--- lbbs/src/str_process.c 2025/06/21 02:15:18 1.18 +++ lbbs/src/str_process.c 2025/11/10 12:53:16 1.29 @@ -1,24 +1,98 @@ -/*************************************************************************** - str_process.c - description - ------------------- - Copyright : (C) 2004-2025 by Leaflet - Email : leaflet@leafok.com - ***************************************************************************/ - -/*************************************************************************** - * * - * 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; either version 3 of the License, or * - * (at your option) any later version. * - * * - ***************************************************************************/ +/* SPDX-License-Identifier: GPL-3.0-or-later */ +/* + * str_process + * - common string processing features with UTF-8 support + * + * Copyright (C) 2004-2025 Leaflet + */ #include "common.h" #include "log.h" #include "str_process.h" +#include #include +#include #include +#include + +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) { @@ -26,6 +100,10 @@ int split_line(const char *buffer, int m *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++) { @@ -46,14 +124,37 @@ int split_line(const char *buffer, int m continue; } - if (c < 0 || c > 127) // GBK chinese character + if (c & 0x80) // head of multi-byte character { - if (*p_display_len + 2 > max_display_len) + 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++; - (*p_display_len) += 2; + + i += (str_len - 1); + (*p_display_len) += wc_len; } else { @@ -76,7 +177,8 @@ int split_line(const char *buffer, int m 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) +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; @@ -89,6 +191,11 @@ long split_data_lines(const char *p_buf, { 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) { @@ -99,7 +206,7 @@ long split_data_lines(const char *p_buf, p_line_offsets[line_cnt + 1] = p_line_offsets[line_cnt] + len; line_cnt++; p_buf += len; - } while (p_buf[0] != '\0'); + } while (p_buf[0] != '\0' || end_of_line); return line_cnt; }