Improved the docs.

[rsync/rsync-patches.git] / filter.diff

After applying this patch and running configure, you MUST run this
command before "make":

    make proto

This patch adds the ability to merge rules into your excludes/includes
using a ". FILE" idiom.  If you specify a name without slashes, that
filename will be looked for in every subdirectory that rsync visits,
and the rules found in that subdirectory's file will affect that dir
and its subdirectories.

For example:

  rsync -av --exclude='. .excl' from/ to

The above will look for a file named ".excl" in every directory of the
hierarchy that rsync visits, and it will exclude (by default) names
based on the rules found therein.  If one of the .excl files contains
this:

  + *.c
  . .excl2
  . ./.excl3
  *.o

Then the file ".excl2" will also be read in the current dir, and all
subdirs of the current dir.  The file ".excl3" would just be read in
for the current dir because its name contained a slash.

..wayne..

--- exclude.c	22 May 2004 05:32:20 -0000	1.82
+++ exclude.c	22 May 2004 06:02:55 -0000
@@ -30,13 +30,56 @@ extern int verbose;
 extern int eol_nulls;
 extern int list_only;
 extern int recurse;
+extern int io_error;
+extern int sanitize_paths;
 
 extern char curr_dir[];
 
 struct exclude_list_struct exclude_list = { 0, 0, "" };
-struct exclude_list_struct local_exclude_list = { 0, 0, "per-dir .cvsignore " };
 struct exclude_list_struct server_exclude_list = { 0, 0, "server " };
 char *exclude_path_prefix = NULL;
+struct exclude_struct **mergelist_parents;
+int mergelist_cnt = 0;
+int mergelist_size = 0;
+
+struct mergelist_save_struct {
+    struct exclude_list_struct *array;
+    int count;
+};
+
+static char dirbuf[MAXPATHLEN];
+static unsigned int dirbuf_offset = 0;
+
+/* Each exclude_list_struct describes a singly-linked list by keeping track
+ * of both the head and tail pointers.  The list is slightly unusual in that
+ * a parent-dir's content can be appended to the end of the local list in a
+ * special way:  the last item in the local list has its "next" pointer set
+ * to point to the inherited list, but the local list's tail pointer points
+ * at the end of the local list.  Thus, if the local list is empty, the head
+ * will be pointing at the inherited content but the tail will be NULL.  To
+ * help you visualize this, here are the possible list arrangements:
+ *
+ * Completely Empty                     Local Content Only
+ * ==================================   ====================================
+ * head -> NULL                         head -> Local1 -> Local2 -> NULL
+ * tail -> NULL                         tail -------------^
+ *
+ * Inherited Content Only               Both Local and Inherited Content
+ * ==================================   ====================================
+ * head -> Parent1 -> Parent2 -> NULL   head -> L1 -> L2 -> P1 -> P2 -> NULL
+ * tail -> NULL                         tail ---------^
+ *
+ * This means that anyone wanting to traverse the whole list to USE it just
+ * needs to start at the head and use the "next" pointers until it goes
+ * NULL.  To add new local content, we insert the item after the tail item
+ * and update the tail (obviously, if "tail" was NULL, we insert it at the
+ * head).  To clear the local list, WE MUST NOT FREE THE INHERITED CONTENT
+ * because it is shared between the current list and our parent list(s).
+ * The easiest way to handle this is to simply truncate the list after the
+ * tail item and then free the local list from the head.  When inheriting
+ * the list for a new local dir, we just save off the exclude_list_struct
+ * values (so we can pop back to them later) and set the tail to NULL.
+ */
 
 /** Build an exclude structure given an exclude pattern. */
 static void make_exclude(struct exclude_list_struct *listp, const char *pat,
@@ -46,6 +89,24 @@ static void make_exclude(struct exclude_
 	const char *cp;
 	unsigned int ex_len;
 
+	if (mflags & MATCHFLG_MERGE_FILE) {
+		int i;
+		/* If the local include file was already mentioned, don't
+		 * add it again. */
+		for (i = 0; i < mergelist_cnt; i++) {
+			struct exclude_struct *ex = mergelist_parents[i];
+			if (strlen(ex->pattern) == pat_len
+			    && memcmp(ex->pattern, pat, pat_len) == 0)
+				return;
+		}
+		if ((pat_len == 10 || (pat_len > 10 && pat[pat_len-11] == '/'))
+		    && strncmp(pat+pat_len-10, ".cvsignore", 10) == 0) {
+			mflags |= MATCHFLG_CVSIGNORE;
+			mflags &= ~MATCHFLG_INCLUDE;
+		} else
+			mflags &= ~MATCHFLG_CVSIGNORE;
+	}
+
 	ret = new(struct exclude_struct);
 	if (!ret)
 		out_of_memory("make_exclude");
@@ -81,14 +142,36 @@ static void make_exclude(struct exclude_
 		mflags |= MATCHFLG_DIRECTORY;
 	}
 
-	for (cp = ret->pattern; (cp = strchr(cp, '/')) != NULL; cp++)
-		ret->slash_cnt++;
+	if (mflags & MATCHFLG_MERGE_FILE) {
+		struct exclude_list_struct *lp
+		    = new_array(struct exclude_list_struct, 1);
+		if (!lp)
+			out_of_memory("make_exclude");
+		lp->head = lp->tail = NULL;
+		if (asprintf(&lp->debug_type, "per-dir %s ", ret->pattern) < 0)
+			out_of_memory("make_exclude");
+		ret->u.mergelist = lp;
+		if (mergelist_cnt == mergelist_size) {
+		    mergelist_size += 5;
+		    mergelist_parents = realloc_array(mergelist_parents,
+						      struct exclude_struct *,
+						      mergelist_size);
+		    if (!mergelist_parents)
+			    out_of_memory("make_exclude");
+		}
+		mergelist_parents[mergelist_cnt++] = ret;
+	} else {
+		for (cp = ret->pattern; (cp = strchr(cp, '/')) != NULL; cp++)
+			ret->u.slash_cnt++;
+	}
 
 	ret->match_flags = mflags;
 
-	if (!listp->tail)
+	if (!listp->tail) {
+		ret->next = listp->head;
 		listp->head = listp->tail = ret;
-	else {
+	} else {
+		ret->next = listp->tail->next;
 		listp->tail->next = ret;
 		listp->tail = ret;
 	}
@@ -96,22 +179,116 @@ static void make_exclude(struct exclude_
 
 static void free_exclude(struct exclude_struct *ex)
 {
+	if (ex->match_flags & MATCHFLG_MERGE_FILE) {
+		free(ex->u.mergelist->debug_type);
+		free(ex->u.mergelist);
+	}
 	free(ex->pattern);
 	free(ex);
 }
 
-void clear_exclude_list(struct exclude_list_struct *listp)
+static void clear_exclude_list(struct exclude_list_struct *listp)
 {
-	struct exclude_struct *ent, *next;
-
-	for (ent = listp->head; ent; ent = next) {
-		next = ent->next;
-		free_exclude(ent);
+	if (listp->tail) {
+		struct exclude_struct *ent, *next;
+		/* Truncate any inherited items from the local list. */
+		listp->tail->next = NULL;
+		/* Now free everything that is left. */
+		for (ent = listp->head; ent; ent = next) {
+			next = ent->next;
+			free_exclude(ent);
+		}
 	}
 
 	listp->head = listp->tail = NULL;
 }
 
+void *push_local_excludes(char *fname, unsigned int offset)
+{
+	struct mergelist_save_struct *push;
+	struct exclude_list_struct *ap;
+	int i;
+
+	/* Make it easy to construct the full path for a merge-file that was
+	 * specified with a relative path by saving off the current dir. */
+	memcpy(dirbuf, fname, offset);
+	dirbuf_offset = offset;
+
+	if (!(push = new_array(struct mergelist_save_struct, 1)))
+		out_of_memory("push_local_excludes");
+
+	push->count = mergelist_cnt;
+	push->array = new_array(struct exclude_list_struct, mergelist_cnt);
+	if (!push->array)
+		out_of_memory("push_local_excludes");
+
+	for (i = 0, ap = push->array; i < mergelist_cnt; i++) {
+		memcpy(ap++, mergelist_parents[i]->u.mergelist,
+		       sizeof (struct exclude_list_struct));
+	}
+
+	/* Note: add_exclude_file() might increase mergelist_cnt, so keep
+	 * this loop separate from the above loop. */
+	for (i = 0; i < mergelist_cnt; i++) {
+		struct exclude_struct *ex = mergelist_parents[i];
+		struct exclude_list_struct *lp = ex->u.mergelist;
+		int flags;
+
+		if (verbose > 2) {
+			rprintf(FINFO, "[%s] pushing %sexclude list\n",
+				who_am_i(), lp->debug_type);
+		}
+
+		if (ex->match_flags & MATCHFLG_CVSIGNORE) {
+			lp->head = NULL; /* CVS doesn't inherit rules. */
+			flags = XFLG_WORD_SPLIT | XFLG_WORDS_ONLY;
+		} else {
+			flags = ex->match_flags & MATCHFLG_INCLUDE
+			    ? XFLG_DEF_INCLUDE : 0;
+		}
+		lp->tail = NULL; /* Switch any local rules to inherited. */
+		if (strlcpy(dirbuf +  dirbuf_offset, ex->pattern,
+		    MAXPATHLEN - dirbuf_offset) < MAXPATHLEN - dirbuf_offset)
+			add_exclude_file(lp, dirbuf, flags);
+		else {
+			io_error |= IOERR_GENERAL;
+			rprintf(FINFO,
+			    "cannot add local excludes in long-named directory %s\n",
+			    full_fname(dirbuf));
+		}
+	}
+
+	return (void*)push;
+}
+
+void pop_local_excludes(void *mem)
+{
+	struct mergelist_save_struct *pop = (struct mergelist_save_struct*)mem;
+	struct exclude_list_struct *ap;
+	int i;
+
+	for (i = mergelist_cnt; i-- > 0; ) {
+		struct exclude_struct *ex = mergelist_parents[i];
+		struct exclude_list_struct *lp = ex->u.mergelist;
+
+		if (verbose > 2) {
+			rprintf(FINFO, "[%s] popping %sexclude list\n",
+				who_am_i(), lp->debug_type);
+		}
+
+		clear_exclude_list(lp);
+	}
+
+	mergelist_cnt = pop->count;
+	for (i = 0, ap = pop->array; i < mergelist_cnt; i++) {
+		memcpy(mergelist_parents[i]->u.mergelist, ap++,
+		       sizeof (struct exclude_list_struct));
+	}
+
+	free(pop->array);
+	free(pop);
+}
+
 static int check_one_exclude(char *name, struct exclude_struct *ex,
                              int name_is_dir)
 {
@@ -122,7 +299,7 @@ static int check_one_exclude(char *name,
 	/* If the pattern does not have any slashes AND it does not have
 	 * a "**" (which could match a slash), then we just match the
 	 * name portion of the path. */
-	if (!ex->slash_cnt && !(ex->match_flags & MATCHFLG_WILD2)) {
+	if (!ex->u.slash_cnt && !(ex->match_flags & MATCHFLG_WILD2)) {
 		if ((p = strrchr(name,'/')) != NULL)
 			name = p+1;
 	}
@@ -148,9 +325,9 @@ static int check_one_exclude(char *name,
 	if (ex->match_flags & MATCHFLG_WILD) {
 		/* A non-anchored match with an infix slash and no "**"
 		 * needs to match the last slash_cnt+1 name elements. */
-		if (!match_start && ex->slash_cnt
+		if (!match_start && ex->u.slash_cnt
 		    && !(ex->match_flags & MATCHFLG_WILD2)) {
-			int cnt = ex->slash_cnt + 1;
+			int cnt = ex->u.slash_cnt + 1;
 			for (p = name + strlen(name) - 1; p >= name; p--) {
 				if (*p == '/' && !--cnt)
 					break;
@@ -221,6 +398,13 @@ int check_exclude(struct exclude_list_st
 	struct exclude_struct *ent;
 
 	for (ent = listp->head; ent; ent = ent->next) {
+		if (ent->match_flags & MATCHFLG_MERGE_FILE) {
+			int rc = check_exclude(ent->u.mergelist, name,
+					       name_is_dir);
+			if (rc)
+				return rc;
+			continue;
+		}
 		if (check_one_exclude(name, ent, name_is_dir)) {
 			report_exclude_result(name, ent, name_is_dir,
 					      listp->debug_type);
@@ -253,11 +437,16 @@ static const char *get_exclude_tok(const
 		p = (const char *)s;
 	}
 
-	/* Is this a '+' or '-' followed by a space (not whitespace)? */
+	/* Is this a +/-/. followed by a space (not whitespace)? */
 	if (!(xflags & XFLG_WORDS_ONLY)
-	    && (*s == '-' || *s == '+') && s[1] == ' ') {
+	    && (*s == '-' || *s == '+' || *s == '.') && s[1] == ' ') {
 		if (*s == '+')
 			mflags |= MATCHFLG_INCLUDE;
+		else if (*s == '.') {
+			mflags |= MATCHFLG_MERGE_FILE;
+			if (xflags & XFLG_DEF_INCLUDE)
+				mflags |= MATCHFLG_INCLUDE;
+		}
 		s += 2;
 	} else if (xflags & XFLG_DEF_INCLUDE)
 		mflags |= MATCHFLG_INCLUDE;
@@ -306,11 +495,42 @@ void add_exclude(struct exclude_list_str
 			continue;
 		}
 
+		if (mflags & MATCHFLG_MERGE_FILE) {
+			char name[MAXPATHLEN];
+			if (pat_len >= sizeof name) {
+				rprintf(FERROR,
+					"merge filename too long: %s\n", cp);
+				continue;
+			}
+			strlcpy(name, cp, pat_len+1);
+			if (strchr(name, '/') != NULL) {
+				if (sanitize_paths)
+					sanitize_path(name, curr_dir);
+				if (*name == '/')
+					cp = name;
+				else {
+					if (strlcpy(dirbuf + dirbuf_offset,
+					    name, MAXPATHLEN - dirbuf_offset)
+					    >= MAXPATHLEN - dirbuf_offset) {
+						rprintf(FERROR,
+						    "merge filename too long: %s...\n",
+						    dirbuf);
+						continue;
+					}
+					cp = dirbuf;
+				}
+				add_exclude_file(listp, cp,
+				    xflags | XFLG_FATAL_ERRORS);
+				continue;
+			}
+		}
+
 		make_exclude(listp, cp, pat_len, mflags);
 
 		if (verbose > 2) {
-			rprintf(FINFO, "[%s] add_exclude(%.*s, %s%sclude)\n",
+			rprintf(FINFO, "[%s] add_exclude(%.*s, %s%s%sclude)\n",
 				who_am_i(), (int)pat_len, cp, listp->debug_type,
+				mflags & MATCHFLG_MERGE_FILE ? "FILE " : "",
 				mflags & MATCHFLG_INCLUDE ? "in" : "ex");
 		}
 	}
@@ -402,7 +622,11 @@ void send_exclude_list(int f)
 		if (ent->match_flags & MATCHFLG_INCLUDE) {
 			write_int(f, l + 2);
 			write_buf(f, "+ ", 2);
-		} else if ((*p == '-' || *p == '+') && p[1] == ' ') {
+		} else if (ent->match_flags & MATCHFLG_MERGE_FILE) {
+			write_int(f, l + 2);
+			write_buf(f, ". ", 2);
+		} else if ((*p == '-' || *p == '+' || *p == '.')
+		    && p[1] == ' ') {
 			write_int(f, l + 2);
 			write_buf(f, "- ", 2);
 		} else
@@ -443,6 +667,7 @@ void add_cvs_excludes(void)
 	char fname[MAXPATHLEN];
 	char *p;
 
+	add_exclude(&exclude_list, ". .cvsignore", 0);
 	add_exclude(&exclude_list, default_cvsignore,
 		    XFLG_WORD_SPLIT | XFLG_WORDS_ONLY);
 
--- flist.c	21 May 2004 23:22:14 -0000	1.225
+++ flist.c	22 May 2004 06:02:55 -0000
@@ -39,8 +39,6 @@ extern int module_id;
 extern int ignore_errors;
 extern int numeric_ids;
 
-extern int cvs_exclude;
-
 extern int recurse;
 extern char curr_dir[MAXPATHLEN];
 extern char *files_from;
@@ -65,7 +63,6 @@ extern int write_batch;
 
 extern struct exclude_list_struct exclude_list;
 extern struct exclude_list_struct server_exclude_list;
-extern struct exclude_list_struct local_exclude_list;
 
 int io_error;
 
@@ -210,8 +207,6 @@ int link_stat(const char *path, STRUCT_S
  */
 static int check_exclude_file(char *fname, int is_dir, int exclude_level)
 {
-	int rc;
-
 #if 0 /* This currently never happens, so avoid a useless compare. */
 	if (exclude_level == NO_EXCLUDES)
 		return 0;
@@ -233,10 +228,7 @@ static int check_exclude_file(char *fnam
 	if (exclude_level != ALL_EXCLUDES)
 		return 0;
 	if (exclude_list.head
-	    && (rc = check_exclude(&exclude_list, fname, is_dir)) != 0)
-		return rc < 0;
-	if (local_exclude_list.head
-	    && check_exclude(&local_exclude_list, fname, is_dir) < 0)
+	    && check_exclude(&exclude_list, fname, is_dir) < 0)
 		return 1;
 	return 0;
 }
@@ -942,15 +934,7 @@ void send_file_name(int f, struct file_l
 
 	if (recursive && S_ISDIR(file->mode)
 	    && !(file->flags & FLAG_MOUNT_POINT)) {
-		struct exclude_list_struct last_list = local_exclude_list;
-		local_exclude_list.head = local_exclude_list.tail = NULL;
 		send_directory(f, flist, f_name_to(file, fbuf));
-		if (verbose > 2) {
-			rprintf(FINFO, "[%s] popping %sexclude list\n",
-				who_am_i(), local_exclude_list.debug_type);
-		}
-		clear_exclude_list(&local_exclude_list);
-		local_exclude_list = last_list;
 	}
 }
 
@@ -961,6 +945,7 @@ static void send_directory(int f, struct
 	struct dirent *di;
 	char fname[MAXPATHLEN];
 	unsigned int offset;
+	void *save_excludes;
 	char *p;
 
 	d = opendir(dir);
@@ -984,18 +969,7 @@ static void send_directory(int f, struct
 		offset++;
 	}
 
-	if (cvs_exclude) {
-		if (strlcpy(p, ".cvsignore", MAXPATHLEN - offset)
-		    < MAXPATHLEN - offset) {
-			add_exclude_file(&local_exclude_list, fname,
-					 XFLG_WORD_SPLIT | XFLG_WORDS_ONLY);
-		} else {
-			io_error |= IOERR_GENERAL;
-			rprintf(FINFO,
-				"cannot cvs-exclude in long-named directory %s\n",
-				full_fname(fname));
-		}
-	}
+	save_excludes = push_local_excludes(fname, offset);
 
 	for (errno = 0, di = readdir(d); di; errno = 0, di = readdir(d)) {
 		char *dname = d_name(di);
@@ -1016,6 +990,8 @@ static void send_directory(int f, struct
 		rsyserr(FERROR, errno, "readdir(%s)", dir);
 	}
 
+	pop_local_excludes(save_excludes);
+
 	closedir(d);
 }
 
--- rsync.h	16 May 2004 07:28:24 -0000	1.204
+++ rsync.h	22 May 2004 06:02:56 -0000
@@ -496,11 +496,16 @@ struct map_struct {
 #define MATCHFLG_INCLUDE	(1<<4) /* this is an include, not an exclude */
 #define MATCHFLG_DIRECTORY	(1<<5) /* this matches only directories */
 #define MATCHFLG_CLEAR_LIST 	(1<<6) /* this item is the "!" token */
+#define MATCHFLG_MERGE_FILE	(1<<7) /* specifies a file to merge */
+#define MATCHFLG_CVSIGNORE	(1<<8) /* parse this as a .cvsignore file */
 struct exclude_struct {
 	struct exclude_struct *next;
 	char *pattern;
 	unsigned int match_flags;
-	int slash_cnt;
+	union {
+		int slash_cnt;
+		struct exclude_list_struct *mergelist;
+	} u;
 };
 
 struct exclude_list_struct {
--- rsync.yo	21 May 2004 09:44:32 -0000	1.170
+++ rsync.yo	22 May 2004 06:02:57 -0000
@@ -1090,6 +1090,74 @@ itemize(
   it would be excluded by the "*")
 )
 
+manpagesection(MERGING EXCLUDE FILES)
+
+You can merge whole files into an exclude file using a rule that starts
+with a ". " (a dot followed by a space) and has a filename in place of the
+pattern.  There are two types of merge rules, single-instance and
+per-directory:
+
+itemize(
+  it() If the filename has no slashes in it, it is a per-directory merge;
+  rsync scans every directory that it traverses for the named file, merging
+  its contents (when it exists) at the start of this per-directory sub-list
+  (subdirectories inherit the contents of their parent directories by
+  default, and each subdirectory's rules have precedence over the parent
+  directory's rules).  When a per-directory merge file adds another
+  per-directory merge file, it is only active in the subtree of the current
+  directory.
+
+  it() If a filename has a slash in it, it is a single-instance merge; the
+  named file's contents will be merged into the current exclude file,
+  replacing the merge rule.  Thus, you should use the name ./foo instead of
+  foo if you don't want to scan for "foo" in all the subdirectories of the
+  current directory.
+)
+
+Note also that you can eliminate all the inherited rules for the current
+per-directory ruleset by putting the list-clearing token (!) in the file.
+This only clears the rules for the current per-directory sub-list (up
+through the ! token) and only for the current directory and its
+subdirectories.
+
+Here's an example exclude file (which you'd specify via the normal
+--exclude-from option):
+
+verb(
+    . /home/user/.global_excludes
+    - *.gz
+    . .excl
+    + *.[ch]
+    - *.o
+)
+
+This will merge the contents of the /home/user/.global_excludes file at the
+start of the list and also turns the ".excl" filename into a per-directory
+exclude file whose local contents will be merged into the list in place of
+the .excl line.
+
+Additionally, you can affect where the --cvs-exclude (-C) option's
+inclusion of a per-directory .cvsignore file gets placed into your rules by
+adding an explicit merge rule for ".cvsignore".  For instance, specifying
+this:
+
+verb(
+  rsync -avC --exclude='. .cvsignore' --exclude-from=foo a/ b
+)
+
+will merge all the per-directory .cvsignore rules at the start of your list
+rather than at the end.  This allows their dir-specific rules to supersede
+your rules instead of being subservient to them.  (The global rules taken
+from the $HOME/.cvsignore file and from $CVSIGNORE are not affected by
+this.)
+
+Note also that the parsing of any merge-file named ".cvsignore" is always
+done in a CVS-compatible manner, even if -C wasn't specified (i.e. the
+rules are always exclude rules (even when specified by an include option);
+they are split on whitespace; no special prefixes, list-clearing tokens, or
+comment characters are honored; and, for per-directory files,
+subdirectories don't inherit the parent directory's rules).
+
 manpagesection(BATCH MODE)
 
 bf(Note:) Batch mode should be considered experimental in this version