/* xbt_str.c - various helping functions to deal with strings */
-/* Copyright (c) 2007, 2008, 2009, 2010. The SimGrid Team.
+/* Copyright (c) 2007-2013. The SimGrid Team.
* All rights reserved. */
/* This program is free software; you can redistribute it and/or modify it
* Strips the whitespaces from the end of s.
* By default (when char_list=NULL), 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)) NULL.
- * - "\x0B" (ASCII 11 (0x0B)) vertical tab.
+ * - " " (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)) NULL.
+ * - "\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.
* Strips the whitespaces from the begining of s.
* By default (when char_list=NULL), 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)) NULL.
- * - "\x0B" (ASCII 11 (0x0B)) vertical tab.
+ * - " " (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)) NULL.
+ * - "\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.
* Strips the whitespaces from both the beginning and the end of s.
* By default (when char_list=NULL), 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)) NULL.
- * - "\x0B" (ASCII 11 (0x0B)) vertical tab.
+ * - " " (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)) NULL.
+ * - "\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.
/** @brief Replaces a set of variables by their values
*
- * @param str where to apply the change
- * @param patterns what to change
+ * @param str The input of the replacement process
+ * @param patterns The changes to apply
* @return The string modified
*
- * Check xbt_strbuff_varsubst() for more details, and remember that the string may be reallocated (moved) in the process.
+ * 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(char *str, xbt_dict_t patterns)
+char *xbt_str_varsubst(const char *str, xbt_dict_t patterns)
{
xbt_strbuff_t buff = xbt_strbuff_new_from(str);
char *res;
*
* By default (with sep=NULL), these characters are used as separator:
*
- * - " " (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)) NULL.
- * - "\x0B" (ASCII 11 (0x0B)) vertical tab.
+ * - " " (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)) NULL.
+ * - "\x0B" (ASCII 11 (0x0B)) vertical tab.
*/
xbt_dynar_t xbt_str_split(const char *s, const char *sep)
//if substring was not found add the entire string
if (NULL == q) {
v = strlen(p);
- to_push = malloc(v + 1);
+ to_push = 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 = malloc(q - p + 1);
+ to_push = xbt_malloc(q - p + 1);
memcpy(to_push, p, q - p);
//add string terminator
to_push[q - p] = '\0';
*
* The string passed as argument must be writable (not const)
* The elements of the dynar are just parts of the string passed as argument.
+ * So if you don't store that argument elsewhere, you should free it in addition
+ * to freeing the dynar. This can be done by simply freeing the first argument
+ * of the dynar:
+ * free(xbt_dynar_get_ptr(dynar,0));
*
- * To free the structure constructed by this function, free the first element and free the dynar:
- *
- * free(xbt_dynar_get_ptr(dynar,0));
- * xbt_dynar_free(&dynar);
+ * Actually this function puts a bunch of \0 in the memory area you passed as
+ * argument to separate the elements, and pushes the address of each chunk
+ * in the resulting dynar. Yes, that's uneven. Yes, that's gory. But that's efficient.
*/
xbt_dynar_t xbt_str_split_quoted_in_place(char *s) {
xbt_dynar_t res = xbt_dynar_new(sizeof(char *), NULL);
beg = s;
- /* do not trim leading spaces: caller responsability to clean his cruft */
+ /* do not trim leading spaces: caller responsibility to clean his cruft */
end = beg;
while (!done) {
return res;
}
-#if defined(SIMGRID_NEED_GETLINE) || defined(DOXYGEN)
/** @brief Get a single line from the stream (reimplementation of the GNU getline)
*
- * This is a redefinition of the GNU getline function, used on platforms where it does not exists.
+ * This is a reimplementation of the GNU getline function, so that our code don't depends on the GNU libc.
+ *
+ * xbt_getline() reads an entire line from stream, storing the address of the
+ * buffer containing the text into *buf. The buffer is null-terminated and
+ * includes the newline character, if one was found.
*
- * getline() reads an entire line from stream, storing the address of the buffer
- * containing the text into *buf. The buffer is null-terminated and includes
- * the newline character, if one was found.
+ * If *buf is NULL, then xbt_getline() will allocate a buffer for storing the
+ * line, which should be freed by the user program.
*
- * If *buf is NULL, then getline() will allocate a buffer for storing the line,
- * which should be freed by the user program. Alternatively, before calling getline(),
- * *buf can contain a pointer to a malloc()-allocated buffer *n bytes in size. If the buffer
- * is not large enough to hold the line, getline() resizes it with realloc(), updating *buf and *n
- * as necessary. In either case, on a successful call, *buf and *n will be updated to
- * reflect the buffer address and allocated size respectively.
+ * Alternatively, before calling xbt_getline(), *buf can contain a pointer to a
+ * malloc()-allocated buffer *n bytes in size. If the buffer is not large
+ * enough to hold the line, xbt_getline() resizes it with realloc(), updating
+ * *buf and *n as necessary.
+ *
+ * In either case, on a successful call, *buf and *n will be updated to reflect
+ * the buffer address and allocated size respectively.
*/
-long getline(char **buf, size_t * n, FILE * stream)
+ssize_t xbt_getline(char **buf, size_t *n, FILE *stream)
{
-
- size_t i;
+ ssize_t i;
int ch;
+ ch = getc(stream);
+ if (ferror(stream) || feof(stream))
+ return -1;
+
if (!*buf) {
- *buf = xbt_malloc(512);
*n = 512;
+ *buf = xbt_malloc(*n);
}
- if (feof(stream))
- return (ssize_t) - 1;
-
- for (i = 0; (ch = fgetc(stream)) != EOF; i++) {
-
- if (i >= (*n) + 1)
+ i = 0;
+ do {
+ if (i == *n)
*buf = xbt_realloc(*buf, *n += 512);
-
- (*buf)[i] = ch;
-
- if ((*buf)[i] == '\n') {
- i++;
- (*buf)[i] = '\0';
- break;
- }
- }
+ (*buf)[i++] = ch;
+ } while (ch != '\n' && (ch = getc(stream)) != EOF);
if (i == *n)
*buf = xbt_realloc(*buf, *n += 1);
-
(*buf)[i] = '\0';
- return (ssize_t) i;
+ return i;
}
-#endif /* HAVE_GETLINE */
-
/*
* Diff related functions
+ *
+ * Implementation of the algorithm described in "An O(NP) Sequence Comparison
+ * Algorithm", by Sun Wu, Udi Manber, Gene Myers, and Webb Miller (Information
+ * Processing Letters 35(6):317-323, 1990), with the linear-space
+ * divide-and-conquer strategy described in "An O(ND) Difference Algorithm and
+ * Its Variations", by Eugene W. Myers (Algorithmica 1:251-266, 1986).
*/
-static xbt_matrix_t diff_build_LCS(xbt_dynar_t da, xbt_dynar_t db)
+
+struct subsequence {
+ int x, y; /* starting coordinates */
+ int len; /* length */
+};
+
+static XBT_INLINE
+void diff_snake(const char *vec_a[], int a0, int len_a,
+ const char *vec_b[], int b0, int len_b,
+ struct subsequence *seqs, int *fp, int k, int limit)
{
- xbt_matrix_t C =
- xbt_matrix_new(xbt_dynar_length(da), xbt_dynar_length(db),
- sizeof(int), NULL);
- unsigned long i, j;
-
- /* Compute the LCS */
- /*
- C = array(0..m, 0..n)
- for i := 0..m
- C[i,0] = 0
- for j := 1..n
- C[0,j] = 0
- for i := 1..m
- for j := 1..n
- if X[i] = Y[j]
- C[i,j] := C[i-1,j-1] + 1
- else:
- C[i,j] := max(C[i,j-1], C[i-1,j])
- return C[m,n]
- */
- if (xbt_dynar_length(db) != 0)
- for (i = 0; i < xbt_dynar_length(da); i++)
- *((int *) xbt_matrix_get_ptr(C, i, 0)) = 0;
-
- if (xbt_dynar_length(da) != 0)
- for (j = 0; j < xbt_dynar_length(db); j++)
- *((int *) xbt_matrix_get_ptr(C, 0, j)) = 0;
-
- for (i = 1; i < xbt_dynar_length(da); i++)
- for (j = 1; j < xbt_dynar_length(db); j++) {
-
- if (!strcmp
- (xbt_dynar_get_as(da, i, char *),
- xbt_dynar_get_as(db, j, char *)))
- *((int *) xbt_matrix_get_ptr(C, i, j)) =
- xbt_matrix_get_as(C, i - 1, j - 1, int) + 1;
- else
- *((int *) xbt_matrix_get_ptr(C, i, j)) =
- max(xbt_matrix_get_as(C, i, j - 1, int),
- xbt_matrix_get_as(C, i - 1, j, int));
- }
- return C;
+ int record_seq;
+ int x, y;
+ int fp_left = fp[k - 1] + 1;
+ int fp_right = fp[k + 1];
+ if (fp_left > fp_right) {
+ x = fp_left;
+ record_seq = k - 1;
+ } else {
+ x = fp_right;
+ record_seq = k + 1;
+ }
+ y = x - k;
+ if (x + y <= limit) {
+ seqs[k].x = x;
+ seqs[k].y = y;
+ record_seq = k;
+ } else {
+ seqs[k] = seqs[record_seq];
+ }
+ while (x < len_a && y < len_b && !strcmp(vec_a[a0 + x], vec_b[b0 + y]))
+ ++x, ++y;
+ fp[k] = x;
+ if (record_seq == k)
+ seqs[k].len = x - seqs[k].x;
}
-static void diff_build_diff(xbt_dynar_t res,
- xbt_matrix_t C,
- xbt_dynar_t da, xbt_dynar_t db, int i, int j)
+/* Returns the length of a shortest edit script, and a common
+ * subsequence from the middle.
+ */
+static int diff_middle_subsequence(const char *vec_a[], int a0, int len_a,
+ const char *vec_b[], int b0, int len_b,
+ struct subsequence *subseq,
+ struct subsequence *seqs, int *fp)
{
- char *topush;
- /* Construct the diff
- function printDiff(C[0..m,0..n], X[1..m], Y[1..n], i, j)
- if i > 0 and j > 0 and X[i] = Y[j]
- printDiff(C, X, Y, i-1, j-1)
- print " " + X[i]
- else
- if j > 0 and (i = 0 or C[i,j-1] >= C[i-1,j])
- printDiff(C, X, Y, i, j-1)
- print "+ " + Y[j]
- else if i > 0 and (j = 0 or C[i,j-1] < C[i-1,j])
- printDiff(C, X, Y, i-1, j)
- print "- " + X[i]
- */
-
- if (i >= 0 && j >= 0 && !strcmp(xbt_dynar_get_as(da, i, char *),
- xbt_dynar_get_as(db, j, char *))) {
- diff_build_diff(res, C, da, db, i - 1, j - 1);
- topush = bprintf(" %s", xbt_dynar_get_as(da, i, char *));
- xbt_dynar_push(res, &topush);
- } else if (j >= 0 &&
- (i <= 0 || j == 0
- || xbt_matrix_get_as(C, i, j - 1,
- int) >= xbt_matrix_get_as(C, i - 1, j,
- int))) {
- diff_build_diff(res, C, da, db, i, j - 1);
- topush = bprintf("+ %s", xbt_dynar_get_as(db, j, char *));
- xbt_dynar_push(res, &topush);
- } else if (i >= 0 &&
- (j <= 0
- || xbt_matrix_get_as(C, i, j - 1, int) < xbt_matrix_get_as(C,
- i
- -
- 1,
- j,
- int)))
- {
- diff_build_diff(res, C, da, db, i - 1, j);
- topush = bprintf("- %s", xbt_dynar_get_as(da, i, char *));
- xbt_dynar_push(res, &topush);
- } else if (i <= 0 && j <= 0) {
- return;
+ const int delta = len_a - len_b;
+ const int limit = (len_a + len_b) / 2;
+ int kmin;
+ int kmax;
+ int k;
+ int p = -1;
+
+ if (delta >= 0) {
+ kmin = 0;
+ kmax = delta;
} else {
- THROWF(arg_error, 0, "Invalid values: i=%d, j=%d", i, j);
+ kmin = delta;
+ kmax = 0;
}
+ for (k = kmin; k <= kmax; k++)
+ fp[k] = -1;
+ do {
+ p++;
+ fp[kmin - p - 1] = fp[kmax + p + 1] = -1;
+ for (k = kmax + p; k > delta; k--)
+ diff_snake(vec_a, a0, len_a, vec_b, b0, len_b, seqs, fp, k, limit);
+ for (k = kmin - p; k <= delta; k++)
+ diff_snake(vec_a, a0, len_a, vec_b, b0, len_b, seqs, fp, k, limit);
+ } while (fp[delta] != len_a);
+
+ subseq->x = a0 + seqs[delta].x;
+ subseq->y = b0 + seqs[delta].y;
+ subseq->len = seqs[delta].len;
+ return abs(delta) + 2 * p;;
+}
+/* Finds a longest common subsequence.
+ * Returns its length.
+ */
+static int diff_compute_lcs(const char *vec_a[], int a0, int len_a,
+ const char *vec_b[], int b0, int len_b,
+ xbt_dynar_t common_sequence,
+ struct subsequence *seqs, int *fp)
+{
+ if (len_a > 0 && len_b > 0) {
+ struct subsequence subseq;
+ int ses_len = diff_middle_subsequence(vec_a, a0, len_a, vec_b, b0, len_b,
+ &subseq, seqs, fp);
+ int lcs_len = (len_a + len_b - ses_len) / 2;
+ if (lcs_len == 0) {
+ return 0;
+ } else if (ses_len > 1) {
+ int lcs_len1 = subseq.len;
+ if (lcs_len1 < lcs_len)
+ lcs_len1 += diff_compute_lcs(vec_a, a0, subseq.x - a0,
+ vec_b, b0, subseq.y - b0,
+ common_sequence, seqs, fp);
+ if (subseq.len > 0)
+ xbt_dynar_push(common_sequence, &subseq);
+ if (lcs_len1 < lcs_len) {
+ int u = subseq.x + subseq.len;
+ int v = subseq.y + subseq.len;
+ diff_compute_lcs(vec_a, u, a0 + len_a - u, vec_b, v, b0 + len_b - v,
+ common_sequence, seqs, fp);
+ }
+ } else {
+ int len = MIN(len_a, len_b) - subseq.len;
+ if (subseq.x == a0 && subseq.y == b0) {
+ if (subseq.len > 0)
+ xbt_dynar_push(common_sequence, &subseq);
+ if (len > 0) {
+ struct subsequence subseq0 = {a0 + len_a - len,
+ b0 + len_b - len, len};
+ xbt_dynar_push(common_sequence, &subseq0);
+ }
+ } else {
+ if (len > 0) {
+ struct subsequence subseq0 = {a0, b0, len};
+ xbt_dynar_push(common_sequence, &subseq0);
+ }
+ if (subseq.len > 0)
+ xbt_dynar_push(common_sequence, &subseq);
+ }
+ }
+ return lcs_len;
+ } else {
+ return 0;
+ }
+}
+
+static int diff_member(const char *s, const char *vec[], int from, int to)
+{
+ for ( ; from < to ; from++)
+ if (!strcmp(s, vec[from]))
+ return 1;
+ return 0;
+}
+
+/* Extract common prefix.
+ */
+static void diff_easy_prefix(const char *vec_a[], int *a0_p, int *len_a_p,
+ const char *vec_b[], int *b0_p, int *len_b_p,
+ xbt_dynar_t common_sequence)
+{
+ int a0 = *a0_p;
+ int b0 = *b0_p;
+ int len_a = *len_a_p;
+ int len_b = *len_b_p;
+
+ while (len_a > 0 && len_b > 0) {
+ struct subsequence subseq = {a0, b0, 0};
+ while (len_a > 0 && len_b > 0 && !strcmp(vec_a[a0], vec_b[b0])) {
+ a0++, len_a--;
+ b0++, len_b--;
+ subseq.len++;
+ }
+ if (subseq.len > 0)
+ xbt_dynar_push(common_sequence, &subseq);
+ if (len_a > 0 && len_b > 0 &&
+ !diff_member(vec_a[a0], vec_b, b0 + 1, b0 + len_b)) {
+ a0++, len_a--;
+ } else {
+ break;
+ }
+ }
+
+ *a0_p = a0;
+ *b0_p = b0;
+ *len_a_p = len_a;
+ *len_b_p = len_b;
+}
+
+/* Extract common suffix.
+ */
+static void diff_easy_suffix(const char *vec_a[], int *a0_p, int *len_a_p,
+ const char *vec_b[], int *b0_p, int *len_b_p,
+ xbt_dynar_t common_suffix)
+{
+ int a0 = *a0_p;
+ int b0 = *b0_p;
+ int len_a = *len_a_p;
+ int len_b = *len_b_p;
+
+ while (len_a > 0 && len_b > 0){
+ struct subsequence subseq;
+ subseq.len = 0;
+ while (len_a > 0 && len_b > 0 &&
+ !strcmp(vec_a[a0 + len_a - 1], vec_b[b0 + len_b - 1])) {
+ len_a--;
+ len_b--;
+ subseq.len++;
+ }
+ if (subseq.len > 0) {
+ subseq.x = a0 + len_a;
+ subseq.y = b0 + len_b;
+ xbt_dynar_push(common_suffix, &subseq);
+ }
+ if (len_a > 0 && len_b > 0 &&
+ !diff_member(vec_b[b0 + len_b - 1], vec_a, a0, a0 + len_a - 1)) {
+ len_b--;
+ } else {
+ break;
+ }
+ }
+
+ *a0_p = a0;
+ *b0_p = b0;
+ *len_a_p = len_a;
+ *len_b_p = len_b;
}
/** @brief Compute the unified diff of two strings */
-char *xbt_str_diff(char *a, char *b)
+char *xbt_str_diff(const char *a, const char *b)
{
xbt_dynar_t da = xbt_str_split(a, "\n");
xbt_dynar_t db = xbt_str_split(b, "\n");
+ xbt_dynar_t common_sequence, common_suffix;
+ size_t len;
+ const char **vec_a, **vec_b;
+ int a0, b0;
+ int len_a, len_b;
+ int max;
+ int *fp_base, *fp;
+ struct subsequence *seqs_base, *seqs;
+ struct subsequence subseq;
+ xbt_dynar_t diff;
+ char *res;
+ int x, y;
+ unsigned s;
+
+ /* Clean empty lines at the end of da and db */
+ len = strlen(a);
+ if (len > 0 && a[len - 1] == '\n')
+ xbt_dynar_pop(da, NULL);
+ len = strlen(b);
+ if (len > 0 && b[len - 1] == '\n')
+ xbt_dynar_pop(db, NULL);
+
+ /* Various initializations */
+ /* Assume that dynar's content is contiguous */
+ a0 = 0;
+ len_a = xbt_dynar_length(da);
+ vec_a = len_a ? xbt_dynar_get_ptr(da, 0) : NULL;
+ b0 = 0;
+ len_b = xbt_dynar_length(db);
+ vec_b = len_b ? xbt_dynar_get_ptr(db, 0) : NULL;
+ max = MAX(len_a, len_b) + 1;
+ fp_base = xbt_new(int, 2 * max + 1);
+ fp = fp_base + max; /* indexes in [-max..max] */
+ seqs_base = xbt_new(struct subsequence, 2 * max + 1);
+ seqs = seqs_base + max; /* indexes in [-max..max] */
+ common_sequence = xbt_dynar_new(sizeof(struct subsequence), NULL);
+ common_suffix = xbt_dynar_new(sizeof(struct subsequence), NULL);
+
+ /* Add a sentinel a the end of the sequence */
+ subseq.x = len_a;
+ subseq.y = len_b;
+ subseq.len = 0;
+ xbt_dynar_push(common_suffix, &subseq);
+
+ /* Compute the Longest Common Subsequence */
+ diff_easy_prefix(vec_a, &a0, &len_a, vec_b, &b0, &len_b, common_sequence);
+ diff_easy_suffix(vec_a, &a0, &len_a, vec_b, &b0, &len_b, common_suffix);
+ diff_compute_lcs(vec_a, a0, len_a, vec_b, b0, len_b, common_sequence, seqs, fp);
+ while (!xbt_dynar_is_empty(common_suffix)) {
+ xbt_dynar_pop(common_suffix, &subseq);
+ xbt_dynar_push(common_sequence, &subseq);
+ }
- xbt_matrix_t C = diff_build_LCS(da, db);
- xbt_dynar_t diff = xbt_dynar_new(sizeof(char *), &xbt_free_ref);
- char *res = NULL;
-
- diff_build_diff(diff, C, da, db, xbt_dynar_length(da) - 1,
- xbt_dynar_length(db) - 1);
- /* Clean empty lines at the end */
- while (xbt_dynar_length(diff) > 0) {
- char *str;
- xbt_dynar_pop(diff, &str);
- if (str[0] == '\0' || !strcmp(str, " ")) {
- free(str);
- } else {
- xbt_dynar_push(diff, &str);
- break;
+ /* Build a Shortest Edit Script, and the final result */
+ diff = xbt_dynar_new(sizeof(char *), &xbt_free_ref);
+ x = 0;
+ y = 0;
+ xbt_dynar_foreach(common_sequence, s, subseq) {
+ while (x < subseq.x) {
+ char *topush = bprintf("- %s", vec_a[x++]);
+ xbt_dynar_push_as(diff, char*, topush);
+ }
+ while (y < subseq.y) {
+ char *topush = bprintf("+ %s", vec_b[y++]);
+ xbt_dynar_push_as(diff, char*, topush);
+ }
+ while (x < subseq.x + subseq.len) {
+ char *topush = bprintf(" %s", vec_a[x++]);
+ xbt_dynar_push_as(diff, char*, topush);
+ y++;
}
}
res = xbt_str_join(diff, "\n");
- xbt_dynar_free(&da);
+ xbt_free(fp_base);
+ xbt_free(seqs_base);
xbt_dynar_free(&db);
+ xbt_dynar_free(&da);
+ xbt_dynar_free(&common_sequence);
+ xbt_dynar_free(&common_suffix);
xbt_dynar_free(&diff);
- xbt_matrix_free(C);
return res;
}
return res;
}
+/* @brief Retrun 1 if string 'str' starts with string 'start'
+ *
+ * \param str a string
+ * \param start the string to search in str
+ *
+ * \return 1 if 'str' starts with 'start'
+ */
+int xbt_str_start_with(const char* str, const char* start)
+{
+ int i;
+ size_t l_str = strlen(str);
+ size_t l_start = strlen(start);
+
+ if(l_start > l_str) return 0;
+
+ for(i = 0; i< l_start; i++){
+ if(str[i] != start[i]) return 0;
+ }
+
+ return 1;
+}
+
#ifdef SIMGRID_TEST
#include "xbt/str.h"
mytest_str("String with no separator in it", "toto", "##", "toto");
mytest_str("Basic test", "toto##tutu", "##", "totoXXXtutu");
}
+
+/* Last args are format string and parameters for xbt_test_add */
+#define mytest_diff(s1, s2, diff, ...) \
+ do { \
+ char *mytest_diff_res; \
+ xbt_test_add(__VA_ARGS__); \
+ mytest_diff_res = xbt_str_diff(s1, s2); \
+ xbt_test_assert(!strcmp(mytest_diff_res, diff), \
+ "Wrong output:\n--- got:\n%s\n--- expected:\n%s\n---", \
+ mytest_diff_res, diff); \
+ free(mytest_diff_res); \
+ } while (0)
+
+XBT_TEST_UNIT("xbt_str_diff", test_diff, "test the function xbt_str_diff")
+{
+ unsigned i;
+
+ /* Trivial cases */
+ mytest_diff("a", "a", " a", "1 word, no difference");
+ mytest_diff("a", "A", "- a\n+ A", "1 word, different");
+ mytest_diff("a\n", "a\n", " a", "1 line, no difference");
+ mytest_diff("a\n", "A\n", "- a\n+ A", "1 line, different");
+
+ /* Empty strings */
+ mytest_diff("", "", "", "empty strings");
+ mytest_diff("", "a", "+ a", "1 word, added");
+ mytest_diff("a", "", "- a", "1 word, removed");
+ mytest_diff("", "a\n", "+ a", "1 line, added");
+ mytest_diff("a\n", "", "- a", "1 line, removed");
+ mytest_diff("", "a\nb\nc\n", "+ a\n+ b\n+ c", "4 lines, all added");
+ mytest_diff("a\nb\nc\n", "", "- a\n- b\n- c", "4 lines, all removed");
+
+ /* Empty lines */
+ mytest_diff("\n", "\n", " ", "empty lines");
+ mytest_diff("", "\n", "+ ", "empty line, added");
+ mytest_diff("\n", "", "- ", "empty line, removed");
+
+ mytest_diff("a", "\na", "+ \n a", "empty line added before word");
+ mytest_diff("a", "a\n\n", " a\n+ ", "empty line added after word");
+ mytest_diff("\na", "a", "- \n a", "empty line removed before word");
+ mytest_diff("a\n\n", "a", " a\n- ", "empty line removed after word");
+
+ mytest_diff("a\n", "\na\n", "+ \n a", "empty line added before line");
+ mytest_diff("a\n", "a\n\n", " a\n+ ", "empty line added after line");
+ mytest_diff("\na\n", "a\n", "- \n a", "empty line removed before line");
+ mytest_diff("a\n\n", "a\n", " a\n- ", "empty line removed after line");
+
+ mytest_diff("a\nb\nc\nd\n", "\na\nb\nc\nd\n", "+ \n a\n b\n c\n d",
+ "empty line added before 4 lines");
+ mytest_diff("a\nb\nc\nd\n", "a\nb\nc\nd\n\n", " a\n b\n c\n d\n+ ",
+ "empty line added after 4 lines");
+ mytest_diff("\na\nb\nc\nd\n", "a\nb\nc\nd\n", "- \n a\n b\n c\n d",
+ "empty line removed before 4 lines");
+ mytest_diff("a\nb\nc\nd\n\n", "a\nb\nc\nd\n", " a\n b\n c\n d\n- ",
+ "empty line removed after 4 lines");
+
+ /* Missing newline at the end of one of the strings */
+ mytest_diff("a\n", "a", " a", "1 line, 1 word, no difference");
+ mytest_diff("a", "a\n", " a", "1 word, 1 line, no difference");
+ mytest_diff("a\n", "A", "- a\n+ A", "1 line, 1 word, different");
+ mytest_diff("a", "A\n", "- a\n+ A", "1 word, 1 line, different");
+
+ mytest_diff("a\nb\nc\nd", "a\nb\nc\nd\n", " a\n b\n c\n d",
+ "4 lines, no newline on first");
+ mytest_diff("a\nb\nc\nd\n", "a\nb\nc\nd", " a\n b\n c\n d",
+ "4 lines, no newline on second");
+
+ /* Four lines, all combinations of differences */
+ for (i = 0 ; i < (1U << 4) ; i++) {
+ char d2[4 + 1];
+ char s2[4 * 2 + 1];
+ char res[4 * 8 + 1];
+ char *pd = d2;
+ char *ps = s2;
+ char *pr = res;
+ unsigned j = 0;
+ while (j < 4) {
+ unsigned k;
+ for (/* j */ ; j < 4 && !(i & (1U << j)) ; j++) {
+ *pd++ = "abcd"[j];
+ ps += sprintf(ps, "%c\n", "abcd"[j]);
+ pr += sprintf(pr, " %c\n", "abcd"[j]);
+ }
+ for (k = j ; k < 4 && (i & (1U << k)) ; k++) {
+ *pd++ = "ABCD"[k];
+ ps += sprintf(ps, "%c\n", "ABCD"[k]);
+ pr += sprintf(pr, "- %c\n", "abcd"[k]);
+ }
+ for (/* j */ ; j < k ; j++) {
+ pr += sprintf(pr, "+ %c\n", "ABCD"[j]);
+ }
+ }
+ *pd = '\0';
+ *--pr = '\0'; /* strip last '\n' from expected result */
+ mytest_diff("a\nb\nc\nd\n", s2, res,
+ "compare (abcd) with changed (%s)", d2);
+ }
+
+ /* Subsets of four lines, do not test for empty subset */
+ for (i = 1 ; i < (1U << 4) ; i++) {
+ char d2[4 + 1];
+ char s2[4 * 2 + 1];
+ char res[4 * 8 + 1];
+ char *pd = d2;
+ char *ps = s2;
+ char *pr = res;
+ unsigned j = 0;
+ while (j < 4) {
+ for (/* j */ ; j < 4 && (i & (1U << j)) ; j++) {
+ *pd++ = "abcd"[j];
+ ps += sprintf(ps, "%c\n", "abcd"[j]);
+ pr += sprintf(pr, " %c\n", "abcd"[j]);
+ }
+ for (/* j */; j < 4 && !(i & (1U << j)) ; j++) {
+ pr += sprintf(pr, "- %c\n", "abcd"[j]);
+ }
+ }
+ *pd = '\0';
+ *--pr = '\0'; /* strip last '\n' from expected result */
+ mytest_diff("a\nb\nc\nd\n", s2, res,
+ "compare (abcd) with subset (%s)", d2);
+
+ for (pr = res ; *pr != '\0' ; pr++)
+ if (*pr == '-')
+ *pr = '+';
+ mytest_diff(s2, "a\nb\nc\nd\n", res,
+ "compare subset (%s) with (abcd)", d2);
+ }
+}
+
#endif /* SIMGRID_TEST */