From 1d23efcd8480b1021ca00fbb14db41b113a2f2b1 Mon Sep 17 00:00:00 2001
From: rswindell <>
Date: Mon, 8 Jul 2019 04:23:48 +0000
Subject: [PATCH] Add function cp437_to_utf8_str().
---
src/encode/utf8.c | 28 ++++++++++++++++++++++++++++
src/encode/utf8.h | 15 +++++++++++++--
2 files changed, 41 insertions(+), 2 deletions(-)
diff --git a/src/encode/utf8.c b/src/encode/utf8.c
index a210e6d0c9..94daa78948 100644
--- a/src/encode/utf8.c
+++ b/src/encode/utf8.c
@@ -34,6 +34,7 @@
****************************************************************************/
#include "utf8.h"
+#include "unicode.h"
#include <stdbool.h>
#include <string.h>
@@ -218,6 +219,33 @@ bool utf8_str_is_valid(const char* str)
return true;
}
+int cp437_to_utf8_str(const char* str, char* dest, size_t maxlen, unsigned char minval)
+{
+ int retval = 0;
+ size_t outlen = 0;
+ for(const unsigned char* p = str; *p != 0; p++) {
+ if(outlen >= maxlen) {
+ retval = -1;
+ break;
+ }
+ uint32_t codepoint = 0;
+ if(*p >= minval)
+ codepoint = cp437_unicode_tbl[*p];
+ if(codepoint) {
+ retval = utf8_putc(dest + outlen, maxlen - outlen, codepoint);
+ if(retval < 1)
+ break;
+ outlen += retval;
+ } else {
+ *(dest + outlen) = *p;
+ outlen++;
+ }
+ }
+ *(dest + outlen) = 0;
+ return retval;
+}
+
+
// From openssl/crypto/asn1/a_utf8.c:
/*
* Copyright 1995-2016 The OpenSSL Project Authors. All Rights Reserved.
diff --git a/src/encode/utf8.h b/src/encode/utf8.h
index c959b4457e..2d3293cb17 100644
--- a/src/encode/utf8.h
+++ b/src/encode/utf8.h
@@ -48,13 +48,24 @@ extern "C" {
// Returns true if the string is valid UTF-8
bool utf8_str_is_valid(const char*);
+
// Normalizes (to ASCII) chars in UTF-8 string 'str', in-place, resulting in string <= original in length
char* utf8_normalize_str(char* str);
-// Replace or strip UTF-8 sequences in str
-// If table ('tbl') of unicode codepoints if non-NULL is an array of 256 codepoints to map to 8-bit chars
+
+// Replace or strip UTF-8 sequences in str (in-place)
+// 'lookup' is a Unicode codepoint look-up function (optional)
+// 'unsupported_ch' is the character used to replace unsupported Unicode codepoints (optional)
+// 'unsupported_zwch' is the character used to replace unsupported zero-width Unicode codepoints (optional)
+// 'error_ch' is the character used to replace invalid UTF-8 sequence bytes (optional)
char* utf8_replace_chars(char* str, char (*lookup)(uint32_t), char unsupported_ch, char unsupported_zwch, char error_ch);
+
+// Convert a CP437 char string (src) to UTF-8 string (dest) up to 'maxlen' chars long (sans NUL-terminator)
+// 'minval' can be used to limit the range of converted chars
+int cp437_to_utf8_str(const char* src, char* dest, size_t maxlen, unsigned char minval);
+
// Decode a UTF-8 sequence to a UNICODE code point
int utf8_getc(const char* str, size_t len, uint32_t* codepoint);
+
// Encode a UNICODE code point into a UTF-8 sequence (str)
int utf8_putc(char* str, size_t len, uint32_t codepoint);
--
GitLab