X-Git-Url: http://info.iut-bm.univ-fcomte.fr/pub/gitweb/simgrid.git/blobdiff_plain/1ca34b76da924ef9743ce07316d18e69ba45374f..5fd684c7bad6ba06081c88aa7054aac4e28d566f:/src/xbt/xbt_str.cpp diff --git a/src/xbt/xbt_str.cpp b/src/xbt/xbt_str.cpp index b1f9a38cfb..bfaeff97ac 100644 --- a/src/xbt/xbt_str.cpp +++ b/src/xbt/xbt_str.cpp @@ -1,170 +1,13 @@ -/* xbt_str.cpp - various helping functions to deal with strings */ +/* xbt_str.cpp - various helping functions to deal with strings */ -/* Copyright (c) 2007-2014. The SimGrid Team. - * All rights reserved. */ +/* Copyright (c) 2007-2019. The SimGrid Team. All rights reserved. */ /* This program is free software; you can redistribute it and/or modify it * under the terms of the license (GNU LGPL) which comes with this package. */ -#include -#include "src/internal_config.h" +#include "simgrid/Exception.hpp" #include "xbt/misc.h" -#include "xbt/sysdep.h" -#include "xbt/str.h" /* headers of these functions */ -#include "xbt/strbuff.h" - -/** @brief Strip whitespace (or other characters) from the end of a string. - * - * Strips the whitespaces from the end of s. - * By default (when char_list=nullptr), these characters get stripped: - * - * - " " (ASCII 32 (0x20)) space. - * - "\t" (ASCII 9 (0x09)) tab. - * - "\n" (ASCII 10 (0x0A)) line feed. - * - "\r" (ASCII 13 (0x0D)) carriage return. - * - "\0" (ASCII 0 (0x00)) nullptr. - * - "\x0B" (ASCII 11 (0x0B)) vertical tab. - * - * @param s The string to strip. Modified in place. - * @param char_list A string which contains the characters you want to strip. - */ -void xbt_str_rtrim(char *s, const char *char_list) -{ - char *cur = s; - const char *__char_list = " \t\n\r\x0B"; - char white_char[256] = { 1, 0 }; - - if (!s) - return; - - if (!char_list) { - while (*__char_list) { - white_char[(unsigned char) *__char_list++] = 1; - } - } else { - while (*char_list) { - white_char[(unsigned char) *char_list++] = 1; - } - } - - while (*cur) - ++cur; - - while ((cur >= s) && white_char[(unsigned char) *cur]) - --cur; - - *++cur = '\0'; -} - -/** @brief Strip whitespace (or other characters) from the beginning of a string. - * - * Strips the whitespaces from the begining of s. - * By default (when char_list=nullptr), these characters get stripped: - * - * - " " (ASCII 32 (0x20)) space. - * - "\t" (ASCII 9 (0x09)) tab. - * - "\n" (ASCII 10 (0x0A)) line feed. - * - "\r" (ASCII 13 (0x0D)) carriage return. - * - "\0" (ASCII 0 (0x00)) nullptr. - * - "\x0B" (ASCII 11 (0x0B)) vertical tab. - * - * @param s The string to strip. Modified in place. - * @param char_list A string which contains the characters you want to strip. - */ -void xbt_str_ltrim(char *s, const char *char_list) -{ - char *cur = s; - const char *__char_list = " \t\n\r\x0B"; - char white_char[256] = { 1, 0 }; - - if (!s) - return; - - if (!char_list) { - while (*__char_list) { - white_char[(unsigned char) *__char_list++] = 1; - } - } else { - while (*char_list) { - white_char[(unsigned char) *char_list++] = 1; - } - } - - while (*cur && white_char[(unsigned char) *cur]) - ++cur; - - memmove(s, cur, strlen(cur) + 1); -} - -/** @brief Strip whitespace (or other characters) from the end and the begining of a string. - * - * Strips the whitespaces from both the beginning and the end of s. - * By default (when char_list=nullptr), these characters get stripped: - * - * - " " (ASCII 32 (0x20)) space. - * - "\t" (ASCII 9 (0x09)) tab. - * - "\n" (ASCII 10 (0x0A)) line feed. - * - "\r" (ASCII 13 (0x0D)) carriage return. - * - "\0" (ASCII 0 (0x00)) nullptr. - * - "\x0B" (ASCII 11 (0x0B)) vertical tab. - * - * @param s The string to strip. - * @param char_list A string which contains the characters you want to strip. - */ -void xbt_str_trim(char *s, const char *char_list) -{ - if (!s) - return; - - xbt_str_rtrim(s, char_list); - xbt_str_ltrim(s, char_list); -} - -/** @brief Substitutes a char for another in a string - * - * @param str the string to modify - * @param from char to search - * @param to char to put instead - * @param occurence number of changes to do (=0 means all) - */ -void xbt_str_subst(char *str, char from, char to, int occurence) -{ - char *p = str; - while (*p != '\0') { - if (*p == from) { - *p = to; - if (occurence == 1) - return; - occurence--; - } - p++; - } -} - -/** @brief Replaces a set of variables by their values - * - * @param str The input of the replacement process - * @param patterns The changes to apply - * @return The string modified - * - * Both '$toto' and '${toto}' are valid (and the two writing are equivalent). - * - * If the variable name contains spaces, use the brace version (ie, ${toto tutu}) - * - * You can provide a default value to use if the variable is not set in the dict by using '${var:=default}' or - * '${var:-default}'. These two forms are equivalent, even if they shouldn't to respect the shell standard (:= form - * should set the value in the dict, but does not) (BUG). - */ -char *xbt_str_varsubst(const char *str, xbt_dict_t patterns) -{ - xbt_strbuff_t buff = xbt_strbuff_new_from(str); - char *res; - xbt_strbuff_varsubst(buff, patterns); - res = buff->data; - xbt_strbuff_free_container(buff); - return res; -} - +#include "xbt/str.h" /* headers of these functions */ /** @brief Splits a string into a dynar of strings * @@ -183,33 +26,31 @@ char *xbt_str_varsubst(const char *str, xbt_dict_t patterns) xbt_dynar_t xbt_str_split(const char *s, const char *sep) { xbt_dynar_t res = xbt_dynar_new(sizeof(char *), &xbt_free_ref); - const char *p, *q; - int done; const char *sep_dflt = " \t\n\r\x0B"; char is_sep[256] = { 1, 0 }; /* check what are the separators */ memset(is_sep, 0, sizeof(is_sep)); - if (!sep) { + if (not sep) { while (*sep_dflt) is_sep[(unsigned char) *sep_dflt++] = 1; } else { while (*sep) is_sep[(unsigned char) *sep++] = 1; } - is_sep[0] = 1; /* End of string is also separator */ + is_sep[0] = 1; /* End of string is also separator */ /* Do the job */ - p = s; - q = s; - done = 0; + const char* p = s; + const char* q = s; + int done = 0; if (s[0] == '\0') return res; - while (!done) { + while (not done) { char *topush; - while (!is_sep[(unsigned char) *q]) { + while (not is_sep[(unsigned char)*q]) { q++; } if (*q == '\0') @@ -225,55 +66,6 @@ xbt_dynar_t xbt_str_split(const char *s, const char *sep) return res; } -/** - * \brief This functions splits a string after using another string as separator - * For example A!!B!!C splitted after !! will return the dynar {A,B,C} - * \return An array of dynars containing the string tokens - */ -xbt_dynar_t xbt_str_split_str(const char *s, const char *sep) -{ - xbt_dynar_t res = xbt_dynar_new(sizeof(char *), &xbt_free_ref); - int done; - const char *p, *q; - - p = s; - q = s; - done = 0; - - if (s[0] == '\0') - return res; - if (sep[0] == '\0') { - s = xbt_strdup(s); - xbt_dynar_push(res, &s); - return res; - } - - while (!done) { - char *to_push; - int v = 0; - //get the start of the first occurence of the substring - q = strstr(p, sep); - //if substring was not found add the entire string - if (nullptr == q) { - v = strlen(p); - to_push = (char*) xbt_malloc(v + 1); - memcpy(to_push, p, v); - to_push[v] = '\0'; - xbt_dynar_push(res, &to_push); - done = 1; - } else { - //get the appearance - to_push = (char*) xbt_malloc(q - p + 1); - memcpy(to_push, p, q - p); - //add string terminator - to_push[q - p] = '\0'; - xbt_dynar_push(res, &to_push); - p = q + strlen(sep); - } - } - return res; -} - /** @brief Just like @ref xbt_str_split_quoted (Splits a string into a dynar of strings), but without memory allocation * * The string passed as argument must be writable (not const) @@ -287,10 +79,12 @@ xbt_dynar_t xbt_str_split_str(const char *s, const char *sep) */ xbt_dynar_t xbt_str_split_quoted_in_place(char *s) { xbt_dynar_t res = xbt_dynar_new(sizeof(char *), nullptr); - char *beg, *end; /* pointers around the parsed chunk */ - int in_simple_quote = 0, in_double_quote = 0; - int done = 0; - int ctn = 0; /* Got something in this block */ + char* beg; + char* end; /* pointers around the parsed chunk */ + int in_simple_quote = 0; + int in_double_quote = 0; + int done = 0; + int ctn = 0; /* Got something in this block */ if (s[0] == '\0') return res; @@ -300,7 +94,7 @@ xbt_dynar_t xbt_str_split_quoted_in_place(char *s) { /* do not trim leading spaces: caller responsibility to clean his cruft */ end = beg; - while (!done) { + while (not done) { switch (*end) { case '\\': ctn = 1; @@ -312,8 +106,8 @@ xbt_dynar_t xbt_str_split_quoted_in_place(char *s) { break; case '\'': ctn = 1; - if (!in_double_quote) { - in_simple_quote = !in_simple_quote; + if (not in_double_quote) { + in_simple_quote = not in_simple_quote; memmove(end, end + 1, strlen(end)); } else { /* simple quote protected by double ones */ @@ -322,8 +116,8 @@ xbt_dynar_t xbt_str_split_quoted_in_place(char *s) { break; case '"': ctn = 1; - if (!in_simple_quote) { - in_double_quote = !in_double_quote; + if (not in_simple_quote) { + in_double_quote = not in_double_quote; memmove(end, end + 1, strlen(end)); } else { /* double quote protected by simple ones */ @@ -392,94 +186,12 @@ xbt_dynar_t xbt_str_split_quoted(const char *s) char *q=xbt_strdup(p); xbt_dynar_push(res,&q); } - free(str_to_free); + xbt_free(str_to_free); xbt_dynar_shrink(res, 0); xbt_dynar_free(&parsed); return res; } -/** @brief Join a set of strings as a single string */ -char *xbt_str_join(xbt_dynar_t dyn, const char *sep) -{ - int len = 1, dyn_len = xbt_dynar_length(dyn); - unsigned int cpt; - char *cursor; - char *res, *p; - - if (!dyn_len) - return xbt_strdup(""); - - /* compute the length */ - xbt_dynar_foreach(dyn, cpt, cursor) { - len += strlen(cursor); - } - len += strlen(sep) * dyn_len; - /* Do the job */ - res = (char*) xbt_malloc(len); - p = res; - xbt_dynar_foreach(dyn, cpt, cursor) { - if ((int) cpt < dyn_len - 1) - p += snprintf(p,len, "%s%s", cursor, sep); - else - p += snprintf(p,len, "%s", cursor); - } - return res; -} - -/** @brief Join a set of strings as a single string - * - * The parameter must be a nullptr-terminated array of chars, - * just like xbt_dynar_to_array() produces - */ -char *xbt_str_join_array(const char *const *strs, const char *sep) -{ - char *res,*q; - int amount_strings=0; - int len=0; - int i; - - if ((!strs) || (!strs[0])) - return xbt_strdup(""); - - /* compute the length before malloc */ - for (i=0;strs[i];i++) { - len += strlen(strs[i]); - amount_strings++; - } - len += strlen(sep) * amount_strings; - - /* Do the job */ - res = (char*) xbt_malloc(len); - q = res; - for (i=0;strs[i];i++) { - if (i!=0) { // not first loop - q += snprintf(q,len, "%s%s", sep, strs[i]); - } else { - q += snprintf(q,len, "%s",strs[i]); - } - } - return res; -} - -/** @brief creates a new string containing what can be read on a fd */ -char *xbt_str_from_file(FILE * file) -{ - xbt_strbuff_t buff = xbt_strbuff_new(); - char *res; - char bread[1024]; - memset(bread, 0, 1024); - - while (!feof(file)) { - int got = fread(bread, 1, 1023, file); - bread[got] = '\0'; - xbt_strbuff_append(buff, bread); - } - - res = buff->data; - xbt_strbuff_free_container(buff); - return res; -} - /** @brief Parse an integer out of a string, or raise an error * * The @a str is passed as argument to your @a error_msg, as follows: @@ -487,7 +199,7 @@ char *xbt_str_from_file(FILE * file) */ long int xbt_str_parse_int(const char* str, const char* error_msg) { - char *endptr; + char* endptr; if (str == nullptr || str[0] == '\0') THROWF(arg_error, 0, error_msg, str); @@ -515,107 +227,3 @@ double xbt_str_parse_double(const char* str, const char* error_msg) return res; } - -#ifdef SIMGRID_TEST -#include -#include "xbt/str.h" - -XBT_TEST_SUITE("xbt_str", "String Handling"); - -#define mytest(name, input, expected) \ - xbt_test_add(name); \ - d=xbt_str_split_quoted(input); \ - s=xbt_str_join(d,"XXX"); \ - xbt_test_assert(!strcmp(s,expected),\ - "Input (%s) leads to (%s) instead of (%s)", \ - input,s,expected);\ - free(s); \ - xbt_dynar_free(&d); -XBT_TEST_UNIT("xbt_str_split_quoted", test_split_quoted, "test the function xbt_str_split_quoted") -{ - xbt_dynar_t d; - char *s; - - mytest("Empty", "", ""); - mytest("Basic test", "toto tutu", "totoXXXtutu"); - mytest("Useless backslashes", "\\t\\o\\t\\o \\t\\u\\t\\u", "totoXXXtutu"); - mytest("Protected space", "toto\\ tutu", "toto tutu"); - mytest("Several spaces", "toto tutu", "totoXXXtutu"); - mytest("LTriming", " toto tatu", "totoXXXtatu"); - mytest("Triming", " toto tutu ", "totoXXXtutu"); - mytest("Single quotes", "'toto tutu' tata", "toto tutuXXXtata"); - mytest("Double quotes", "\"toto tutu\" tata", "toto tutuXXXtata"); - mytest("Mixed quotes", "\"toto' 'tutu\" tata", "toto' 'tutuXXXtata"); - mytest("Backslashed quotes", "\\'toto tutu\\' tata", "'totoXXXtutu'XXXtata"); - mytest("Backslashed quotes + quotes", "'toto \\'tutu' tata", "toto 'tutuXXXtata"); -} - -#define mytest_str(name, input, separator, expected) \ - xbt_test_add(name); \ - d=xbt_str_split_str(input, separator); \ - s=xbt_str_join(d,"XXX"); \ - xbt_test_assert(!strcmp(s,expected),\ - "Input (%s) leads to (%s) instead of (%s)", \ - input,s,expected);\ - free(s); \ - xbt_dynar_free(&d); - -XBT_TEST_UNIT("xbt_str_split_str", test_split_str, "test the function xbt_str_split_str") -{ - xbt_dynar_t d; - char *s; - - mytest_str("Empty string and separator", "", "", ""); - mytest_str("Empty string", "", "##", ""); - mytest_str("Empty separator", "toto", "", "toto"); - mytest_str("String with no separator in it", "toto", "##", "toto"); - mytest_str("Basic test", "toto##tutu", "##", "totoXXXtutu"); -} - -#define test_parse_error(function, name, variable, str) \ - do { \ - xbt_test_add(name); \ - try { \ - variable = function(str, "Parse error"); \ - xbt_test_fail("The test '%s' did not detect the problem",name ); \ - } catch(xbt_ex& e) { \ - if (e.category != arg_error) { \ - xbt_test_exception(e); \ - } \ - } \ - } while (0) -#define test_parse_ok(function, name, variable, str, value) \ - do { \ - xbt_test_add(name); \ - try { \ - variable = function(str, "Parse error"); \ - } catch(xbt_ex& e) { \ - xbt_test_exception(e); \ - } \ - xbt_test_assert(variable == value, "Fail to parse '%s'", str); \ - } while (0) - -XBT_TEST_UNIT("xbt_str_parse", test_parse, "Test the parsing functions") -{ - int rint = -9999; - test_parse_ok(xbt_str_parse_int, "Parse int", rint, "42", 42); - test_parse_ok(xbt_str_parse_int, "Parse 0 as an int", rint, "0", 0); - test_parse_ok(xbt_str_parse_int, "Parse -1 as an int", rint, "-1", -1); - - test_parse_error(xbt_str_parse_int, "Parse int + noise", rint, "342 cruft"); - test_parse_error(xbt_str_parse_int, "Parse nullptr as an int", rint, nullptr); - test_parse_error(xbt_str_parse_int, "Parse '' as an int", rint, ""); - test_parse_error(xbt_str_parse_int, "Parse cruft as an int", rint, "cruft"); - - double rdouble = -9999; - test_parse_ok(xbt_str_parse_double, "Parse 42 as a double", rdouble, "42", 42); - test_parse_ok(xbt_str_parse_double, "Parse 42.5 as a double", rdouble, "42.5", 42.5); - test_parse_ok(xbt_str_parse_double, "Parse 0 as a double", rdouble, "0", 0); - test_parse_ok(xbt_str_parse_double, "Parse -1 as a double", rdouble, "-1", -1); - - test_parse_error(xbt_str_parse_double, "Parse double + noise", rdouble, "342 cruft"); - test_parse_error(xbt_str_parse_double, "Parse nullptr as a double", rdouble, nullptr); - test_parse_error(xbt_str_parse_double, "Parse '' as a double", rdouble, ""); - test_parse_error(xbt_str_parse_double, "Parse cruft as a double", rdouble, "cruft"); -} -#endif /* SIMGRID_TEST */