dump: support gzipped and atomic output

The main goal is to support gzipped output for future internal
calls (e.g. from notmuch-new) to notmuch_database_dump.

The additional dependency is not very heavy since xapian already pulls
in zlib.

We want the dump to be "atomic", in the sense that after running the
dump file is either present and complete, or not present.  This avoids
certain classes of mishaps involving overwriting a good backup with a
bad or partial one.

1 files changed, 82 insertions, 19 deletions

diff --git a/notmuch-dump.c b/notmuch-dump.c
index 21702d79..2849eaba 100644
--- a/notmuch-dump.c
+++ b/notmuch-dump.c
@@ -21,9 +21,11 @@
 #include "notmuch-client.h"
 #include "hex-escape.h"
 #include "string-util.h"
+#include <zlib.h>
+
 
 static int
-database_dump_file (notmuch_database_t *notmuch, FILE *output,
+database_dump_file (notmuch_database_t *notmuch, gzFile output,
 		    const char *query_str, int output_format)
 {
     notmuch_query_t *query;
@@ -69,7 +71,7 @@ database_dump_file (notmuch_database_t *notmuch, FILE *output,
 	}
 
 	if (output_format == DUMP_FORMAT_SUP) {
-	    fprintf (output, "%s (", message_id);
+	    gzprintf (output, "%s (", message_id);
 	}
 
 	for (tags = notmuch_message_get_tags (message);
@@ -78,12 +80,12 @@ database_dump_file (notmuch_database_t *notmuch, FILE *output,
 	    const char *tag_str = notmuch_tags_get (tags);
 
 	    if (! first)
-		fputs (" ", output);
+		gzputs (output, " ");
 
 	    first = 0;
 
 	    if (output_format == DUMP_FORMAT_SUP) {
-		fputs (tag_str, output);
+		gzputs (output, tag_str);
 	    } else {
 		if (hex_encode (notmuch, tag_str,
 				&buffer, &buffer_size) != HEX_SUCCESS) {
@@ -91,12 +93,12 @@ database_dump_file (notmuch_database_t *notmuch, FILE *output,
 			     tag_str);
 		    return EXIT_FAILURE;
 		}
-		fprintf (output, "+%s", buffer);
+		gzprintf (output, "+%s", buffer);
 	    }
 	}
 
 	if (output_format == DUMP_FORMAT_SUP) {
-	    fputs (")\n", output);
+	    gzputs (output, ")\n");
 	} else {
 	    if (make_boolean_term (notmuch, "id", message_id,
 				   &buffer, &buffer_size)) {
@@ -104,7 +106,7 @@ database_dump_file (notmuch_database_t *notmuch, FILE *output,
 			     message_id, strerror (errno));
 		    return EXIT_FAILURE;
 	    }
-	    fprintf (output, " -- %s\n", buffer);
+	    gzprintf (output, " -- %s\n", buffer);
 	}
 
 	notmuch_message_destroy (message);
@@ -121,24 +123,83 @@ database_dump_file (notmuch_database_t *notmuch, FILE *output,
 int
 notmuch_database_dump (notmuch_database_t *notmuch,
 		       const char *output_file_name,
-		       const char *query_str, dump_format_t output_format)
+		       const char *query_str,
+		       dump_format_t output_format,
+		       notmuch_bool_t gzip_output)
 {
-    FILE *output = stdout;
-    int ret;
+    gzFile output = NULL;
+    const char *mode = gzip_output ? "w9" : "wT";
+    const char *name_for_error = output_file_name ? output_file_name : "stdout";
+
+    char *tempname = NULL;
+    int outfd = -1;
+
+    int ret = -1;
 
     if (output_file_name) {
-	output = fopen (output_file_name, "w");
-	if (output == NULL) {
-	    fprintf (stderr, "Error opening %s for writing: %s\n",
-		     output_file_name, strerror (errno));
-	    return EXIT_FAILURE;
-	}
+	tempname = talloc_asprintf (notmuch, "%s.XXXXXX", output_file_name);
+	outfd = mkstemp (tempname);
+    } else {
+	outfd = dup (STDOUT_FILENO);
+    }
+
+    if (outfd < 0) {
+	fprintf (stderr, "Bad output file %s\n", name_for_error);
+	goto DONE;
+    }
+
+    output = gzdopen (outfd, mode);
+
+    if (output == NULL) {
+	fprintf (stderr, "Error opening %s for (gzip) writing: %s\n",
+		 name_for_error, strerror (errno));
+	if (close (outfd))
+	    fprintf (stderr, "Error closing %s during shutdown: %s\n",
+		 name_for_error, strerror (errno));
+	goto DONE;
     }
 
     ret = database_dump_file (notmuch, output, query_str, output_format);
+    if (ret) goto DONE;
+
+    ret = gzflush (output, Z_FINISH);
+    if (ret) {
+	fprintf (stderr, "Error flushing output: %s\n", gzerror (output, NULL));
+	goto DONE;
+    }
+
+    if (output_file_name) {
+	ret = fdatasync (outfd);
+	if (ret) {
+	    fprintf (stderr, "Error syncing %s to disk: %s\n",
+		     name_for_error, strerror (errno));
+	    goto DONE;
+	}
+    }
+
+    if (gzclose_w (output) != Z_OK) {
+	fprintf (stderr, "Error closing %s: %s\n", name_for_error,
+		 gzerror (output, NULL));
+	ret = EXIT_FAILURE;
+	output = NULL;
+	goto DONE;
+    }
+
+    if (output_file_name) {
+	ret = rename (tempname, output_file_name);
+	if (ret) {
+	    fprintf (stderr, "Error renaming %s to %s: %s\n",
+		     tempname, output_file_name, strerror (errno));
+	    goto DONE;
+	}
+
+    }
+ DONE:
+    if (ret != EXIT_SUCCESS && output)
+	(void) gzclose_w (output);
 
-    if (output != stdout)
-	fclose (output);
+    if (ret != EXIT_SUCCESS && output_file_name)
+	(void) unlink (tempname);
 
     return ret;
 }
@@ -158,6 +219,7 @@ notmuch_dump_command (notmuch_config_t *config, int argc, char *argv[])
     int opt_index;
 
     int output_format = DUMP_FORMAT_BATCH_TAG;
+    notmuch_bool_t gzip_output = 0;
 
     notmuch_opt_desc_t options[] = {
 	{ NOTMUCH_OPT_KEYWORD, &output_format, "format", 'f',
@@ -165,6 +227,7 @@ notmuch_dump_command (notmuch_config_t *config, int argc, char *argv[])
 				  { "batch-tag", DUMP_FORMAT_BATCH_TAG },
 				  { 0, 0 } } },
 	{ NOTMUCH_OPT_STRING, &output_file_name, "output", 'o', 0  },
+	{ NOTMUCH_OPT_BOOLEAN, &gzip_output, "gzip", 'z', 0 },
 	{ 0, 0, 0, 0, 0 }
     };
 
@@ -181,7 +244,7 @@ notmuch_dump_command (notmuch_config_t *config, int argc, char *argv[])
     }
 
     ret = notmuch_database_dump (notmuch, output_file_name, query_str,
-				 output_format);
+				 output_format, gzip_output);
 
     notmuch_database_destroy (notmuch);