From 9d2effe206b7eb7e59f78bfcde0fe42bee72523a Mon Sep 17 00:00:00 2001 From: Sage Vaillancourt Date: Wed, 20 Apr 2022 17:01:02 -0400 Subject: [PATCH] Better stringing. --- src/examples/forbble.pbl | 10 +++- src/object.c | 14 +++-- src/tests.sh | 5 +- src/tokens.c | 109 +++++++++++++++++++++++++++------------ 4 files changed, 96 insertions(+), 42 deletions(-) diff --git a/src/examples/forbble.pbl b/src/examples/forbble.pbl index ce353bb..5e63a85 100644 --- a/src/examples/forbble.pbl +++ b/src/examples/forbble.pbl @@ -10,7 +10,7 @@ ))) (def pop (fn () - "Remove the top of the stack" + "Remove and print the top of the stack" (if (> (len stack) 0) ( (def top (at 0 stack)) (set stack (rest stack)) @@ -18,6 +18,13 @@ ) (prnl "pop: STACK EMPTY")) )) +(def drop (fn () + "Remove the top of the stack" + (if (> (len stack) 0) ( + (set stack (rest stack)) + ) (prnl "pop: STACK EMPTY")) +)) + (def twop (fn (op) "Apply the given operation to the top two stack elements" ( (if (< (len stack) 2) (prnl "stack too small!") ( @@ -92,6 +99,7 @@ (h-insert operations name op) ))) (add-op "cls" '(sys "clear")) +(add-op "drop" '(drop)) (add-op "over" '(over)) (add-op "rot" '(rot)) (add-op "dup" '(dup)) diff --git a/src/object.c b/src/object.c index 84d18ed..683713e 100644 --- a/src/object.c +++ b/src/object.c @@ -739,22 +739,20 @@ inline Object nullTerminated(const char* string) inline Object stringFromSlice(const char* string, int len) { #ifdef STANDALONE + //eprintf("ay: %s\n", string); Object o = withLen(len, TYPE_STRING); int c = 0; for (int i = 0; i < len; i++) { if (string[i] == '\\') { - switch (string[i + 1]) { - case '"': - o.string[c] = '"'; - break; - case '\\': - o.string[c] = '\\'; + i++; + switch (string[i]) { + case 'n': + o.string[c] = '\n'; break; default: - o.string[c] = string[i + 1]; + o.string[c] = string[i]; break; } - i++; } else { o.string[c] = string[i]; } diff --git a/src/tests.sh b/src/tests.sh index 08421e2..9a3ad37 100755 --- a/src/tests.sh +++ b/src/tests.sh @@ -49,7 +49,7 @@ pass() { fail() { echo -n "X" - FAIL_OUTPUT="$FAIL_OUTPUT $1 FAILED\n When running \"$2\"" + FAIL_OUTPUT="$FAIL_OUTPUT $1 FAILED\n When running $2" ((FAILS++)) ((TOTAL_FAILS++)) } @@ -115,6 +115,9 @@ check "String Inequality" '(= "Beans" "Bean" )' "F" check "Null Inequality" '(= "Beans" "" )' "F" title "Complex Strings" +check "Escaping Quotes" '"\"Hello\""' '"Hello"' +check "Escaping Backslash" '"\\Hello\\"' '\Hello\' +check "Escaping Many" '"\\\"\\\\\\Hello\\\"\\\\\\"' '\"\\\Hello\"\\\' check "Triple Quoting" '"""Hello"""' 'Hello' check "Triple Quoting With Quotes" '"""My name is "yoink"."""' 'My name is "yoink".' diff --git a/src/tokens.c b/src/tokens.c index 7a3086c..9e514cc 100644 --- a/src/tokens.c +++ b/src/tokens.c @@ -43,6 +43,11 @@ void buildErrFromInput(struct Error* err, enum errorCode code, int i, const char strncpy(err->context, &input[start], ERR_LEN); } +void collectSymbol(const char* input, int* i, int* length); + +void processString(const char* input, struct Error* err, struct Slice* slices, int slice, int* i, int* lineNumber, + int* length); + struct Slice* nf_tokenize(const char* input, struct Error* err) { if (!input) { @@ -94,39 +99,13 @@ struct Slice* nf_tokenize(const char* input, struct Error* err) } continue; } else if (input[i] == '"') { - if (input[i + 1] == '"' && input[i + 2] == '"') { - // Triple-quoted block - i += 2; - slices[slice].text += 2; - for (;;) { - i++; - if (input[i] == '"' && input[i + 1] == '"' && input[i + 2] == '"') { - break; - } - if (input[i] == '\n') { - lineNumber++; - } - length++; - if (input[i] == '\0' || input[i + 1] == '\0' || input[i + 2] == '\0') { - buildErrFromInput(err, UNEXPECTED_EOF, i, input); - //free(slices); - return NULL; - } - } - } else { - // Simple string - while ((input[++i] != '"' || input[i - 1] == '\\') && input[i] != '\0') { - if (input[i] == '\n') { - lineNumber++; - } - length++; - } - } - i++; + processString(input, err, slices, slice, &i, &lineNumber, &length); } else { - while (!isWhitespace(input[++i]) && !isSingle(input[i]) && input[i] != '"' && input[i] != '\0') { - length++; - } + collectSymbol(input, &i, &length); + } + + if (err->context != NULL) { + return NULL; } slices[slice].length = length; @@ -147,3 +126,69 @@ struct Slice* nf_tokenize(const char* input, struct Error* err) return allocated; } + +void singleQuotedString(const char* input, int* i, int* lineNumber, int* length); + +void tripleQuotedString(const char* input, struct Error* err, struct Slice* slice, int* i, int* lineNumber, + int* length); + +void processString(const char* input, struct Error* err, struct Slice* slices, int slice, int* i, int* lineNumber, + int* length) +{ + if (input[(*i) + 1] == '"' && input[(*i) + 2] == '"') { + tripleQuotedString(input, err, &slices[slice], i, lineNumber, length); + } else { + singleQuotedString(input, i, lineNumber, length); + } + (*i)++; +} + +void collectSymbol(const char* input, int* i, int* length) +{ + while (!isWhitespace(input[++(*i)]) && !isSingle(input[(*i)]) && input[(*i)] != '"' && input[(*i)] != '\0') { + (*length)++; + } +} + +void tripleQuotedString(const char* input, struct Error* err, struct Slice* slice, int* i, int* lineNumber, int* length) +{ + // Skip past the extra opening quotes + (*i) += 2; + slice->text += 2; + + for (;;) { + (*i)++; + const int c = *i; + if (input[c] == '"' && input[c + 1] == '"' && input[c + 2] == '"') { + break; + } + if (input[c] == '\n') { + (*lineNumber)++; + } + (*length)++; + if (input[c] == '\0' || input[c + 1] == '\0' || input[c + 2] == '\0') { + buildErrFromInput(err, UNEXPECTED_EOF, c, input); + return; + } + } +} + +void singleQuotedString(const char* input, int* i, int* lineNumber, int* length) +{ + while (input[++(*i)] != '\0') { + if (input[(*i)] == '"') { + int backslashes = 0; + while (input[((*i) - 1) - backslashes] == '\\') { + backslashes++; + } + // \\\" => Odd number of backslashes, quote IS escaped + // \\" => Even number of backslashes, quote is NOT escaped + if (backslashes % 2 == 0) { + break; + } + } else if (input[(*i)] == '\n') { + (*lineNumber)++; + } + (*length)++; + } +}