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)
                        {
#ifdef _DEBUG
                                log_error("mbstowcs(%s) error\n", input_str);
#endif
                                wc_len = (UTF8_fixed_width ? 2 : 1); // Fallback
                        }
                        else
                        {
                                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)
                        {
#ifdef _DEBUG
                                log_error("mbstowcs(%s) error\n", input_str);
#endif
                                wc_len = (UTF8_fixed_width ? 2 : 1); // Fallback
                        }
                        else
                        {
                                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	#ifdef _DEBUG
76	log_error("mbstowcs(%s) error\n", input_str);
77	#endif
78	wc_len = (UTF8_fixed_width ? 2 : 1); // Fallback
79	}
80	else
81	{
82	wc_len = (UTF8_fixed_width ? 2 : wcwidth(wcs[0]));
83	}
84
85	i += (str_len - 1);
86	ret += wc_len;
87	}
88	else
89	{
90	ret++;
91	}
92	}
93
94	return ret;
95	}
96
97	int split_line(const char buffer, int max_display_len, int p_eol, int *p_display_len, int skip_ctrl_seq)
98	{
99	int i;
100	*p_eol = 0;
101	*p_display_len = 0;
102	char c;
103	int str_len;
104	char input_str[5];
105	wchar_t wcs[2];
106	int wc_len;
107
108	for (i = 0; buffer[i] != '\0'; i++)
109	{
110	c = buffer[i];
111
112	if (c == '\r' \|\| c == '\7') // skip
113	{
114	continue;
115	}
116
117	if (skip_ctrl_seq && c == '\033' && buffer[i + 1] == '[') // Skip control sequence
118	{
119	i += 2;
120	while (buffer[i] != '\0' && buffer[i] != 'm')
121	{
122	i++;
123	}
124	continue;
125	}
126
127	if (c & 0x80) // head of multi-byte character
128	{
129	str_len = 0;
130	c = (char)(c & 0xf0);
131	while (c & 0x80)
132	{
133	input_str[str_len] = buffer[i + str_len];
134	str_len++;
135	c = (c & 0x7f) << 1;
136	}
137	input_str[str_len] = '\0';
138
139	if (mbstowcs(wcs, input_str, 1) == (size_t)-1)
140	{
141	#ifdef _DEBUG
142	log_error("mbstowcs(%s) error\n", input_str);
143	#endif
144	wc_len = (UTF8_fixed_width ? 2 : 1); // Fallback
145	}
146	else
147	{
148	wc_len = (UTF8_fixed_width ? 2 : wcwidth(wcs[0]));
149	}
150
151	if (*p_display_len + wc_len > max_display_len)
152	{
153	break;
154	}
155
156	i += (str_len - 1);
157	(*p_display_len) += wc_len;
158	}
159	else
160	{
161	if (*p_display_len + 1 > max_display_len)
162	{
163	break;
164	}
165	(*p_display_len)++;
166
167	// \n is regarded as 1 character wide in terminal editor, which is different from Web version
168	if (c == '\n')
169	{
170	i++;
171	*p_eol = 1;
172	break;
173	}
174	}
175	}
176
177	return i;
178	}
179
180	long split_data_lines(const char p_buf, int max_display_len, long p_line_offsets, long line_offsets_count,
181	int skip_ctrl_seq, int *p_line_widths)
182	{
183	int line_cnt = 0;
184	int len;
185	int end_of_line = 0;
186	int display_len = 0;
187
188	p_line_offsets[line_cnt] = 0L;
189
190	do
191	{
192	len = split_line(p_buf, max_display_len, &end_of_line, &display_len, skip_ctrl_seq);
193
194	if (p_line_widths)
195	{
196	p_line_widths[line_cnt] = display_len;
197	}
198
199	// Exceed max_line_cnt
200	if (line_cnt + 1 >= line_offsets_count)
201	{
202	// log_error("Line count %d reaches limit %d\n", line_cnt + 1, line_offsets_count);
203	return line_cnt;
204	}
205
206	p_line_offsets[line_cnt + 1] = p_line_offsets[line_cnt] + len;
207	line_cnt++;
208	p_buf += len;
209	} while (p_buf[0] != '\0' \|\| end_of_line);
210
211	return line_cnt;
212	}
213
214	int str_filter(char *buffer, int skip_ctrl_seq)
215	{
216	int i;
217	int j;
218
219	for (i = 0, j = 0; buffer[i] != '\0'; i++)
220	{
221	if (buffer[i] == '\r' \|\| buffer[i] == '\7') // skip
222	{
223	continue;
224	}
225
226	if (skip_ctrl_seq && buffer[i] == '\033' && buffer[i + 1] == '[') // Skip control sequence
227	{
228	i += 2;
229	while (buffer[i] != '\0' && buffer[i] != 'm')
230	{
231	i++;
232	}
233	continue;
234	}
235
236	buffer[j] = buffer[i];
237	j++;
238	}
239
240	buffer[j] = '\0';
241
242	return j;
243	}
webmaster@leafok.com	ViewVC Help
Powered by ViewVC 1.3.0-beta1