notmuch/test/random-corpus.c
Austin Clements d08c714b6a dump: Disallow \n in message IDs
When we switch to using regular Xapian queries in the dump format, \n
will cause problems, so we disallow it.  Specially, while Xapian can
quote and parse queries containing \n without difficultly, quoted
queries containing \n still span multiple lines, which breaks the
line-orientedness of the dump format.  Strictly speaking, we could
still round-trip these, but it would significantly complicate restore
as well as scripts that deal with tag dumps.  This complexity would
come at absolutely no benefit: because of the RFC 2822 unfolding
rules, no amount of standards negligence can produce a message with a
message ID containing a line break (not even Outlook can do it!).

Hence, we simply disallow it.
2013-01-06 22:40:01 -04:00

211 lines
5.2 KiB
C

/*
* Generate a random corpus of stub messages.
*
* Initial use case is testing dump and restore, so we only have
* message-ids and tags.
*
* Generated message-id's and tags are intentionally nasty.
*
* Copyright (c) 2012 David Bremner
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see http://www.gnu.org/licenses/ .
*
* Author: David Bremner <david@tethera.net>
*/
#include <stdlib.h>
#include <assert.h>
#include <talloc.h>
#include <string.h>
#include <glib.h>
#include <math.h>
#include "notmuch-client.h"
#include "command-line-arguments.h"
#include "database-test.h"
/* Current largest Unicode value defined. Note that most of these will
* be printed as boxes in most fonts.
*/
#define GLYPH_MAX 0x10FFFE
typedef struct {
int weight;
int start;
int stop;
} char_class_t;
/*
* Choose about half ascii as test characters, as ascii
* punctation and whitespace is the main cause of problems for
* the (old) restore parser.
*
* We then favour code points with 2 byte encodings. Note that
* code points 0xD800-0xDFFF are forbidden in UTF-8.
*/
static const
char_class_t char_class[] = { { 0.50 * GLYPH_MAX, 0x0001, 0x007f },
{ 0.75 * GLYPH_MAX, 0x0080, 0x07ff },
{ 0.88 * GLYPH_MAX, 0x0800, 0xd7ff },
{ 0.90 * GLYPH_MAX, 0xE000, 0xffff },
{ GLYPH_MAX, 0x10000, GLYPH_MAX } };
static gunichar
random_unichar ()
{
int i;
int class = random () % GLYPH_MAX;
int size;
for (i = 0; char_class[i].weight < class; i++) /* nothing */;
size = char_class[i].stop - char_class[i].start + 1;
return char_class[i].start + (random () % size);
}
static char *
random_utf8_string (void *ctx, size_t char_count)
{
size_t offset = 0;
size_t i;
gchar *buf = NULL;
size_t buf_size = 0;
for (i = 0; i < char_count; i++) {
gunichar randomchar;
size_t written;
/* 6 for one glyph, one for null, one for luck */
while (buf_size <= offset + 8) {
buf_size = 2 * buf_size + 8;
buf = talloc_realloc (ctx, buf, gchar, buf_size);
}
do {
randomchar = random_unichar ();
} while (randomchar == '\n');
written = g_unichar_to_utf8 (randomchar, buf + offset);
if (written <= 0) {
fprintf (stderr, "error converting to utf8\n");
exit (1);
}
offset += written;
}
buf[offset] = 0;
return buf;
}
int
main (int argc, char **argv)
{
void *ctx = talloc_new (NULL);
char *config_path = NULL;
notmuch_config_t *config;
notmuch_database_t *notmuch;
int num_messages = 500;
int max_tags = 10;
// leave room for UTF-8 encoding.
int tag_len = NOTMUCH_TAG_MAX / 6;
// NOTMUCH_MESSAGE_ID_MAX is not exported, so we make a
// conservative guess.
int message_id_len = (NOTMUCH_TAG_MAX - 20) / 6;
int seed = 734569;
notmuch_opt_desc_t options[] = {
{ NOTMUCH_OPT_STRING, &config_path, "config-path", 'c', 0 },
{ NOTMUCH_OPT_INT, &num_messages, "num-messages", 'n', 0 },
{ NOTMUCH_OPT_INT, &max_tags, "max-tags", 'm', 0 },
{ NOTMUCH_OPT_INT, &message_id_len, "message-id-len", 'M', 0 },
{ NOTMUCH_OPT_INT, &tag_len, "tag-len", 't', 0 },
{ NOTMUCH_OPT_INT, &seed, "seed", 's', 0 },
{ 0, 0, 0, 0, 0 }
};
int opt_index = parse_arguments (argc, argv, options, 1);
if (opt_index < 0)
exit (1);
if (message_id_len < 1) {
fprintf (stderr, "message id's must be least length 1\n");
exit (1);
}
if (config_path == NULL) {
fprintf (stderr, "configuration path must be specified");
exit (1);
}
config = notmuch_config_open (ctx, config_path, NULL);
if (config == NULL)
return 1;
if (notmuch_database_open (notmuch_config_get_database_path (config),
NOTMUCH_DATABASE_MODE_READ_WRITE, &notmuch))
return 1;
srandom (seed);
int count;
for (count = 0; count < num_messages; count++) {
int j;
/* explicitly allow zero tags */
int num_tags = random () % (max_tags + 1);
/* message ids should be non-empty */
int this_mid_len = (random () % message_id_len) + 1;
const char **tag_list;
char *mid;
notmuch_status_t status;
do {
mid = random_utf8_string (ctx, this_mid_len);
tag_list = talloc_realloc (ctx, NULL, const char *, num_tags + 1);
for (j = 0; j < num_tags; j++) {
int this_tag_len = random () % tag_len + 1;
tag_list[j] = random_utf8_string (ctx, this_tag_len);
}
tag_list[j] = NULL;
status = notmuch_database_add_stub_message (notmuch, mid, tag_list);
} while (status == NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID);
if (status != NOTMUCH_STATUS_SUCCESS) {
fprintf (stderr, "error %d adding message", status);
exit (status);
}
}
notmuch_database_destroy (notmuch);
talloc_free (ctx);
return 0;
}