/*
- * ucm2cp.c - Convert IBM ".ucm" files into code page handling classes.
+ * ucm2cp.c - Convert IBM ".ucm" files or hexadecimal mapping ".TXT" files
+ * into code page handling classes.
*
* Copyright (c) 2002 Southern Storm Software, Pty Ltd
+ * Copyright (c) 2006 Bruno Haible
+ * Copyright (c) 2013 Mikko Korkalo
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the "Software"),
windowsCodePage = codePage;
}
- /* Open the UCM file */
+ /* Open the UCM or TXT file */
file = fopen(argv[1], "r");
if(!file)
{
}
/*
- * Load the character mapping information from a UCM file.
+ * Load the character mapping information from a UCM or TXT file.
*/
static void loadCharMaps(FILE *file)
{
+ enum { unknown, ucm, txt } syntax;
unsigned long posn;
unsigned long byteValue;
int level;
charToByte[posn] = -1;
}
+ syntax = unknown;
+
/* Read the contents of the file */
while(fgets(buffer, BUFSIZ, file))
{
- /* Lines of interest begin with "<U" */
- if(buffer[0] != '<' || buffer[1] != 'U')
+ /* Syntax recognition */
+ if (syntax == unknown)
{
- continue;
+ if (memcmp(buffer, "CHARMAP", 7) == 0)
+ syntax = ucm;
+ else if (memcmp(buffer, "0x", 2) == 0)
+ syntax = txt;
}
- /* Parse the fields on the line */
- buf = buffer + 2;
- buf += parseHex(buf, &posn);
- if(posn >= 65536)
- {
- continue;
- }
- while(*buf != '\0' && *buf != '\\')
- {
- ++buf;
- }
- if(*buf != '\\' || buf[1] != 'x')
+ if (syntax == ucm)
{
- continue;
- }
- buf += 2;
- buf += parseHex(buf, &byteValue);
- if(byteValue >= 256)
- {
- continue;
+ /* Lines of interest begin with "<U" */
+ if(buffer[0] != '<' || buffer[1] != 'U')
+ {
+ continue;
+ }
+
+ /* Parse the fields on the line */
+ buf = buffer + 2;
+ buf += parseHex(buf, &posn);
+ if(posn >= 65536)
+ {
+ continue;
+ }
+ while(*buf != '\0' && *buf != '\\')
+ {
+ ++buf;
+ }
+ if(*buf != '\\' || buf[1] != 'x')
+ {
+ continue;
+ }
+ buf += 2;
+ buf += parseHex(buf, &byteValue);
+ if(byteValue >= 256)
+ {
+ continue;
+ }
+ while(*buf != '\0' && *buf != '|')
+ {
+ ++buf;
+ }
+ if(*buf != '|')
+ {
+ continue;
+ }
+ level = (int)(buf[1] - '0');
}
- while(*buf != '\0' && *buf != '|')
+ else
+ if (syntax == txt)
{
- ++buf;
+ unsigned int x;
+ int cnt;
+
+ /* Lines of interest begin with "0x" */
+ if(buffer[0] != '0' || buffer[1] != 'x')
+ continue;
+
+ /* Parse the fields on the line */
+ if(sscanf(buffer, "0x%x%n", &x, &cnt) <= 0)
+ exit(1);
+ if(!(x < 0x100))
+ exit(1);
+ byteValue = x;
+ while (buffer[cnt] == ' ' || buffer[cnt] == '\t')
+ cnt++;
+ if(sscanf(buffer+cnt, "0x%x", &x) != 1)
+ continue;
+ if(!(x < 0x10000))
+ exit(1);
+ posn = x;
+ level = 0;
}
- if(*buf != '|')
- {
+ else
continue;
- }
- level = (int)(buf[1] - '0');
/* Update the byte->char mapping table */
if(level < byteToCharLevel[byteValue])
printf("/*\n * CP%d.cs - %s code page.\n", codePage, name);
fputs(COPYRIGHT_MSG, stdout);
printf("// Generated from \"%s\".\n\n", filename);
+ printf("// WARNING: Modifying this file directly might be a bad idea.\n");
+ printf("// You should edit the code generator tools/ucm2cp.c instead for your changes\n");
+ printf("// to appear in all relevant classes.\n");
printf("namespace I18N.%s\n{\n\n", region);
printf("using System;\n");
+ printf("using System.Text;\n");
printf("using I18N.Common;\n\n");
+ printf("[Serializable]\n");
printf("public class CP%d : ByteEncoding\n{\n", codePage);
printf("\tpublic CP%d()\n", codePage);
printf("\t\t: base(%d, ToChars, \"%s\",\n", codePage, name);
static void printFooter(void)
{
printf("}; // class CP%d\n\n", codePage);
+ printf("[Serializable]\n");
printf("public class ENC");
printEncodingName(webName);
printf(" : CP%d\n{\n", codePage);
* Print a "switch" statement that converts "ch" from
* a character value into a byte value.
*/
-static void printConvertSwitch(void)
+static void printConvertSwitch(int forString)
{
unsigned long directLimit;
unsigned long posn;
/* Print the switch footer */
if(!haveFullWidth)
{
- printf("\t\t\t\tdefault: ch = 0x3F; break;\n");
+ if(forString)
+ printf("\t\t\t\tdefault: ch = 0x3F; break;\n");
+ else {
+ printf("\t\t\t\tdefault:\n");
+ printf("#if NET_2_0\n");
+ printf("\t\t\t\t\tHandleFallback (ref buffer, chars, ref charIndex, ref charCount, bytes, ref byteIndex, ref byteCount);\n");
+ printf("\t\t\t\t\tfallback = true;\n");
+ printf("#else\n");
+ printf("\t\t\t\t\t\tch = 0x3F;\n");
+ printf("#endif\n");
+ printf("\t\t\t\t\tbreak;\n");
+ }
}
else
{
printf("\t\t\t\tdefault:\n");
printf("\t\t\t\t{\n");
printf("\t\t\t\t\tif(ch >= 0xFF01 && ch <= 0xFF5E)\n");
+ printf("\t\t\t\t\t{\n");
printf("\t\t\t\t\t\tch -= 0xFEE0;\n");
+ printf("\t\t\t\t\t}\n");
printf("\t\t\t\t\telse\n");
- printf("\t\t\t\t\t\tch = 0x3F;\n");
+ printf("\t\t\t\t\t{\n");
+ if(forString) /* this is basically meaningless, just to make diff for unused code minimum */
+ printf("\t\t\t\t\t\tch = 0x3F;\n");
+ else {
+ printf("#if NET_2_0\n");
+ printf("\t\t\t\t\t\t//Execute fallback routine and set fallback flag on\n");
+ printf("\t\t\t\t\t\tHandleFallback (ref buffer, chars, ref charIndex, ref charCount, bytes, ref byteIndex, ref byteCount);\n");
+ printf("\t\t\t\t\t\tfallback = true;\n");
+ printf("#else\n");
+ printf("\t\t\t\t\t\tch = 0x3F;\n");
+ printf("#endif\n");
+ }
+ printf("\t\t\t\t\t}\n");
printf("\t\t\t\t}\n");
printf("\t\t\t\tbreak;\n");
}
*/
static void printCharToByte(void)
{
+ printf("\t// Get the number of bytes needed to encode a character buffer.\n");
+ printf("\tpublic unsafe override int GetByteCountImpl (char* chars, int count)\n");
+ printf("\t{\n");
+ printf("\t\tif (this.EncoderFallback != null)");
+ printf("\t\t{\n");
+ printf("\t\t\t//Calculate byte count by actually doing encoding and discarding the data.\n");
+ printf("\t\t\treturn GetBytesImpl(chars, count, null, 0);\n");
+ printf("\t\t}\n");
+ printf("\t\telse\n");
+ printf("\t\t\n");
+ printf("\t\t{\n");
+ printf("\t\t\treturn count;\n");
+ printf("\t\t}\n");
+ printf("\t}\n");
+ printf("\t\n");
+ printf("\t// Get the number of bytes needed to encode a character buffer.\n");
+ printf("\tpublic override int GetByteCount (String s)\n");
+ printf("\t{\n");
+ printf("\t\tif (this.EncoderFallback != null)\n");
+ printf("\t\t{\n");
+ printf("\t\t\t//Calculate byte count by actually doing encoding and discarding the data.\n");
+ printf("\t\t\tunsafe\n");
+ printf("\t\t\t{\n");
+ printf("\t\t\t\tfixed (char *s_ptr = s)\n");
+ printf("\t\t\t\t{\n");
+ printf("\t\t\t\t\treturn GetBytesImpl(s_ptr, s.Length, null, 0);\n");
+ printf("\t\t\t\t}\n");
+ printf("\t\t\t}\n");
+ printf("\t\t}\n");
+ printf("\t\telse\n");
+ printf("\t\t{\n");
+ printf("\t\t\t//byte count equals character count because no EncoderFallback set\n");
+ printf("\t\t\treturn s.Length;\n");
+ printf("\t\t}\n");
+ printf("\t}\n");
+ printf("\t\n");
+ printf("\t//ToBytes is just an alias for GetBytesImpl, but doesn't return byte count\n");
+ printf("\tprotected unsafe override void ToBytes(char* chars, int charCount,\n");
+ printf("\t byte* bytes, int byteCount)\n");
+ printf("\t{\n");
+ printf("\t\t//Calling ToBytes with null destination buffer doesn't make any sense\n");
+ printf("\t\tif (bytes == null)\n");
+ printf("\t\t\tthrow new ArgumentNullException(\"bytes\");\n");
+ printf("\t\tGetBytesImpl(chars, charCount, bytes, byteCount);\n");
+ printf("\t}\n");
+ printf("\t\n");
+
/* Print the conversion method for character buffers */
- printf("\tprotected override void ToBytes(char[] chars, int charIndex, int charCount,\n");
- printf("\t byte[] bytes, int byteIndex)\n");
+ //printf("\tprotected unsafe override void ToBytes(char* chars, int charCount,\n");
+ //printf("\t byte* bytes, int byteCount)\n");
+ printf("\tpublic unsafe override int GetBytesImpl (char* chars, int charCount,\n");
+ printf("\t byte* bytes, int byteCount)\n");
printf("\t{\n");
printf("\t\tint ch;\n");
- printf("\t\twhile(charCount > 0)\n");
+ printf("\t\tint charIndex;\n");
+ printf("\t\tint byteIndex = 0;\n");
+ printf("#if NET_2_0\n");
+ printf("\t\tEncoderFallbackBuffer buffer = null;\n");
+ printf("#endif\n");
+ printf("\t\tfor (charIndex=0; charCount > 0; charIndex++, charCount--)\n");
printf("\t\t{\n");
- printf("\t\t\tch = (int)(chars[charIndex++]);\n");
- printConvertSwitch();
- printf("\t\t\tbytes[byteIndex++] = (byte)ch;\n");
- printf("\t\t\t--charCount;\n");
+ printf("\t\t\tch = (int)(chars[charIndex]);\n");
+ printf("\t\t\tbool fallback = false;\n");
+ printConvertSwitch(0);
+ printf("\t\t\t//Write encoded byte to buffer, if buffer is defined and fallback was not used\n");
+ printf("\t\t\tif (bytes != null && !fallback)\n");
+ printf("\t\t\t{\n");
+ printf("\t\t\t\tbytes[byteIndex] = (byte)ch;\n");
+ printf("\t\t\t}\n");
+ printf("\t\t\t\n");
+ printf("\t\t\t//Bump counters if fallback was not used\n");
+ printf("\t\t\tif (!fallback)\n");
+ printf("\t\t\t{\n");
+ printf("\t\t\t\tbyteIndex++;\n");
+ printf("\t\t\t\tbyteCount--;\n");
+ printf("\t\t\t}\n");
+ //printf("\t\t\tbytes[byteIndex++] = (byte)ch;\n");
+ //printf("\t\t\t--charCount;\n");
+ //printf("\t\t\t--byteCount;\n");
printf("\t\t}\n");
+ printf("\t\treturn byteIndex;\n");
printf("\t}\n\n");
/* Print the conversion method for string buffers */
+ printf("\t/*\n");
printf("\tprotected override void ToBytes(String s, int charIndex, int charCount,\n");
printf("\t byte[] bytes, int byteIndex)\n");
printf("\t{\n");
printf("\t\twhile(charCount > 0)\n");
printf("\t\t{\n");
printf("\t\t\tch = (int)(s[charIndex++]);\n");
- printConvertSwitch();
+ printConvertSwitch(1);
printf("\t\t\tbytes[byteIndex++] = (byte)ch;\n");
printf("\t\t\t--charCount;\n");
printf("\t\t}\n");
- printf("\t}\n\n");
+ printf("\t}\n");
+ printf("\t*/\n\n");
}
projects / mono.git / blobdiff