From 5257358958cca6a8c44aed0a03aa962bcf4563a5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jonathan=20Sch=C3=B6bel?= <jonathan@xn--schbel-yxa.info>
Date: Sun, 2 Jul 2023 12:31:07 +0200
Subject: [PATCH] Text: added stupid replacement function
The copy_and_replace function replaces a single character with a string,
while copying. This may be replaced by an elaborate function as
manipulating a text normally means that manipulating is deferred until
needed, which this function contradicts to.
---
sefht.geany | 10 +--
src/lib/sefht/text.c | 31 +++++++
src/lib/sefht/text.h | 11 +++
src/lib/sefht/text_segment.c | 155 +++++++++++++++++++++++++++++++++++
tests/test_text.c | 28 +++++++
5 files changed, 230 insertions(+), 5 deletions(-)
diff --git a/sefht.geany b/sefht.geany
index f42fa15..59edf8e 100644
--- a/sefht.geany
+++ b/sefht.geany
@@ -28,7 +28,7 @@ long_line_behaviour=1
long_line_column=72
[files]
-current_page=25
+current_page=45
FILE_NAME_0=139;None;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2FREADME;0;8
FILE_NAME_1=134;None;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2F.gitignore;0;8
FILE_NAME_2=1737;Sh;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fconfigure.ac;0;8
@@ -51,10 +51,10 @@ FILE_NAME_18=19;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprg
FILE_NAME_19=19;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Fattr.h;0;8
FILE_NAME_20=26;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Fattr_static.c;0;8
FILE_NAME_21=24;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Fattr_data.h;0;8
-FILE_NAME_22=2765;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext.c;0;8
-FILE_NAME_23=19;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext.h;0;8
+FILE_NAME_22=2929;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext.c;0;8
+FILE_NAME_23=2065;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext.h;0;8
FILE_NAME_24=1036;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext_data.h;0;8
-FILE_NAME_25=4650;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext_segment.c;0;8
+FILE_NAME_25=8479;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext_segment.c;0;8
FILE_NAME_26=1083;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext_segment.h;0;8
FILE_NAME_27=900;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext_segment_mark.c;0;8
FILE_NAME_28=1867;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Fsrc%2Flib%2Fsefht%2Ftext_mark_static.c;0;8
@@ -74,7 +74,7 @@ FILE_NAME_41=8232;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fp
FILE_NAME_42=33;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_node_fragment.c;0;8
FILE_NAME_43=4771;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_text_fragment.c;0;8
FILE_NAME_44=24;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_attr.c;0;8
-FILE_NAME_45=3584;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_text.c;0;8
+FILE_NAME_45=4221;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_text.c;0;8
FILE_NAME_46=994;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_text_mark.c;0;8
FILE_NAME_47=29;C;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftests%2Ftest_validator.c;0;8
FILE_NAME_48=536;None;0;EUTF-8;1;1;0;%2Fhome%2Fjonathan%2FDokumente%2Fprojekte%2Fprgm%2Finternet%2Fweb%2FSeFHT%2Ftodo.txt;0;8
diff --git a/src/lib/sefht/text.c b/src/lib/sefht/text.c
index 5b97cb0..988efb4 100644
--- a/src/lib/sefht/text.c
+++ b/src/lib/sefht/text.c
@@ -144,6 +144,37 @@ SH_Text_copy (const struct SH_Text * text,
return copy;
}
+/*@null@*/
+/*@only@*/
+struct SH_Text *
+SH_Text_copy_and_replace (const struct SH_Text * text,
+ char old, char * new,
+ /*@null@*/ /*@out@*/ struct SH_Status * status)
+ /*@globals fileSystem@*/
+ /*@modifies fileSystem@*/
+ /*@modifies status@*/
+{
+ struct SH_Text * copy;
+
+ copy = malloc (sizeof (struct SH_Text));
+ if (NULL == copy)
+ {
+ set_status (status, E_ALLOC, 3, "malloc failed");
+ return NULL;
+ }
+
+ copy->data = copy_and_replace_text_segment (text->data,
+ old, new, status);
+ if (NULL == copy->data)
+ {
+ free (copy);
+ return NULL;
+ }
+
+ set_success (status);
+ return copy;
+}
+
size_t
SH_Text_get_length (const struct SH_Text * text,
/*@null@*/ /*@out@*/ struct SH_Status * status)
diff --git a/src/lib/sefht/text.h b/src/lib/sefht/text.h
index 59e3173..278fccf 100644
--- a/src/lib/sefht/text.h
+++ b/src/lib/sefht/text.h
@@ -68,6 +68,17 @@ SH_Text_copy (const SH_Text * text,
/*@modifies fileSystem@*/
/*@modifies status@*/;
+
+/*@null@*/
+/*@only@*/
+SH_Text *
+SH_Text_copy_and_replace (const SH_Text * text,
+ char old, char * new,
+ /*@null@*/ /*@out@*/ struct SH_Status * status)
+ /*@globals fileSystem@*/
+ /*@modifies fileSystem@*/
+ /*@modifies status@*/;
+
size_t
SH_Text_get_length (const struct SH_Text * text,
/*@null@*/ /*@out@*/ struct SH_Status * status)
diff --git a/src/lib/sefht/text_segment.c b/src/lib/sefht/text_segment.c
index 0d911cd..4cda050 100644
--- a/src/lib/sefht/text_segment.c
+++ b/src/lib/sefht/text_segment.c
@@ -367,3 +367,158 @@ squash_text_segment (const struct text_segment * segment,
copy_seg->next = NULL;
}
}
+
+static inline
+/*@null@*/
+/*@only@*/
+struct text_segment *
+copy_and_replace_text_segment (const struct text_segment * segment,
+ char old, char * new,
+ /*@null@*/ /*@out@*/
+ struct SH_Status * status)
+ /*@globals fileSystem@*/
+ /*@modifies fileSystem@*/
+ /*@modifies status@*/
+{
+ const struct text_segment * start;
+ const struct text_segment * end;
+ struct text_segment * copy;
+ struct text_segment * copy_seg;
+
+
+ copy = malloc (sizeof (struct text_segment));
+ if (NULL == copy)
+ {
+ set_status (status, E_ALLOC, 3, "malloc failed");
+ return NULL;
+ }
+
+ copy->next = NULL;
+
+ /* copy_seg is the segment we're currently copying to */
+ copy_seg = copy;
+
+ /* start to end are the segments we're copying from. */
+ /* start points to the first segment to copy.
+ * end points to the first segment not being copied in
+ * the current iteration. */
+ end = segment;
+ start = end;
+
+ /* At the beginning of each iteration start == end.
+ * This is ensured before the loop and at the end of the outer
+ * loop with the last for loop, which does the actual copying.
+ *
+ * Then end is adjusted to point after the last segment to be
+ * copied. If it isn't NULL afterwards another loop iteration
+ * has to take place.
+ *
+ * With start and end properly set, and the length and size of
+ * the resulting segment determined, the copying can start,
+ * which is just a repeated string concatenation.
+ *
+ * If another iteration is following, the next segment has to be
+ * allocated. This can't be done at the beginning, because
+ * before the first iteration the pointer to the allocated
+ * segment must be written to copy directly to create the
+ * anchor of the linked list. (So there is no ?->next
+ * to write to.)
+ * Actually copy_seg could be typed (struct text_segment **)
+ * and then be set to &(copy) before the loop and to
+ * &((*copy_seg)->next) after each iteration, but that would
+ * complicate things with another indirection...
+ * Implementing the loop as for loop and writing the allocation
+ * code into their head is also not possible, because compound
+ * statements aren't allowed there.
+ * Thus the allocation has to take place at the end of the loop. */
+ while (TRUE)
+ {
+ char * text;
+ size_t r_length;
+
+ /* calculate length, adjust end */
+ for (copy_seg->length = 0; end != NULL; end = end->next)
+ {
+ /* catch overflow, use <= instead of <
+ * to prevent overflow of copy_seg->size */
+ if (SIZE_MAX - end->length <= copy_seg->length)
+ {
+ /* stop copying */
+ end = end->next;
+
+ /*@innerbreak@*/
+ break;
+ }
+
+ copy_seg->length += end->length;
+ }
+
+ /* initialize marks
+ * TODO: copy and adjust the marks regarding to
+ * the concatenated segment */
+ init_marks (copy_seg);
+
+ /* This addition can not overflow, because in the
+ * for-loop it was tested with > instead of >= which
+ * would be appropriate. See comment there. */
+ copy_seg->size = get_alloc_size (copy_seg->length + 1);
+
+ copy_seg->text = malloc (copy_seg->size);
+ if (NULL == copy_seg->text)
+ {
+ set_status (status, E_ALLOC, 3, "malloc failed");
+ free_text_segment (copy);
+ return NULL;
+ }
+
+
+ text = copy_seg->text;
+ r_length = strlen (new);
+ /* actual copying */
+ /* Initialization was done since the beginning of
+ * the outer loop. So think of all this code standing
+ * | here.
+ * v */
+ for (; start != end; start = start->next)
+ {
+ char * s_start;
+ char * s_end;
+
+ s_start = start->text,
+ s_end = s_start + start->length;
+
+ while (s_end != s_start)
+ {
+ if (old == *s_start)
+ {
+ memcpy (text, new, r_length);
+ text += r_length;
+ copy_seg->length += r_length-1;
+ s_start++;
+ }
+ else
+ {
+ *text++ = *s_start++;
+ }
+ }
+ }
+ text[0] = '\0';
+
+ /* See comment of outer loop. (4th paragraph) */
+ if (end == NULL)
+ {
+ return copy;
+ }
+
+ copy_seg->next = malloc (sizeof (struct text_segment));
+ if (NULL == copy_seg->next)
+ {
+ set_status (status, E_ALLOC, 3, "malloc failed");
+ free_text_segment (copy);
+ return NULL;
+ }
+
+ copy_seg = copy_seg->next;
+ copy_seg->next = NULL;
+ }
+}
diff --git a/tests/test_text.c b/tests/test_text.c
index 4f03a78..247beb1 100644
--- a/tests/test_text.c
+++ b/tests/test_text.c
@@ -161,6 +161,33 @@ START_TEST (test_text_copy_with_status)
}
END_TEST
+START_TEST (test_text_copy_replace_with_status)
+{
+ struct SH_Status status;
+ struct SH_Text * text;
+ struct SH_Text * copy;
+
+ /* setup */
+ text = SH_Text_new_from_string ("Text12345", NULL);
+ ck_assert_ptr_ne (NULL, text);
+
+ /* test */
+ _status_preinit (status);
+ copy = SH_Text_copy_and_replace (text, '1', "000", &status);
+ ck_assert_int_eq (status.status, SUCCESS);
+ ck_assert_ptr_ne (copy, NULL);
+
+ ck_assert_ptr_ne (copy, text);
+ ck_assert_ptr_ne (copy->data->text, text->data->text);
+ ck_assert_str_ne (copy->data->text, text->data->text);
+ ck_assert_str_eq ("Text0002345", copy->data->text);
+
+ /* cleanup */
+ SH_Text_free (copy);
+ SH_Text_free (text);
+}
+END_TEST
+
START_TEST(test_text_get_length_no_status)
{
struct SH_Text * text;
@@ -671,6 +698,7 @@ Suite * text_suite (void)
tcase_add_test (tc_core, test_text_string_with_status);
tcase_add_test (tc_core, test_text_copy_no_status);
tcase_add_test (tc_core, test_text_copy_with_status);
+ tcase_add_test (tc_core, test_text_copy_replace_with_status);
tcase_add_test (tc_core, test_text_get_length_no_status);
tcase_add_test (tc_core, test_text_get_length_with_status);
tcase_add_test (tc_core, test_text_get_char_no_status);
--
GitLab