lbbs/src/str_process.c

/* SPDX-License-Identifier: GPL-3.0-or-later */
/*
 * str_process
 *   - common string processing features with UTF-8 support
 *
 * Copyright (C) 2004-2025  Leaflet <leaflet@leafok.com>
 */

#include "common.h"
#include "log.h"
#include "str_process.h"
#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <wchar.h>

int UTF8_fixed_width = 1;

int str_length(const char *str, int skip_ctrl_seq)
{
        int str_len;
        char input_str[5];
        wchar_t wcs[2];
        int wc_len;
        int i;
        char c;
        int ret = 0;

        for (i = 0; str[i] != '\0'; i++)
        {
                c = str[i];

                if (c == '\r' || c == '\7') // skip
                {
                        continue;
                }

                if (skip_ctrl_seq && c == '\033' && str[i + 1] == '[') // Skip control sequence
                {
                        for (i = i + 2; isdigit(str[i]) || str[i] == ';' || str[i] == '?'; i++)
                                ;

                        if (str[i] == 'm') // valid
                        {
                                // skip
                        }
                        else if (isalpha(str[i]))
                        {
                                // unsupported ANSI CSI command
                        }
                        else
                        {
                                i--;
                        }

                        continue;
                }

                // Process UTF-8 Chinese characters
                if (c & 0x80) // head of multi-byte character
                {
                        str_len = 0;
                        c = (char)(c & 0xf0);
                        while (c & 0x80)
                        {
                                input_str[str_len] = str[i + str_len];
                                str_len++;
                                c = (c & 0x7f) << 1;
                        }
                        input_str[str_len] = '\0';

                        if (mbstowcs(wcs, input_str, 1) == (size_t)-1)
                        {
                                log_error("mbstowcs(%s) error\n", input_str);
                        }
                        wc_len = (UTF8_fixed_width ? 2 : wcwidth(wcs[0]));

                        i += (str_len - 1);
                        ret += wc_len;
                }
                else
                {
                        ret++;
                }
        }

        return ret;
}

int split_line(const char *buffer, int max_display_len, int *p_eol, int *p_display_len, int skip_ctrl_seq)
{
        int i;
        *p_eol = 0;
        *p_display_len = 0;
        char c;
        int str_len;
        char input_str[5];
        wchar_t wcs[2];
        int wc_len;

        for (i = 0; buffer[i] != '\0'; i++)
        {
                c = buffer[i];

                if (c == '\r' || c == '\7') // skip
                {
                        continue;
                }

                if (skip_ctrl_seq && c == '\033' && buffer[i + 1] == '[') // Skip control sequence
                {
                        i += 2;
                        while (buffer[i] != '\0' && buffer[i] != 'm')
                        {
                                i++;
                        }
                        continue;
                }

                if (c & 0x80) // head of multi-byte character
                {
                        str_len = 0;
                        c = (char)(c & 0xf0);
                        while (c & 0x80)
                        {
                                input_str[str_len] = buffer[i + str_len];
                                str_len++;
                                c = (c & 0x7f) << 1;
                        }
                        input_str[str_len] = '\0';

                        if (mbstowcs(wcs, input_str, 1) == (size_t)-1)
                        {
                                log_error("mbstowcs(%s) error\n", input_str);
                        }
                        wc_len = (UTF8_fixed_width ? 2 : wcwidth(wcs[0]));
                        if (*p_display_len + wc_len > max_display_len)
                        {
                                break;
                        }

                        i += (str_len - 1);
                        (*p_display_len) += wc_len;
                }
                else
                {
                        if (*p_display_len + 1 > max_display_len)
                        {
                                break;
                        }
                        (*p_display_len)++;

                        // \n is regarded as 1 character wide in terminal editor, which is different from Web version
                        if (c == '\n')
                        {
                                i++;
                                *p_eol = 1;
                                break;
                        }
                }
        }

        return i;
}

long split_data_lines(const char *p_buf, int max_display_len, long *p_line_offsets, long line_offsets_count,
                                          int skip_ctrl_seq, int *p_line_widths)
{
        int line_cnt = 0;
        int len;
        int end_of_line = 0;
        int display_len = 0;

        p_line_offsets[line_cnt] = 0L;

        do
        {
                len = split_line(p_buf, max_display_len, &end_of_line, &display_len, skip_ctrl_seq);

                if (p_line_widths)
                {
                        p_line_widths[line_cnt] = display_len;
                }

                // Exceed max_line_cnt
                if (line_cnt + 1 >= line_offsets_count)
                {
                        // log_error("Line count %d reaches limit %d\n", line_cnt + 1, line_offsets_count);
                        return line_cnt;
                }

                p_line_offsets[line_cnt + 1] = p_line_offsets[line_cnt] + len;
                line_cnt++;
                p_buf += len;
        } while (p_buf[0] != '\0' || end_of_line);

        return line_cnt;
}

int str_filter(char *buffer, int skip_ctrl_seq)
{
        int i;
        int j;

        for (i = 0, j = 0; buffer[i] != '\0'; i++)
        {
                if (buffer[i] == '\r' || buffer[i] == '\7') // skip
                {
                        continue;
                }

                if (skip_ctrl_seq && buffer[i] == '\033' && buffer[i + 1] == '[') // Skip control sequence
                {
                        i += 2;
                        while (buffer[i] != '\0' && buffer[i] != 'm')
                        {
                                i++;
                        }
                        continue;
                }

                buffer[j] = buffer[i];
                j++;
        }

        buffer[j] = '\0';

        return j;
}
1	/* SPDX-License-Identifier: GPL-3.0-or-later */
2	/*
3	* str_process
4	* - common string processing features with UTF-8 support
5	*
6	* Copyright (C) 2004-2025 Leaflet <leaflet@leafok.com>
7	*/
8
9	#include "common.h"
10	#include "log.h"
11	#include "str_process.h"
12	#include <ctype.h>
13	#include <stdio.h>
14	#include <stdlib.h>
15	#include <string.h>
16	#include <wchar.h>
17
18	int UTF8_fixed_width = 1;
19
20	int str_length(const char *str, int skip_ctrl_seq)
21	{
22	int str_len;
23	char input_str[5];
24	wchar_t wcs[2];
25	int wc_len;
26	int i;
27	char c;
28	int ret = 0;
29
30	for (i = 0; str[i] != '\0'; i++)
31	{
32	c = str[i];
33
34	if (c == '\r' \|\| c == '\7') // skip
35	{
36	continue;
37	}
38
39	if (skip_ctrl_seq && c == '\033' && str[i + 1] == '[') // Skip control sequence
40	{
41	for (i = i + 2; isdigit(str[i]) \|\| str[i] == ';' \|\| str[i] == '?'; i++)
42	;
43
44	if (str[i] == 'm') // valid
45	{
46	// skip
47	}
48	else if (isalpha(str[i]))
49	{
50	// unsupported ANSI CSI command
51	}
52	else
53	{
54	i--;
55	}
56
57	continue;
58	}
59
60	// Process UTF-8 Chinese characters
61	if (c & 0x80) // head of multi-byte character
62	{
63	str_len = 0;
64	c = (char)(c & 0xf0);
65	while (c & 0x80)
66	{
67	input_str[str_len] = str[i + str_len];
68	str_len++;
69	c = (c & 0x7f) << 1;
70	}
71	input_str[str_len] = '\0';
72
73	if (mbstowcs(wcs, input_str, 1) == (size_t)-1)
74	{
75	log_error("mbstowcs(%s) error\n", input_str);
76	}
77	wc_len = (UTF8_fixed_width ? 2 : wcwidth(wcs[0]));
78
79	i += (str_len - 1);
80	ret += wc_len;
81	}
82	else
83	{
84	ret++;
85	}
86	}
87
88	return ret;
89	}
90
91	int split_line(const char buffer, int max_display_len, int p_eol, int *p_display_len, int skip_ctrl_seq)
92	{
93	int i;
94	*p_eol = 0;
95	*p_display_len = 0;
96	char c;
97	int str_len;
98	char input_str[5];
99	wchar_t wcs[2];
100	int wc_len;
101
102	for (i = 0; buffer[i] != '\0'; i++)
103	{
104	c = buffer[i];
105
106	if (c == '\r' \|\| c == '\7') // skip
107	{
108	continue;
109	}
110
111	if (skip_ctrl_seq && c == '\033' && buffer[i + 1] == '[') // Skip control sequence
112	{
113	i += 2;
114	while (buffer[i] != '\0' && buffer[i] != 'm')
115	{
116	i++;
117	}
118	continue;
119	}
120
121	if (c & 0x80) // head of multi-byte character
122	{
123	str_len = 0;
124	c = (char)(c & 0xf0);
125	while (c & 0x80)
126	{
127	input_str[str_len] = buffer[i + str_len];
128	str_len++;
129	c = (c & 0x7f) << 1;
130	}
131	input_str[str_len] = '\0';
132
133	if (mbstowcs(wcs, input_str, 1) == (size_t)-1)
134	{
135	log_error("mbstowcs(%s) error\n", input_str);
136	}
137	wc_len = (UTF8_fixed_width ? 2 : wcwidth(wcs[0]));
138	if (*p_display_len + wc_len > max_display_len)
139	{
140	break;
141	}
142
143	i += (str_len - 1);
144	(*p_display_len) += wc_len;
145	}
146	else
147	{
148	if (*p_display_len + 1 > max_display_len)
149	{
150	break;
151	}
152	(*p_display_len)++;
153
154	// \n is regarded as 1 character wide in terminal editor, which is different from Web version
155	if (c == '\n')
156	{
157	i++;
158	*p_eol = 1;
159	break;
160	}
161	}
162	}
163
164	return i;
165	}
166
167	long split_data_lines(const char p_buf, int max_display_len, long p_line_offsets, long line_offsets_count,
168	int skip_ctrl_seq, int *p_line_widths)
169	{
170	int line_cnt = 0;
171	int len;
172	int end_of_line = 0;
173	int display_len = 0;
174
175	p_line_offsets[line_cnt] = 0L;
176
177	do
178	{
179	len = split_line(p_buf, max_display_len, &end_of_line, &display_len, skip_ctrl_seq);
180
181	if (p_line_widths)
182	{
183	p_line_widths[line_cnt] = display_len;
184	}
185
186	// Exceed max_line_cnt
187	if (line_cnt + 1 >= line_offsets_count)
188	{
189	// log_error("Line count %d reaches limit %d\n", line_cnt + 1, line_offsets_count);
190	return line_cnt;
191	}
192
193	p_line_offsets[line_cnt + 1] = p_line_offsets[line_cnt] + len;
194	line_cnt++;
195	p_buf += len;
196	} while (p_buf[0] != '\0' \|\| end_of_line);
197
198	return line_cnt;
199	}
200
201	int str_filter(char *buffer, int skip_ctrl_seq)
202	{
203	int i;
204	int j;
205
206	for (i = 0, j = 0; buffer[i] != '\0'; i++)
207	{
208	if (buffer[i] == '\r' \|\| buffer[i] == '\7') // skip
209	{
210	continue;
211	}
212
213	if (skip_ctrl_seq && buffer[i] == '\033' && buffer[i + 1] == '[') // Skip control sequence
214	{
215	i += 2;
216	while (buffer[i] != '\0' && buffer[i] != 'm')
217	{
218	i++;
219	}
220	continue;
221	}
222
223	buffer[j] = buffer[i];
224	j++;
225	}
226
227	buffer[j] = '\0';
228
229	return j;
230	}
webmaster@leafok.com	ViewVC Help
Powered by ViewVC 1.3.0-beta1