nmrtf.c

Linux下的多协议即时通讯程序源代码

/*
 * nmrtf.c
 *
 * Copyright (c) 2004 Novell, Inc. All Rights Reserved.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; version 2 of the License.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA	02111-1307	USA
 *
 */

/* This code was adapted from the sample RTF reader found here:
 * http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnrtfspec/html/rtfspec.asp
 */

#include <glib.h>
#include <stdlib.h>
#include <stdio.h>
#include <stddef.h>
#include <ctype.h>
#include <string.h>
#include "nmrtf.h"
#include "debug.h"

/* Internal RTF parser error codes */
#define NMRTF_OK 0                      /* Everything's fine! */
#define NMRTF_STACK_UNDERFLOW    1       /* Unmatched '}' */
#define NMRTF_STACK_OVERFLOW     2       /* Too many '{' -- memory exhausted */
#define NMRTF_UNMATCHED_BRACE    3       /* RTF ended during an open group. */
#define NMRTF_INVALID_HEX        4       /* invalid hex character found in data */
#define NMRTF_BAD_TABLE          5       /* RTF table (sym or prop) invalid */
#define NMRTF_ASSERTION		     6       /* Assertion failure */
#define NMRTF_EOF		         7       /* End of file reached while reading RTF */
#define NMRTF_CONVERT_ERROR		 8		 /* Error converting text  */

#define NMRTF_MAX_DEPTH 256

typedef enum
{
	NMRTF_STATE_NORMAL,
	NMRTF_STATE_SKIP,
	NMRTF_STATE_FONTTABLE,
	NMRTF_STATE_BIN,
	NMRTF_STATE_HEX
} NMRtfState;  /* Rtf State */

/* Property types that we care about */
typedef enum
{
	NMRTF_PROP_FONT_IDX,
	NMRTF_PROP_FONT_CHARSET,
	NMRTF_PROP_MAX
} NMRtfProperty;

typedef enum
{
	NMRTF_SPECIAL_BIN,
	NMRTF_SPECIAL_HEX,
	NMRTF_SPECIAL_UNICODE,
	NMRTF_SPECIAL_SKIP
} NMRtfSpecialKwd;

typedef enum
{
	NMRTF_DEST_FONTTABLE,
	NMRTF_DEST_SKIP
} NMRtfDestinationType;

typedef enum
{
	NMRTF_KWD_CHAR,
	NMRTF_KWD_DEST,
	NMRTF_KWD_PROP,
	NMRTF_KWD_SPEC
} NMRtfKeywordType;

typedef struct _NMRTFCharProp
{
	/* All we care about for now is the font.
	 * bold, italic, underline, etc. should be
	 * added here
	 */
	int font_idx;
	int	font_charset;
} NMRtfCharProp;

typedef struct _NMRtfStateSave
{
    NMRtfCharProp chp;
    NMRtfState rds;
    NMRtfState ris;
} NMRtfStateSave;

typedef struct _NMRtfSymbol
{
    char *keyword;          	/* RTF keyword */
    int  default_val;          	/* default value to use */
    gboolean pass_default;  	/* true to use default value from this table */
    NMRtfKeywordType kwd_type;  /* the type of the keyword */
    int  action;               	/* property type if the keyword represents a property */
                            	/* destination type if the keyword represents a destination */
                            	/* character to print if the keyword represents a character */
} NMRtfSymbol;


typedef struct _NMRtfFont
{
	int number;
	char *name;
	int charset;
} NMRtfFont;

/* RTF Context */
struct _NMRtfContext
{
	NMRtfState rds; 		/* destination state */
	NMRtfState ris; 		/* internal state */
	NMRtfCharProp chp; 		/* current character properties (ie. font, bold, italic, etc.) */
	GSList *font_table;		/* the font table */
	GSList *saved;			/* saved state stack */
	int param;				/* numeric parameter for the current keyword */
	long bytes_to_skip; 	/* number of bytes to skip (after encountering \bin) */
	int depth;				/* how many groups deep are we */
	gboolean skip_unknown;	/* if true, skip any unknown destinations (this is set after encountering '\*') */
	char *input;			/* input string */
	char nextch;			/* next char in input */
	GString *ansi;   		/* Temporary ansi text, will be convert/flushed to the output string */
	GString *output; 		/* The plain text UTF8 string */
};

static int rtf_parse(NMRtfContext *ctx);
static int rtf_push_state(NMRtfContext *ctx);
static int rtf_pop_state(NMRtfContext *ctx);
static NMRtfFont *rtf_get_font(NMRtfContext *ctx, int index);
static int rtf_get_char(NMRtfContext *ctx, guchar *ch);
static int rtf_unget_char(NMRtfContext *ctx, guchar ch);
static int rtf_flush_data(NMRtfContext *ctx);
static int rtf_parse_keyword(NMRtfContext *ctx);
static int rtf_dispatch_control(NMRtfContext *ctx, char *keyword, int param, gboolean param_set);
static int rtf_dispatch_char(NMRtfContext *ctx, guchar ch);
static int rtf_dispatch_unicode_char(NMRtfContext *ctx, gunichar ch);
static int rtf_print_char(NMRtfContext *ctx, guchar ch);
static int rtf_print_unicode_char(NMRtfContext *ctx, gunichar ch);
static int rtf_change_destination(NMRtfContext *ctx, NMRtfDestinationType dest);
static int rtf_dispatch_special(NMRtfContext *ctx, NMRtfSpecialKwd special);
static int rtf_apply_property(NMRtfContext *ctx, NMRtfProperty prop, int val);

/* RTF parser tables */

/* Keyword descriptions */
NMRtfSymbol rtf_symbols[] = {
	/* keyword, default, pass_default, keyword_type, action */
    {"fonttbl",  0,   	FALSE,    	NMRTF_KWD_DEST, NMRTF_DEST_FONTTABLE},
	{"f",		 0,		FALSE,		NMRTF_KWD_PROP, NMRTF_PROP_FONT_IDX},
	{"fcharset",  0,  	FALSE,		NMRTF_KWD_PROP,	NMRTF_PROP_FONT_CHARSET},
    {"par",      0,     FALSE,		NMRTF_KWD_CHAR, 0x0a},
    {"line",      0,    FALSE,     	NMRTF_KWD_CHAR, 0x0a},
    {"\0x0a",    0,     FALSE,     	NMRTF_KWD_CHAR, 0x0a},
    {"\0x0d",    0,     FALSE,     	NMRTF_KWD_CHAR, 0x0a},
    {"tab",      0,     FALSE,     	NMRTF_KWD_CHAR, 0x09},
	{"\r",		0,		FALSE,		NMRTF_KWD_CHAR,	'\r'},
	{"\n",		0,		FALSE,		NMRTF_KWD_CHAR,	'\n'},
    {"ldblquote",0,     FALSE,     	NMRTF_KWD_CHAR, '"'},
    {"rdblquote",0,     FALSE,     	NMRTF_KWD_CHAR, '"'},
    {"{",        0,     FALSE,     	NMRTF_KWD_CHAR, '{'},
    {"}",        0,     FALSE,     	NMRTF_KWD_CHAR, '}'},
    {"\\",       0,     FALSE,     	NMRTF_KWD_CHAR,  '\\'},
    {"bin",      0,     FALSE,     	NMRTF_KWD_SPEC, NMRTF_SPECIAL_BIN},
    {"*",        0,     FALSE,     	NMRTF_KWD_SPEC, NMRTF_SPECIAL_SKIP},
    {"'",        0,     FALSE,     	NMRTF_KWD_SPEC, NMRTF_SPECIAL_HEX},
	{"u",		0,		FALSE,		NMRTF_KWD_SPEC,	NMRTF_SPECIAL_UNICODE},
    {"colortbl", 0,     FALSE,     NMRTF_KWD_DEST,	NMRTF_DEST_SKIP},
    {"author",   0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"buptim",   0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"comment",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"creatim",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"doccomm",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"footer",   0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"footerf",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"footerl",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"footerr",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"footnote", 0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"ftncn",    0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"ftnsep",   0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"ftnsepc",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"header",   0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"headerf",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"headerl",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"headerr",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"info",     0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"keywords", 0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"operator", 0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"pict",     0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"printim",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"private1", 0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"revtim",   0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"rxe",      0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"stylesheet",   0, FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"subject",  0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"tc",       0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"title",    0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"txe",      0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP},
    {"xe",       0,     FALSE,     NMRTF_KWD_DEST,  NMRTF_DEST_SKIP}
};
int table_size = sizeof(rtf_symbols) / sizeof(NMRtfSymbol);

NMRtfContext *
nm_rtf_init()
{
	NMRtfContext *ctx = g_new0(NMRtfContext, 1);
	ctx->nextch = -1;
	ctx->ansi = g_string_new("");
	ctx->output = g_string_new("");
	return ctx;
}

char *
nm_rtf_strip_formatting(NMRtfContext *ctx, const char *input)
{
	int status;

	ctx->input = (char *)input;
	status = rtf_parse(ctx);
	if (status == NMRTF_OK)
		return g_strdup(ctx->output->str);

	purple_debug_info("novell", "RTF parser failed with error code %d", status);
	return NULL;
}

void
nm_rtf_deinit(NMRtfContext *ctx)
{
	GSList *node;
	NMRtfFont *font;
	NMRtfStateSave *save;

	if (ctx) {
		for (node = ctx->font_table; node; node = node->next) {
			font = node->data;
			g_free(font->name);
			g_free(font);
			node->data = NULL;
		}
		g_slist_free(ctx->font_table);
		for (node = ctx->saved; node; node = node->next) {
			save = node->data;
			g_free(save);
			node->data = NULL;
		}
		g_slist_free(ctx->saved);
		g_string_free(ctx->ansi, TRUE);
		g_string_free(ctx->output, TRUE);
		g_free(ctx);
	}
}

static const char *
get_current_encoding(NMRtfContext *ctx)
{
	NMRtfFont *font;

	font = rtf_get_font(ctx, ctx->chp.font_idx);

	switch (font->charset) {
		case 0:
			return "CP1252";
		case 77:
			return "MACINTOSH";
		case 78:
			return "SJIS";
		case 128:
			return "CP932";
		case 129:
			return "CP949";
		case 130:
			return "CP1361";
		case 134:
			return "CP936";
		case 136:
			return "CP950";
		case 161:
			return "CP1253";
		case 162:
			return "CP1254";
		case 163:
			return "CP1258";
		case 181:
		case 177:
			return "CP1255";
		case 178:
		case 179:
		case 180:
			return "CP1256";
		case 186:
			return "CP1257";
		case 204:
			return "CP1251";
		case 222:
			return "CP874";
		case 238:
			return "CP1250";
		case 254:
			return "CP437";
		default:
			purple_debug_info("novell", "Unhandled font charset %d\n", font->charset);
			return "CP1252";
	}
	return "CP1252";
}


/*
 * Add an entry to the font table
 */
static int
rtf_add_font_entry(NMRtfContext *ctx, int number, const char *name, int charset)
{
    NMRtfFont *font = g_new0(NMRtfFont, 1);

	font->number = number;
    font->name = g_strdup(name);
	font->charset = charset;

	purple_debug_info("novell", "Adding font to table: #%d\t%s\t%d\n",
					font->number, font->name, font->charset);

	ctx->font_table = g_slist_append(ctx->font_table, font);

    return NMRTF_OK;
}

/*
 * Return the nth entry in the font table
 */
static NMRtfFont *
rtf_get_font(NMRtfContext *ctx, int nth)
{
	NMRtfFont *font;

	font = g_slist_nth_data(ctx->font_table, nth);

	return font;
}

/*
 * Step 1:
 * Isolate RTF keywords and send them to rtf_parse_keyword;
 * Push and pop state at the start and end of RTF groups;
 * Send text to rtf_dispatch_char for further processing.
 */
static int
rtf_parse(NMRtfContext *ctx)
{
    int status;
    guchar ch;
    guchar hex_byte = 0;
    int hex_count = 2;
	int len;

	if (ctx->input == NULL)
		return NMRTF_OK;

    while (rtf_get_char(ctx, &ch) == NMRTF_OK) {
        if (ctx->depth < 0)
            return NMRTF_STACK_UNDERFLOW;

		/* if we're parsing binary data, handle it directly */
        if (ctx->ris == NMRTF_STATE_BIN) {
            if ((status = rtf_dispatch_char(ctx, ch)) != NMRTF_OK)
                return status;
        } else {
            switch (ch) {
				case '{':
					if (ctx->depth > NMRTF_MAX_DEPTH)
						return NMRTF_STACK_OVERFLOW;
                    rtf_flush_data(ctx);
					if ((status = rtf_push_state(ctx)) != NMRTF_OK)
						return status;
					break;
				case '}':
					rtf_flush_data(ctx);

					/* for some reason there is always an unwanted '\par' at the end */
					if (ctx->rds == NMRTF_STATE_NORMAL) {
						len = ctx->output->len;
						if (ctx->output->str[len-1] == '\n')
							ctx->output = g_string_truncate(ctx->output, len-1);
					}

					if ((status = rtf_pop_state(ctx)) != NMRTF_OK)
						return status;

					if (ctx->depth < 0)
						return NMRTF_STACK_OVERFLOW;
					break;
				case '\\':
					if ((status = rtf_parse_keyword(ctx)) != NMRTF_OK)
						return status;
					break;
				case 0x0d:
				case 0x0a:          /*  cr and lf are noise characters... */
					break;
				default:
					if (ctx->ris == NMRTF_STATE_NORMAL) {
						if ((status = rtf_dispatch_char(ctx, ch)) != NMRTF_OK)