fixed utf-8 characters returning the wrong width with strlen. added a bunch of stuff

Makefile.am

@@ -3,8 +3,10 @@ bin_PROGRAMS = mikusays
 mikusays_SOURCES = \
 	src/main.c \
 	src/colors.c \
+	src/helper.c \
 	defs/all.h \
-	defs/mikuart.h
+	defs/mikuart.h \
+	defs/helper.h
 
 AM_CFLAGS = -Wall -Wextra
 

Makefile.in

@@ -104,7 +104,8 @@ CONFIG_CLEAN_VPATH_FILES =
 am__installdirs = "$(DESTDIR)$(bindir)"
 PROGRAMS = $(bin_PROGRAMS)
 am__dirstamp = $(am__leading_dot)dirstamp
-am_mikusays_OBJECTS = src/main.$(OBJEXT) src/colors.$(OBJEXT)
+am_mikusays_OBJECTS = src/main.$(OBJEXT) src/colors.$(OBJEXT) \
+	src/helper.$(OBJEXT)
 mikusays_OBJECTS = $(am_mikusays_OBJECTS)
 mikusays_LDADD = $(LDADD)
 AM_V_P = $(am__v_P_@AM_V@)
@@ -122,7 +123,8 @@ am__v_at_1 =
 DEFAULT_INCLUDES = -I.@am__isrc@
 depcomp = $(SHELL) $(top_srcdir)/build-aux/depcomp
 am__maybe_remake_depfiles = depfiles
-am__depfiles_remade = src/$(DEPDIR)/colors.Po src/$(DEPDIR)/main.Po
+am__depfiles_remade = src/$(DEPDIR)/colors.Po src/$(DEPDIR)/helper.Po \
+	src/$(DEPDIR)/main.Po
 am__mv = mv -f
 COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \
 	$(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)
@@ -291,8 +293,10 @@ top_srcdir = @top_srcdir@
 mikusays_SOURCES = \
 	src/main.c \
 	src/colors.c \
+	src/helper.c \
 	defs/all.h \
-	defs/mikuart.h
+	defs/mikuart.h \
+	defs/helper.h
 
 AM_CFLAGS = -Wall -Wextra
 EXTRA_DIST = LICENSE README.md
@@ -384,6 +388,8 @@ src/$(DEPDIR)/$(am__dirstamp):
 src/main.$(OBJEXT): src/$(am__dirstamp) src/$(DEPDIR)/$(am__dirstamp)
 src/colors.$(OBJEXT): src/$(am__dirstamp) \
 	src/$(DEPDIR)/$(am__dirstamp)
+src/helper.$(OBJEXT): src/$(am__dirstamp) \
+	src/$(DEPDIR)/$(am__dirstamp)
 
 mikusays$(EXEEXT): $(mikusays_OBJECTS) $(mikusays_DEPENDENCIES) $(EXTRA_mikusays_DEPENDENCIES) 
 	@rm -f mikusays$(EXEEXT)
@@ -397,6 +403,7 @@ distclean-compile:
 	-rm -f *.tab.c
 
 @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/colors.Po@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/helper.Po@am__quote@ # am--include-marker
 @AMDEP_TRUE@@am__include@ @am__quote@src/$(DEPDIR)/main.Po@am__quote@ # am--include-marker
 
 $(am__depfiles_remade):
@@ -749,6 +756,7 @@ clean-am: clean-binPROGRAMS clean-generic mostlyclean-am
 distclean: distclean-am
 	-rm -f $(am__CONFIG_DISTCLEAN_FILES)
 	-rm -f src/$(DEPDIR)/colors.Po
+	-rm -f src/$(DEPDIR)/helper.Po
 	-rm -f src/$(DEPDIR)/main.Po
 	-rm -f Makefile
 distclean-am: clean-am distclean-compile distclean-generic \
@@ -798,6 +806,7 @@ maintainer-clean: maintainer-clean-am
 	-rm -f $(am__CONFIG_DISTCLEAN_FILES)
 	-rm -rf $(top_srcdir)/autom4te.cache
 	-rm -f src/$(DEPDIR)/colors.Po
+	-rm -f src/$(DEPDIR)/helper.Po
 	-rm -f src/$(DEPDIR)/main.Po
 	-rm -f Makefile
 maintainer-clean-am: distclean-am maintainer-clean-generic

configure.ac

@@ -1,5 +1,5 @@
 AC_PREREQ([2.69])
-AC_INIT([mikusays], [0.1.1], [], [mikusays], [https://github.com/davideldios7/mikusays])
+AC_INIT([mikusays], [0.2.0], [], [mikusays], [https://github.com/davideldios7/mikusays])
 AC_CONFIG_SRCDIR([src/main.c])
 AC_CONFIG_AUX_DIR([build-aux])
 AM_INIT_AUTOMAKE([foreign subdir-objects -Wall])

defs/all.h

@@ -7,6 +7,7 @@
 #include <string.h>
 #include <locale.h>
 
+#include "helper.h"
 
 //COLORS.C 
 void Void(void); //function that does FUCKING NOTHING BECAUSE MY STUPID LSP THINK I'M NOT USING THIS FUCKING FILE AND GIVES ME A WARNINGGENROENIGGHUIOAGHUIHUIAUHIDFG

defs/helper.h

@@ -0,0 +1,6 @@
+#ifndef helper_h
+#define helper_h
+
+int displaywidth(const char *text);
+
+#endif 

src/helper.c

@@ -0,0 +1,78 @@
+#include "../defs/all.h"
+#include "../defs/widechar.h"
+
+/* this reads one utf81 character from the start of bytes
+   it puts the character's number in *codepoint and returns how many bytes it used (its always 1 or more)
+   invalid input it stores U+FFFD (�) and returns 1, so the caller always moves forward. */
+static int utf8_decode(const unsigned char *bytes, unsigned long *codepoint){
+    unsigned long value = bytes[0];
+    unsigned long minallowed = 0;   //rejects overlong encodings 
+    int sequence_length;
+    int byte_index;
+
+    if(value < 0x80){
+        *codepoint = value;
+        return 1;
+    } else if(value >= 0xC2 && value <= 0xDF){
+        sequence_length = 2;
+        value &= 0x1F;
+        minallowed = 0x80;
+    } else if(value >= 0xE0 && value <= 0xEF){
+        sequence_length = 3;
+        value &= 0x0F;
+        minallowed = 0x800;
+    } else if(value >= 0xF0 && value <= 0xF4){
+        sequence_length = 4;
+        value &= 0x07;
+        minallowed = 0x10000;
+    } else {
+        *codepoint = 0xFFFD;
+        return 1;
+    }
+
+    for(byte_index = 1; byte_index < sequence_length; byte_index++){
+        if((bytes[byte_index] & 0xC0) != 0x80){
+            *codepoint = 0xFFFD;
+            return 1;
+        }
+        value = (value << 6) | (bytes[byte_index] & 0x3F);
+    }
+
+    if(value < minallowed || value > 0x10FFFF || (value >= 0xD800 && value <= 0xDFFF)){
+        *codepoint = 0xFFFD;
+        return 1;
+    }
+
+    *codepoint = value;
+    return sequence_length;
+}
+
+/* this turns the library's return value into a column count so 0 1 or 2 */
+static int codepoint_width(unsigned long codepoint){
+    int width = widechar_wcwidth(codepoint);
+    if(width > 0)
+        return width;
+
+    switch(width){
+        case widechar_ambiguous:     
+        case widechar_private_use:
+            return 1;
+        case widechar_widened_in_9:
+            return 2;
+        default:                     
+            return 0;
+    }
+}
+
+/* how many terminal columns a string takes up */
+int displaywidth(const char *text){
+    const unsigned char *current_byte = (const unsigned char *)text;
+    int total_width = 0;
+
+    while(*current_byte){
+        unsigned long codepoint;
+        current_byte += utf8_decode(current_byte, &codepoint);
+        total_width += codepoint_width(codepoint);
+    }
+    return total_width;
+}

src/main.c

@@ -63,7 +63,7 @@ char **wraptext(const char *text, int max, int *count){
 
     char *tok = strtok(copy, " \t\n");   
     while(tok != NULL){ 
-        if(strlen(current_line)+strlen(tok) < (size_t)max){
+        if((size_t)displaywidth(current_line)+(size_t)displaywidth(tok) < (size_t)max){
             if(current_line[0] != '\0'){
                 strcat(current_line, " ");
             }
@@ -101,7 +101,7 @@ char **getbubblelines(char *text){
 
     size_t content_width = 0;
     for (int i = 0; templines[i] != NULL; i++) {
-        size_t current_len = strlen(templines[i]);
+        size_t current_len = displaywidth(templines[i]);
         if (current_len > content_width)
             content_width = current_len;
     }
@@ -128,13 +128,13 @@ char **getbubblelines(char *text){
     strcat(bubblelines[0], top_right);
     
     for (int i = 0; lines[i] != NULL; i++) {
-        int line_width = strlen(lines[i]);
+        int line_width = displaywidth(lines[i]);
         int padding_needed = (bubble_width - 2) - line_width;
         if (padding_needed < 0) padding_needed = 0;
         int left_padding = padding_needed / 2;
         int right_padding = padding_needed - left_padding;
-
-        bubblelines[i + 1] = malloc(strlen(vertical) * 2 + 2 + line_width + padding_needed + 1);
+        
+        bubblelines[i + 1] = malloc(strlen(vertical) * 2 + 2 + strlen(lines[i]) + padding_needed + 1);
         sprintf(bubblelines[i + 1], "%s %*s%s%*s %s",vertical, left_padding, "", lines[i], right_padding, "", vertical);
     }
 
@@ -177,18 +177,6 @@ static void list_all_art_styles(void){
     printf("\nUse --style <number> to select a specific style, or omit for random selection.\n");
 }
 
-
-//display width in columns: counts utf 8 codepoints (skips continuation bytes)
-//assumes every character is 1 column wide
-//emojis and other bullshit won't work bu tthat's okay since who the fuck will try to put '🥺' into this
-static int displaywidth(const char *s){
-    int w = 0;
-    for(; *s; s++)
-        if(((unsigned char)*s & 0xC0) != 0x80)
-            w++;
-    return w;
-}
-
 void draw_miku_says(char *text, int style){
     char **speech_bubble_lines = getbubblelines(text);
     if(speech_bubble_lines == NULL)