) {
+			free(buf);
+			return false;
+		}
+
+		if (snbt_allowed_ident(c) == false) {
+			if (len == 0) {
+				free(buf);
+				return false;
+			} else {
+				ret = c;
+				break;
+			}
+		}
+
+		if (len == capacity) {
+			capacity *= 2;
+			buf = xrealloc(buf, capacity);
+		}
+
+		buf[len++] = c;
+	}
+
+	token->data.text.data = xalloc(len * sizeof(char));
+	token->data.text.len = len;
+	memcpy(token->data.text.data, buf, len * sizeof(char));
+	free(buf);
+
+	return true;
+}
+
+static bool snbt_next_token(token_t *token, const stream_t *stream) {
+
+	memset(token, 0, sizeof(token_t));
+
+	char c;
+	bool ok = true;
+
+retry:
+
+	c = ret;
+	ret = '\0';
+
+	if (c == '\0' && stream_read(stream, &c, 1) == false)
+		return false;
+
+	switch (c) {
+		case ' ':
+		case '\t':
+		case '\n':
+		case '\r':
+			goto retry;
+		case '{':
+			token->type = TOK_LBRACE;
+			break;
+		case '}':
+			token->type = TOK_RBRACE;
+			break;
+		case '[':
+			token->type = TOK_LBRACK;
+			break;
+		case ']':
+			token->type = TOK_RBRACK;
+			break;
+		case ':':
+			token->type = TOK_COLON;
+			break;
+		case ';':
+			token->type = TOK_SEMI_COLON;
+			break;
+		case ',':
+			token->type = TOK_COMMA;
+			break;
+		case '"':
+		case '\'':
+			token->type = TOK_STRING;
+			ok = snbt_parse_string(token, stream, c);
+			break;
+		default:
+			token->type = TOK_IDENT;
+			ok = snbt_parse_ident(token, stream, c);
+			break;
+	}
+
+	return ok;
+}
+
+static bool snbt_read_value(tag_t *tag, const stream_t *stream, token_t *first);
+static bool snbt_convert_ident(tag_t *tag, token_t token);
+
+static bool snbt_read_byte_array(tag_t *tag, const stream_t *stream) {
+	int capacity = 8;
+	int len = 0;
+	int8_t *buf = xalloc(capacity * sizeof(int8_t));
+
+	while(1) {
+		tag_t value;
+		if (snbt_read_value(&value, stream, NULL) == false) {
+			free(buf);
+			return false;
+		}
+
+		if (value.type != TAG_BYTE) {
+			tag_free(&value);
+			free(buf);
+			return false;
+		}
+		if (len == capacity) {
+			capacity *= 2;
+			buf = xrealloc(buf, capacity * sizeof(int8_t));
+		}
+		buf[len++] = value.data.b;
+		token_t token;
+		if (snbt_next_token(&token, stream) == false) {
+			free(buf);
+			return false;
+		}
+		if (token.type == TOK_COMMA)
+			continue;
+		if (token.type == TOK_RBRACK)
+			break;
+		free(buf);
+		return false;
+	}
+	tag->type = TAG_BYTE_ARRAY;
+	tag->data.b_arr.data = xalloc(len * sizeof(int8_t));
+	tag->data.b_arr.size = len;
+	memcpy(tag->data.b_arr.data, buf, len * sizeof(int8_t));
+	free(buf);
+
+	return true;
+}
+
+static bool snbt_read_int_array(tag_t *tag, const stream_t *stream) {
+	int capacity = 8;
+	int len = 0;
+	int32_t *buf = xalloc(capacity * sizeof(int32_t));
+
+	while(1) {
+		tag_t value;
+		if (snbt_read_value(&value, stream, NULL) == false) {
+			free(buf);
+			return false;
+		}
+		if (value.type != TAG_INT) {
+			tag_free(&value);
+			free(buf);
+			return false;
+		}
+		if (len == capacity) {
+			capacity *= 2;
+			buf = xrealloc(buf, capacity * sizeof(int32_t));
+		}
+		buf[len++] = value.data.i;
+		token_t token;
+		if (snbt_next_token(&token, stream) == false) {
+			free(buf);
+			return false;
+		}
+		if (token.type == TOK_COMMA)
+			continue;
+		if (token.type == TOK_RBRACK)
+			break;
+		free(buf);
+		return false;
+	}
+	tag->type = TAG_INT_ARRAY;
+	tag->data.i_arr.data = xalloc(len * sizeof(int32_t));
+	tag->data.i_arr.size = len;
+	memcpy(tag->data.i_arr.data, buf, len * sizeof(int32_t));
+	free(buf);
+
+	return true;
+}
+
+static bool snbt_read_long_array(tag_t *tag, const stream_t *stream) {
+	int capacity = 8;
+	int len = 0;
+	int64_t *buf = xalloc(capacity * sizeof(int64_t));
+
+	while(1) {
+		tag_t value;
+		if (snbt_read_value(&value, stream, NULL) == false) {
+			free(buf);
+			return false;
+		}
+		if (value.type != TAG_LONG) {
+			tag_free(&value);
+			free(buf);
+			return false;
+		}
+		if (len == capacity) {
+			capacity *= 2;
+			buf = xrealloc(buf, capacity * sizeof(int64_t));
+		}
+		buf[len++] = value.data.l;
+		token_t token;
+		if (snbt_next_token(&token, stream) == false) {
+			free(buf);
+			return false;
+		}
+		if (token.type == TOK_COMMA)
+			continue;
+		if (token.type == TOK_RBRACK)
+			break;
+		free(buf);
+		return false;
+	}
+	tag->type = TAG_LONG_ARRAY;
+	tag->data.l_arr.data = xalloc(len * sizeof(int64_t));
+	tag->data.l_arr.size = len;
+	memcpy(tag->data.l_arr.data, buf, len * sizeof(int64_t));
+	free(buf);
+
+	return true;
+}
+
+static bool snbt_read_array(tag_t *tag, tagtype_t type, const stream_t *stream) {
+	switch(type) {
+		case TAG_BYTE:
+			return snbt_read_byte_array(tag, stream);
+		case TAG_INT:
+			return snbt_read_int_array(tag, stream);
+		case TAG_LONG:
+			return snbt_read_long_array(tag, stream);
+		default:
+			return false;
+	}
+}
+
+static bool snbt_read_list(tag_t *tag, tag_t *first, token_t *tok_sav, const stream_t *stream) {
+	int capacity = 8;
+	int len = 0;
+	tag_t *tags = xalloc(sizeof(tag_t) * capacity);
+	tagtype_t type = TAG_END;
+
+	if (first != NULL) {
+		first->name_len = 0;
+		first->name = "";
+		tags[len++] = *first;
+		type = first->type;
+		goto endcheck;
+	}
+
+	if (tok_sav == NULL) {
+		if (snbt_next_token(tok_sav, stream) == false) {
+			free(tags);
+			return false;
+		}
+		if (tok_sav->type == TOK_RBRACK) {
+			goto end;
+		}
+	}
+
+	while(1) {
+
+		tag_t value;
+		token_t check;
+
+		if (snbt_read_value(&value, stream, tok_sav) == false) {
+			snbt_free_token(tok_sav);
+			free(tags);
+			return false;
+		}
+
+		tok_sav = NULL;
+
+		if (type != TAG_END && value.type != type) {
+			free(tags);
+			return false;
+		}
+
+		if (len == capacity) {
+			capacity *= 2;
+			tags = xrealloc(tags, capacity * sizeof(tag_t));
+		}
+
+		tags[len++] = value;
+
+endcheck:
+
+		if (snbt_next_token(&check, stream) == false) {
+			free(tags);
+			return false;
+		}
+
+		if (check.type == TOK_COMMA)
+			continue;
+
+		if (check.type == TOK_RBRACK)
+			break;
+			
+		free(tags);
+		return false;
+	}
+end:
+	tag->type = TAG_LIST;
+	tag->data.list.type = type;
+	tag->data.list.size = len;
+	tag->data.list.tags = xalloc(len * sizeof(tag_t));
+	memcpy(tag->data.list.tags, tags, len *sizeof(tag_t));
+	free(tags);
+
+	return true;
+}
+
+static bool snbt_read_collection(tag_t *tag, const stream_t *stream) {
+
+	token_t first;
+
+	if (snbt_next_token(&first, stream) == false)
+		return false;
+
+	if (first.type == TOK_RBRACK) {
+		tag->type = TAG_LIST;
+		tag->data.list.type = TAG_END;
+		tag->data.list.size = 0;
+		tag->data.list.tags = NULL;
+		return true;
+	}
+
+	if (first.type == TOK_IDENT && first.data.text.len == 1) {
+		tagtype_t type;
+		char c = first.data.text.data[0];
+		switch (c) {
+			case 'B':
+				type = TAG_BYTE;
+				break;
+			case 'I':
+				type = TAG_INT;
+				break;
+			case 'L':
+				type = TAG_LONG;
+				break;
+			default: {
+				tag_t new;
+				if (snbt_convert_ident(&new, first) == false) {
+					snbt_free_token(&first);
+					return false;
+				}
+				return snbt_read_list(tag, &new, NULL, stream);
+			}
+		}
+
+		token_t second;
+		if (snbt_next_token(&second, stream) == false) {
+			snbt_free_token(&first);
+			return false;
+		}
+
+		if (second.type == TOK_COMMA) {
+			tag_t new;
+			new.type = TAG_STRING;
+			new.data.string.data = first.data.text.data;
+			new.data.string.size = first.data.text.len;
+			return snbt_read_list(tag, &new, NULL, stream);
+		}
+		
+		snbt_free_token(&first);
+
+		if (second.type == TOK_SEMI_COLON) {
+			return snbt_read_array(tag, type, stream);
+		}
+
+		snbt_free_token(&second);
+		return false;
+
+	}
+
+	return snbt_read_list(tag, NULL, &first, stream);
+}
+
+static bool snbt_read_compound(tagdata_t *data, const stream_t *stream) {
+	map_t map;
+	map_init(&map);
+
+	token_t next = {0};
+	if (snbt_next_token(&next, stream) == false) {
+		snbt_free_token(&next);
+		return false;
+	}
+
+	if (next.type == TOK_RBRACE) {
+		data->compound = map;
+		return true;
+	}
+
+	while (1) {
+
+		if (next.type != TOK_STRING && next.type != TOK_IDENT) {
+			map_free(&map);
+			snbt_free_token(&next);
+			return false;
+		}
+
+		char *name = next.data.text.data;
+		int name_len = next.data.text.len;
+
+		if (name_len < 1) {
+			map_free(&map);
+			free(name);
+			return false;
+		}
+		
+		if (snbt_next_token(&next, stream) == false || next.type != TOK_COLON) {
+			map_free(&map);
+			free(name);
+			return false;
+		}
+		
+		tag_t value;
+		if (snbt_read_value(&value, stream, NULL) == false) {
+			map_free(&map);
+			free(name);
+			return false;
+		}
+
+		value.name = name;
+		value.name_len = name_len;
+
+		map_put(&map, &value);
+
+		if (snbt_next_token(&next, stream) == false) {
+			map_free(&map);
+			snbt_free_token(&next);
+			return false;
+		}
+
+		if (next.type == TOK_COMMA) {
+			if (snbt_next_token(&next, stream) == false) {
+				map_free(&map);
+				return false;
+			}
+			continue;
+		} else if (next.type == TOK_RBRACE) {
+			break;
+		} else {
+			map_free(&map);
+			snbt_free_token(&next);
+			return false;
+		}
+
+	}
+
+	data->compound = map;
+
+	return true;
+}
+
+static bool snbt_convert_decimal(tag_t *tag, char *text) {
+	uint16_t len = strlen(text);
+
+	char *end = NULL;
+
+	double d = strtod(text, &end);
+
+	char c = *end;
+	bool check1 = end == &text[len - 1] && (
+		c == 'f' ||
+		c == 'F' ||
+		c == 'd' ||
+		c == 'D'
+	);
+	bool check2 = end == &text[len] && c == '\0';
+	if (!check1 && !check2) {
+		return false;
+	}
+	
+	if (tag->type == TAG_FLOAT) {
+		tag->data.f = (float)d;
+	} else {
+		tag->data.d = d;
+	}
+
+	return true;
+}
+
+static bool snbt_convert_int(tag_t *tag, char *text) {
+	uint16_t len = strlen(text);
+
+	char *end = NULL;
+
+	uint64_t i = strtol(text, &end, 10);
+
+	char c = *end;
+	bool check1 = end == &text[len - 1] && (
+		c == 'b' ||
+		c == 'B' ||
+		c == 's' ||
+		c == 'S' ||
+		c == 'l' ||
+		c == 'L'
+	);
+	bool check2 = end == &text[len] && c == '\0';
+	if (!check1 && !check2) {
+		return false;
+	}
+
+	if (tag->type == TAG_BYTE) {
+		tag->data.b = (uint8_t)i;
+	} else if (tag->type == TAG_SHORT) {
+		tag->data.s = (uint16_t)i;
+	} else if (tag->type == TAG_INT) {
+		tag->data.i = (uint32_t)i;
+	} else {
+		tag->data.l = i;
+	}
+
+	return true;
+}
+
+static bool snbt_convert_ident(tag_t *tag, token_t token) {
+	uint16_t len = token.data.text.len + 1;
+	char *text = xalloc(len * sizeof(char));
+	memcpy(text, token.data.text.data, len - 1);
+	text[len - 1] = '\0';
+
+	if (len == 4 && memcmp(text, "true", 4) == 0) {
+		tag->type = TAG_BYTE;
+		tag->data.b = 1;
+	} else if (len == 5 && memcmp(text, "false", 5) == 0) {
+		tag->type = TAG_BYTE;
+		tag->data.b = 0;
+	}
+
+	// try to parse as a number lol he he ha ha
+	char end = text[len - 2];
+
+	bool ok = false;
+
+	switch (end) {
+		case 'f':
+		case 'F':
+			tag->type = TAG_FLOAT;
+			ok = snbt_convert_decimal(tag, text);
+			break;
+		case 'd':
+		case 'D':
+			tag->type = TAG_DOUBLE;
+			ok = snbt_convert_decimal(tag, text);
+			break;
+		case 'b':
+		case 'B':
+			tag->type = TAG_BYTE;
+			ok = snbt_convert_int(tag, text);
+			break;
+		case 's':
+		case 'S':
+			tag->type = TAG_SHORT;
+			ok = snbt_convert_int(tag, text);
+			break;
+		case 'l':
+		case 'L':
+			tag->type = TAG_LONG;
+			ok = snbt_convert_int(tag, text);
+			break;
+		default:
+			break;
+	}
+
+	if (ok) {
+		free(text);
+		snbt_free_token(&token);
+		return true;
+	}
+
+	// unkown type try long >> double >> string (fallback)
+	
+	tag->type = TAG_INT;
+	if (snbt_convert_int(tag, text) == true) {
+		free(text);
+		snbt_free_token(&token);
+		return true;
+	}
+
+	tag->type = TAG_DOUBLE;
+	if (snbt_convert_decimal(tag, text) == true) {
+		free(text);
+		snbt_free_token(&token);
+		return true;
+	}
+
+	tag->type = TAG_STRING;
+	tag->data.string.data = token.data.text.data;
+	tag->data.string.size = token.data.text.len;
+	free(text);
+	return true;
+}
+
+static bool snbt_read_value(tag_t *tag, const stream_t *stream, token_t *first) {
+	
+	token_t token;
+
+	if (first != NULL)
+		token = *first;
+	else if (snbt_next_token(&token, stream) == false)
+		return false;
+
+	tag->name = "";
+	tag->name_len = 0;
+
+	bool ok = true;
+
+	switch (token.type) {
+        case TOK_RBRACK:
+        case TOK_RBRACE:
+        case TOK_COLON:
+		case TOK_SEMI_COLON:
+        case TOK_COMMA:
+			ok = false;
+			break;
+        case TOK_LBRACK:
+			tag->type = TAG_LIST;
+			ok = snbt_read_collection(tag, stream);
+			break;
+        case TOK_LBRACE:
+			tag->type = TAG_COMPOUND;
+			ok = snbt_read_compound(&tag->data, stream);
+			break;
+        case TOK_STRING:
+			tag->type = TAG_STRING;
+			tag->data.string.data = token.data.text.data;
+			tag->data.string.size = token.data.text.len;
+			break;
+		case TOK_IDENT:
+			ok = snbt_convert_ident(tag, token);
+			break;
+    }
+
+	return ok;
+}
+
+bool snbt_read(tag_t *tag, const stream_t *stream) {
+	return snbt_read_value(tag, stream, NULL);
+}