From ba22a750093f5b918aefd58ba698f38dff6a14e1 Mon Sep 17 00:00:00 2001
From: Jan Breuer <jan.breuer@jaybee.cz>
Date: ćšć, 20 6æ 2013 22:03:49 +0800
Subject: [PATCH] Unexpected character handling
---
libscpi/src/lexer.c | 337 ++++++++++++++++++++++++++++++-------------------------
1 files changed, 184 insertions(+), 153 deletions(-)
diff --git a/libscpi/src/lexer.c b/libscpi/src/lexer.c
index 0b3bf09..65824df 100644
--- a/libscpi/src/lexer.c
+++ b/libscpi/src/lexer.c
@@ -40,23 +40,22 @@
#include "scpi/lexer.h"
-
/* identify character */
-static int isws ( int c ) {
+static int isws(int c) {
if ((c == ' ') || (c == '\t')) {
return 1;
}
return 0;
}
-static int isbdigit ( int c ) {
+static int isbdigit(int c) {
if ((c == '0') || (c == '1')) {
return 1;
}
return 0;
}
-static int isqdigit ( int c ) {
+static int isqdigit(int c) {
if ((c == '0') || (c == '1') || (c == '2') || (c == '3') || (c == '4') || (c == '5') || (c == '6') || (c == '7')) {
return 1;
}
@@ -71,6 +70,10 @@
}
}
+int SCPI_LexIsEos(lex_state_t * state) {
+ return iseos(state);
+}
+
static int ischr(lex_state_t * state, char chr) {
return (state->pos[0] == chr);
}
@@ -80,31 +83,36 @@
}
static int isH(int c) {
- return c == 'h' || c == 'H';
+ return c == 'h' || c == 'H';
}
static int isB(int c) {
- return c == 'b' || c == 'B';
+ return c == 'b' || c == 'B';
}
static int isQ(int c) {
- return c == 'q' || c == 'Q';
+ return c == 'q' || c == 'Q';
}
static int isE(int c) {
- return c == 'e' || c == 'E';
+ return c == 'e' || c == 'E';
}
+
+#define SKIP_NONE 0
+#define SKIP_OK 1
+#define SKIP_INCOMPLETE -1
/* skip characters */
/* 7.4.1 <PROGRAM MESSAGE UNIT SEPARATOR>*/
// TODO: static int skipProgramMessageUnitSeparator(lex_state_t * state)
+
static int skipWs(lex_state_t * state) {
int someSpace = 0;
- while(!iseos(state) && isws(state->pos[0])) {
+ while (!iseos(state) && isws(state->pos[0])) {
state->pos++;
someSpace++;
}
-
+
return someSpace;
}
@@ -115,17 +123,17 @@
// static int skipProgramMessageTerminator(lex_state_t * state)
static int skipDigit(lex_state_t * state) {
- if(!iseos(state) && isdigit(state->pos[0])) {
+ if (!iseos(state) && isdigit(state->pos[0])) {
state->pos++;
- return 1;
+ return SKIP_OK;
} else {
- return 0;
+ return SKIP_NONE;
}
}
static int skipNumbers(lex_state_t * state) {
int someNumbers = 0;
- while(!iseos(state) && isdigit(state->pos[0])) {
+ while (!iseos(state) && isdigit(state->pos[0])) {
state->pos++;
someNumbers++;
}
@@ -133,258 +141,283 @@
}
static int skipPlusmn(lex_state_t * state) {
- if(!iseos(state) && isplusmn(state->pos[0])) {
+ if (!iseos(state) && isplusmn(state->pos[0])) {
state->pos++;
- return 1;
+ return SKIP_OK;
} else {
- return 0;
+ return SKIP_NONE;
}
}
static int skipAlpha(lex_state_t * state) {
int someLetters = 0;
- while(!iseos(state) && isalpha(state->pos[0])) {
+ while (!iseos(state) && isalpha(state->pos[0])) {
state->pos++;
someLetters++;
}
- return someLetters;
+ return someLetters;
}
static int skipChr(lex_state_t * state, int chr) {
- if(!iseos(state) && ischr(state, chr)) {
+ if (!iseos(state) && ischr(state, chr)) {
state->pos++;
- return 1;
+ return SKIP_OK;
} else {
- return 0;
+ return SKIP_NONE;
}
}
static int skipSlashDot(lex_state_t * state) {
- if(!iseos(state) && (ischr(state, '/') | ischr(state, '.'))) {
+ if (!iseos(state) && (ischr(state, '/') | ischr(state, '.'))) {
state->pos++;
- return 1;
+ return SKIP_OK;
} else {
- return 0;
- }
+ return SKIP_NONE;
+ }
}
static int skipStar(lex_state_t * state) {
- if(!iseos(state) && ischr(state, '*')) {
+ if (!iseos(state) && ischr(state, '*')) {
state->pos++;
- return 1;
+ return SKIP_OK;
} else {
- return 0;
- }
+ return SKIP_NONE;
+ }
}
static int skipColon(lex_state_t * state) {
- if(!iseos(state) && ischr(state, ':')) {
+ if (!iseos(state) && ischr(state, ':')) {
state->pos++;
- return 1;
+ return SKIP_OK;
} else {
- return 0;
- }
+ return SKIP_NONE;
+ }
}
/* 7.6.1.2 <COMMAND PROGRAM HEADER> */
static int skipProgramMnemonic(lex_state_t * state) {
const char * startPos = state->pos;
- if(!iseos(state) && isalpha(state->pos[0])) {
+ if (!iseos(state) && isalpha(state->pos[0])) {
state->pos++;
- while(!iseos(state) && (isalnum(state->pos[0]) || ischr(state, '_'))) {
+ while (!iseos(state) && (isalnum(state->pos[0]) || ischr(state, '_'))) {
state->pos++;
}
}
-
- return state->pos - startPos;
+
+ if (iseos(state)) {
+ return (state->pos - startPos) * SKIP_INCOMPLETE;
+ } else {
+ return (state->pos - startPos) * SKIP_OK;
+ }
}
/* tokens */
int SCPI_LexWhiteSpace(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
skipWs(state);
-
- token->len = state->pos - token->ptr;
-
+
+ token->len = state->pos - token->ptr;
+
if (token->len > 0) {
token->type = TokWhiteSpace;
} else {
token->type = TokUnknown;
- }
-
+ }
+
return token->len;
}
/* 7.6.1 <COMMAND PROGRAM HEADER> */
static int skipCommonProgramHeader(lex_state_t * state) {
+ int res;
if (skipStar(state)) {
- if(!skipProgramMnemonic(state)) {
- state->pos--;
+ res = skipProgramMnemonic(state);
+ if (res == SKIP_NONE && iseos(state)) {
+ return SKIP_INCOMPLETE;
+ } else if (res <= SKIP_INCOMPLETE) {
+ return SKIP_OK;
+ } else if (res >= SKIP_OK) {
+ return SKIP_OK;
} else {
- return 1;
+ return SKIP_INCOMPLETE;
}
}
- return 0;
+ return SKIP_NONE;
}
static int skipCompoundProgramHeader(lex_state_t * state) {
- const char * rollback = state->pos;
-
- skipColon(state);
-
- if(skipProgramMnemonic(state)) {
- while(skipColon(state)) {
- if(!skipProgramMnemonic(state)) {
- state->pos = rollback;
- return 0;
+ int res;
+ int firstColon = skipColon(state);
+
+ res = skipProgramMnemonic(state);
+ if (res >= SKIP_OK) {
+ while (skipColon(state)) {
+ res = skipProgramMnemonic(state);
+ if (res <= SKIP_INCOMPLETE) {
+ return SKIP_OK;
+ } else if (res == SKIP_NONE) {
+ return SKIP_INCOMPLETE;
}
}
- return 1;
+ return SKIP_OK;
+ } else if (res <= SKIP_INCOMPLETE) {
+ return SKIP_OK;
+ } else if (firstColon) {
+ return SKIP_INCOMPLETE;
} else {
- state->pos = rollback;
- return 0;
+ return SKIP_NONE;
}
}
-int SCPI_LexProgramHeader(lex_state_t * state, token_t * token) {
+int SCPI_LexProgramHeader(lex_state_t * state, token_t * token) {
+ int res;
token->ptr = state->pos;
token->type = TokUnknown;
-
- if(skipCommonProgramHeader(state)) {
- if (skipChr(state, '?')) {
+
+ res = skipCommonProgramHeader(state);
+ if (res >= SKIP_OK) {
+ if (skipChr(state, '?') >= SKIP_OK) {
token->type = TokCommonQueryProgramHeader;
} else {
token->type = TokCommonProgramHeader;
}
- } else if(skipCompoundProgramHeader(state)) {
- if (skipChr(state, '?')) {
- token->type = TokCompoundQueryProgramHeader;
- } else {
- token->type = TokCompoundProgramHeader;
- }
+ } else if (res <= SKIP_INCOMPLETE) {
+ token->type = TokIncompleteCommonProgramHeader;
+ } else if (res == SKIP_NONE) {
+ res = skipCompoundProgramHeader(state);
+
+ if (res >= SKIP_OK) {
+ if (skipChr(state, '?') >= SKIP_OK) {
+ token->type = TokCompoundQueryProgramHeader;
+ } else {
+ token->type = TokCompoundProgramHeader;
+ }
+ } else if (res <= SKIP_INCOMPLETE) {
+ token->type = TokIncompleteCompoundProgramHeader;
+ }
}
-
+
if (token->type != TokUnknown) {
token->len = state->pos - token->ptr;
} else {
token->len = 0;
state->pos = token->ptr;
}
-
- return token->len;
+
+ return token->len;
}
/* 7.7.1 <CHARACTER PROGRAM DATA> */
int SCPI_LexCharacterProgramData(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
- if(!iseos(state) && isalpha(state->pos[0])) {
+
+ if (!iseos(state) && isalpha(state->pos[0])) {
state->pos++;
- while(!iseos(state) && (isalnum(state->pos[0]) || ischr(state, '_'))) {
+ while (!iseos(state) && (isalnum(state->pos[0]) || ischr(state, '_'))) {
state->pos++;
}
}
-
+
token->len = state->pos - token->ptr;
- if(token->len > 0) {
+ if (token->len > 0) {
token->type = TokProgramMnemonic;
} else {
token->type = TokUnknown;
}
-
+
return token->len;
}
/* 7.7.2 <DECIMAL NUMERIC PROGRAM DATA> */
static int skipMantisa(lex_state_t * state) {
int someNumbers = 0;
-
+
skipPlusmn(state);
-
+
someNumbers += skipNumbers(state);
-
- if(skipChr(state, '.')) {
+
+ if (skipChr(state, '.')) {
someNumbers += skipNumbers(state);
}
-
+
return someNumbers;
}
static int skipExponent(lex_state_t * state) {
int someNumbers = 0;
-
- if(!iseos(state) && isE(state->pos[0])) {
+
+ if (!iseos(state) && isE(state->pos[0])) {
state->pos++;
-
+
skipWs(state);
-
+
skipPlusmn(state);
-
+
someNumbers = skipNumbers(state);
}
-
+
return someNumbers;
}
int SCPI_LexDecimalNumericProgramData(lex_state_t * state, token_t * token) {
const char * rollback;
token->ptr = state->pos;
-
+
if (skipMantisa(state)) {
rollback = state->pos;
skipWs(state);
- if(!skipExponent(state)) {
+ if (!skipExponent(state)) {
state->pos = rollback;
}
} else {
state->pos = token->ptr;
}
-
+
token->len = state->pos - token->ptr;
- if(token->len > 0) {
+ if (token->len > 0) {
token->type = TokDecimalNumericProgramData;
} else {
token->type = TokUnknown;
}
-
+
return token->len;
}
/* 7.7.3 <SUFFIX PROGRAM DATA> */
int SCPI_LexSuffixProgramData(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
skipChr(state, '/');
-
+
// TODO: strict parsing : SLASH? (ALPHA+ (MINUS? DIGIT)?) ((SLASH | DOT) (ALPHA+ (MINUS? DIGIT)?))*
if (skipAlpha(state)) {
skipChr(state, '-');
skipDigit(state);
-
+
while (skipSlashDot(state)) {
skipAlpha(state);
skipChr(state, '-');
- skipDigit(state);
+ skipDigit(state);
}
}
-
+
token->len = state->pos - token->ptr;
- if((token->len > 0)) {
+ if ((token->len > 0)) {
token->type = TokSuffixProgramData;
} else {
token->type = TokUnknown;
state->pos = token->ptr;
token->len = 0;
}
-
+
return token->len;
}
/* 7.7.4 <NONDECIMAL NUMERIC PROGRAM DATA> */
static int skipHexNum(lex_state_t * state) {
int someNumbers = 0;
- while(!iseos(state) && isxdigit(state->pos[0])) {
+ while (!iseos(state) && isxdigit(state->pos[0])) {
state->pos++;
someNumbers++;
}
@@ -393,7 +426,7 @@
static int skipOctNum(lex_state_t * state) {
int someNumbers = 0;
- while(!iseos(state) && isqdigit(state->pos[0])) {
+ while (!iseos(state) && isqdigit(state->pos[0])) {
state->pos++;
someNumbers++;
}
@@ -402,35 +435,34 @@
static int skipBinNum(lex_state_t * state) {
int someNumbers = 0;
- while(!iseos(state) && isbdigit(state->pos[0])) {
+ while (!iseos(state) && isbdigit(state->pos[0])) {
state->pos++;
someNumbers++;
}
return someNumbers;
}
-
int SCPI_LexNondecimalNumericData(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
int someNumbers = 0;
- if(skipChr(state, '#')) {
- if(!iseos(state)) {
- if(isH(state->pos[0])) {
+ if (skipChr(state, '#')) {
+ if (!iseos(state)) {
+ if (isH(state->pos[0])) {
state->pos++;
someNumbers = skipHexNum(state);
token->type = TokHexnum;
- } else if(isQ(state->pos[0])) {
+ } else if (isQ(state->pos[0])) {
state->pos++;
someNumbers = skipOctNum(state);
token->type = TokOctnum;
- } else if(isB(state->pos[0])) {
+ } else if (isB(state->pos[0])) {
state->pos++;
someNumbers = skipBinNum(state);
token->type = TokBinnum;
}
}
}
-
+
if (someNumbers) {
token->ptr += 2; // ignore number prefix
token->len = state->pos - token->ptr;
@@ -442,15 +474,13 @@
return token->len > 0 ? token->len + 2 : 0;
}
-
/* 7.7.5 <STRING PROGRAM DATA> */
static int isascii7bit(int c) {
return (c >= 0) && (c <= 0x7f);
}
-
static int skipQuoteProgramData(lex_state_t * state, int quote) {
- while(!iseos(state)) {
+ while (!iseos(state)) {
if (isascii7bit(state->pos[0]) && !ischr(state, quote)) {
state->pos++;
} else if (ischr(state, quote)) {
@@ -464,7 +494,7 @@
}
}
}
-
+
static int skipDoubleQuoteProgramData(lex_state_t * state) {
skipQuoteProgramData(state, '"');
}
@@ -473,9 +503,9 @@
skipQuoteProgramData(state, '\'');
}
-int SCPI_LexStringProgramData(lex_state_t * state, token_t * token) {
+int SCPI_LexStringProgramData(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
if (!iseos(state)) {
if (ischr(state, '"')) {
state->pos++;
@@ -499,10 +529,10 @@
}
}
}
-
+
token->len = state->pos - token->ptr;
-
- if((token->len > 0)) {
+
+ if ((token->len > 0)) {
token->ptr++;
token->len -= 2;
} else {
@@ -510,7 +540,7 @@
state->pos = token->ptr;
token->len = 0;
}
-
+
return token->len > 0 ? token->len + 2 : 0;
}
@@ -530,8 +560,8 @@
/* Get number of digits */
i = state->pos[0] - '0';
state->pos++;
-
- for(; i > 0; i--) {
+
+ for (; i > 0; i--) {
if (!iseos(state) && isdigit(state->pos[0])) {
j *= 10;
j += (state->pos[0] - '0');
@@ -540,8 +570,8 @@
break;
}
}
-
- if(i == 0) {
+
+ if (i == 0) {
state->pos += j;
if ((state->buffer + state->len) < (state->pos)) {
token->len = 0;
@@ -557,47 +587,48 @@
}
}
- if((token->len > 0)) {
+ if ((token->len > 0)) {
token->type = TokArbitraryBlockProgramData;
} else {
token->type = TokUnknown;
state->pos = token->ptr;
token->len = 0;
}
-
+
return token->len + (token->ptr - ptr);
}
/* 7.7.7 <EXPRESSION PROGRAM DATA> */
static int isProgramExpression(int c) {
- if ((c >= 0x20) && (c<=0x7e)) {
- if ((c!= 0x22)
- && (c!= 0x23)
- && (c!= 0x27)
- && (c!= 0x28)
- && (c!= 0x29)
- && (c!= 0x3B)) {
+ if ((c >= 0x20) && (c <= 0x7e)) {
+ if ((c != 0x22)
+ && (c != 0x23)
+ && (c != 0x27)
+ && (c != 0x28)
+ && (c != 0x29)
+ && (c != 0x3B)) {
return 1;
}
}
-
+
return 0;
}
static void skipProgramExpression(lex_state_t * state) {
- while(!iseos(state) && isProgramExpression(state->pos[0])) {
+ while (!iseos(state) && isProgramExpression(state->pos[0])) {
state->pos++;
}
}
// TODO: 7.7.7.2-2 recursive - any program data
-int SCPI_LexProgramExpression(lex_state_t * state, token_t * token) {
+
+int SCPI_LexProgramExpression(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
if (!iseos(state) && ischr(state, '(')) {
state->pos++;
skipProgramExpression(state);
-
+
if (!iseos(state) && ischr(state, ')')) {
state->pos++;
token->len = state->pos - token->ptr;
@@ -605,21 +636,21 @@
token->len = 0;
}
}
-
- if((token->len > 0)) {
+
+ if ((token->len > 0)) {
token->type = TokProgramExpression;
} else {
token->type = TokUnknown;
state->pos = token->ptr;
token->len = 0;
}
-
+
return token->len;
}
int SCPI_LexComma(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
if (skipChr(state, ',')) {
token->len = 1;
token->type = TokComma;
@@ -627,13 +658,13 @@
token->len = 0;
token->type = TokUnknown;
}
-
+
return token->len;
}
int SCPI_LexSemicolon(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
if (skipChr(state, ';')) {
token->len = 1;
token->type = TokSemicolon;
@@ -641,26 +672,26 @@
token->len = 0;
token->type = TokUnknown;
}
-
+
return token->len;
}
-int SCPI_LexNewLine(lex_state_t * state, token_t * token) {
+int SCPI_LexNewLine(lex_state_t * state, token_t * token) {
token->ptr = state->pos;
-
+
skipChr(state, '\r');
skipChr(state, '\n');
-
+
token->len = state->pos - token->ptr;
-
- if((token->len > 0)) {
+
+ if ((token->len > 0)) {
token->type = TokNewLine;
} else {
token->type = TokUnknown;
state->pos = token->ptr;
token->len = 0;
}
-
+
return token->len;
}
--
Gitblit v1.9.1