Debug function to dump the whole index to stdout.

2000-04-12  NotZed  <NotZed@HelixCode.com>

        * find.c (ibex_dump_all): Debug function to dump the whole index
        to stdout.

        * words.c (get_ibex_file): Use g_strdup(), not strdup().

2000-04-11  NotZed  <NotZed@HelixCode.com>

        * file.c (write_word): Always write out all words we have (even if
        its 0 ... the file expects it).  No longer check for removed files.
        (store_word): Check for removed files here, and only add to the
        ordered tree if we have references left to this word.
        (ibex_write): First insert into the tree, to determine the
        wordcount to be saved in the output file, and then write that.
        (ibex_open): Remove some debug.

        * words.c (ibex_index_buffer): Always set 'unread', if it is a
        valid pointer (dont rely on caller to initialise it).

svn path=/trunk/; revision=2409

5 files changed, 82 insertions, 31 deletions

diff --git a/libibex/ChangeLog b/libibex/ChangeLog
index ef27870817..bc6d1d520a 100644
--- a/libibex/ChangeLog
+++ b/libibex/ChangeLog
@@ -1,3 +1,23 @@
+2000-04-12  NotZed  <NotZed@HelixCode.com>
+
+	* find.c (ibex_dump_all): Debug function to dump the whole index
+	to stdout.
+
+	* words.c (get_ibex_file): Use g_strdup(), not strdup().
+
+2000-04-11  NotZed  <NotZed@HelixCode.com>
+
+	* file.c (write_word): Always write out all words we have (even if
+	its 0 ... the file expects it).  No longer check for removed files.	
+	(store_word): Check for removed files here, and only add to the
+	ordered tree if we have references left to this word.
+	(ibex_write): First insert into the tree, to determine the
+	wordcount to be saved in the output file, and then write that.
+	(ibex_open): Remove some debug.
+
+	* words.c (ibex_index_buffer): Always set 'unread', if it is a
+	valid pointer (dont rely on caller to initialise it).
+
 2000-03-26  NotZed  <NotZed@HelixCode.com>
 
 	* lookup.c (main): Fixed call to ibex_open.
diff --git a/libibex/file.c b/libibex/file.c
index f011312cab..5c292e198f 100644
--- a/libibex/file.c
+++ b/libibex/file.c
@@ -81,7 +81,6 @@ ibex_open (char *file, int flags, int mode)
 
 	fd = open(file, flags, mode);
 	if (fd == -1) {
-		printf("open failed :(\n");
 		return NULL;
 	}
 
@@ -113,7 +112,6 @@ ibex_open (char *file, int flags, int mode)
 
 	f = fdopen(fd, modestr);
 	if (f == NULL) {
-		printf("fdopen failed, modestr = '%s'\n", modestr);
 		if (errno == 0)
 			errno = ENOMEM;
 		return NULL;
@@ -147,8 +145,9 @@ ibex_open (char *file, int flags, int mode)
 	for (i = 0; i < nfiles; i++) {
 		ibfs[i] = g_malloc (sizeof (ibex_file));
 		ibfs[i]->name = get_compressed_word (f, &lastword);
-		if (!ibfs[i]->name)
+		if (!ibfs[i]->name) {
 			goto errout;
+		}
 		ibfs[i]->index = 0;
 		g_tree_insert (ib->files, ibfs[i]->name, ibfs[i]);
 	}
@@ -158,16 +157,18 @@ ibex_open (char *file, int flags, int mode)
 	lastword = NULL;
 	for (i = 0; i < nwords; i++) {
 		word = get_compressed_word (f, &lastword);
-		if (!word)
+		if (!word) {
 			goto errout;
+		}
 
 		nrefs = read_number (f);
 		refs = g_ptr_array_new ();
 		g_ptr_array_set_size (refs, nrefs);
 		while (nrefs--) {
 			ref = read_number (f);
-			if (ref >= nfiles)
+			if (ref >= nfiles) {
 				goto errout;
+			}
 			refs->pdata[nrefs] = ibfs[ref];
 		}
 
@@ -230,43 +231,49 @@ write_file (gpointer key, gpointer value, gpointer data)
 	return FALSE;
 }
 
+/* scans for words which still exist in the index (after
+   index removals), and adds them to the ordered tree for
+   writing out in order */
 static void
 store_word (gpointer key, gpointer value, gpointer data)
 {
 	GTree *wtree = data;
+	GPtrArray *refs = value;
+	int i;
+	ibex_file *ibf;
 
-	g_tree_insert (wtree, key, value);
+	for (i = 0; i < refs->len; i++) {
+		ibf = g_ptr_array_index (refs, i);
+		if (ibf->index == -1) {
+			g_ptr_array_remove_index_fast (refs, i);
+			i--;
+		}
+	}
+
+	if (refs->len > 0) {
+		g_tree_insert (wtree, key, value);
+	}
 }
 
+/* writes a word out, in order */
 static gint
 write_word (gpointer key, gpointer value, gpointer data)
 {
 	char *word = key;
 	GPtrArray *refs = value;
 	struct ibex_write_data *iwd = data;
+	int i, prefix;
 	ibex_file *ibf;
-	int i, ind, prefix;
 
-	for (i = ind = 0; i < refs->len; i++) {
-		ibf = g_ptr_array_index (refs, i);
-		if (ibf->index == -1) {
-			g_ptr_array_remove_index_fast (refs, i);
-			i--;
-		} else
-			ind++;
-	}
-
-	if (ind != 0) {
-		prefix = get_prefix (iwd, word);
-		fprintf (iwd->f, "%c%s", prefix, word + prefix);
-		fputc (0, iwd->f);
+	prefix = get_prefix (iwd, word);
+	fprintf (iwd->f, "%c%s", prefix, word + prefix);
+	fputc (0, iwd->f);
 
-		write_number (iwd->f, ind);
+	write_number (iwd->f, refs->len);
 
-		for (i = 0; i < refs->len; i++) {
-			ibf = g_ptr_array_index (refs, i);
-			write_number (iwd->f, ibf->index);
-		}
+	for (i = 0; i < refs->len; i++) {
+		ibf = g_ptr_array_index (refs, i);
+		write_number (iwd->f, ibf->index);
 	}
 	return FALSE;
 }
@@ -310,11 +317,11 @@ ibex_write (ibex *ib)
 		goto lose;
 
 	iwd.lastname = NULL;
-	write_number (iwd.f, g_hash_table_size (ib->words));
-	if (ferror (iwd.f))
-		goto lose;
 	wtree = g_tree_new (strcmp);
 	g_hash_table_foreach (ib->words, store_word, wtree);
+	write_number (iwd.f, g_tree_nnodes(wtree));
+	if (ferror (iwd.f))
+		goto lose;
 	g_tree_traverse (wtree, write_word, G_IN_ORDER, &iwd);
 	g_tree_destroy (wtree);
 	if (ferror (iwd.f))
diff --git a/libibex/find.c b/libibex/find.c
index 91e611401c..e0b5d263c8 100644
--- a/libibex/find.c
+++ b/libibex/find.c
@@ -160,3 +160,23 @@ ibex_find_all (ibex *ib, GPtrArray *words)
 	g_tree_destroy (work);
 	return ret;
 }
+
+static void
+ibex_dump_foo(char *key, GPtrArray *refs, void *data)
+{
+	int i;
+
+	printf("%s: ", key);
+	for (i=0;i<refs->len;i++) {
+		ibex_file *ibf = g_ptr_array_index (refs, i);
+		printf("%c%s", ibf->index==-1?'-':' ', ibf->name);
+	}
+	printf("\n");
+}
+
+/* debug function to dump the tree, in key order */
+void
+ibex_dump_all (ibex *ib)
+{
+	g_hash_table_foreach(ib->words, ibex_dump_foo, 0);
+}
diff --git a/libibex/ibex.h b/libibex/ibex.h
index bb818e6d34..e24508aaf7 100644
--- a/libibex/ibex.h
+++ b/libibex/ibex.h
@@ -92,5 +92,8 @@ gboolean ibex_find_name (ibex *ib, char *name, char *word);
  */
 GPtrArray *ibex_find_all (ibex *ib, GPtrArray *words);
 
+/* Debug function, dumps the whole index tree, showing removed files as well */
+void ibex_dump_all (ibex *ib);
+
 #endif /* ! IBEX_H */
 
diff --git a/libibex/words.c b/libibex/words.c
index 40e0452403..7642b0a305 100644
--- a/libibex/words.c
+++ b/libibex/words.c
@@ -159,7 +159,7 @@ get_ibex_file (ibex *ib, char *name)
 	ibf = g_tree_lookup (ib->files, name);
 	if (!ibf) {
 		ibf = g_malloc (sizeof (ibex_file));
-		ibf->name = strdup (name);
+		ibf->name = g_strdup (name);
 		ibf->index = 0;
 		g_tree_insert (ib->files, ibf->name, ibf);
 		ib->dirty = TRUE;
@@ -213,6 +213,9 @@ ibex_index_buffer (ibex *ib, char *name, char *buffer,
 	ibex_file *ibf = get_ibex_file (ib, name);
 	int wordsiz, cat;
 
+	if (unread)
+		*unread = 0;
+
 	end = buffer + len;
 	wordsiz = 20;
 	word = g_malloc (wordsiz);
@@ -261,8 +264,6 @@ ibex_index_buffer (ibex *ib, char *name, char *buffer,
 		p = q;
 	}
 
-	if (unread)
-		*unread = 0;
 	g_free (word);
 	return 0;
 }