Files
cproof/tests/bench/bench_export_import.c
Jabber Developer 3f36c303c2
All checks were successful
CI Code / Check spelling (push) Successful in 21s
CI Code / Check coding style (push) Successful in 31s
CI Code / Code Coverage (push) Successful in 2m21s
CI Code / Linux (ubuntu) (push) Successful in 4m30s
CI Code / Linux (debian) (push) Successful in 6m43s
CI Code / Linux (arch) (push) Successful in 10m8s
feat(history): flat-file backend with bidirectional SQLite migration
A flat-file alternative to the SQLite chatlog backend with runtime
switching, full migration tooling, integrity verification, and a
synthetic load harness. SQLite remains the default; both backends share
one dispatch layer (db_backend_t vtable) so callers don't change.

Storage layout
- Per-contact append-only `flatlog/<account>/<contact>/history.log`
  under XDG_DATA_HOME, one line per message
- Single-line file header with embedded format-version marker
  (FLATFILE_FORMAT_VERSION); reader warns on missing or mismatched
  marker, writer and checker stay in sync via preprocessor
  stringification
- Deterministic key=value metadata (`id`, `aid`, `corrects`, `to`,
  `to_res`, `read`) plus escaped body \u2014 `\|`, `\]`, `\\`, `\n`, `\r`
  literals prevent log injection
- Sparse byte-offset index (FF_INDEX_STEP=500) per contact for
  O(log n) time-range lookups; rebuilt on inode / size / mtime
  change, extended in-place when the file just grew
- Per-contact GHashTable caches for archive_id presence and
  stanza_id \u2192 from_jid mapping (O(1) MAM dedup, O(1) LMC sender
  validation)

Hardening
- Path-traversal protection: JID directory name normalisation
  (`@` \u2192 `_at_`, slashes and `..` rejected at construction); every
  per-contact path is anchored under the account's flatlog/
  directory and validated before open
- Symlink-attack protection: every fopen / open uses O_NOFOLLOW; on
  ELOOP the operation aborts with an error rather than following
- Filesystem permissions: log files created with mode 0600,
  directories with mode 0700; both enforced at creation, verified
  on each open and reported on drift by `/history verify`
- Atomic crash-safe export: write to a temp file via mkstemp (mode
  0600, random suffix, no name collisions between concurrent
  exports), fsync, then rename \u2014 partial state never replaces the
  live file
- Concurrency: advisory flock(LOCK_EX) held for the duration of
  every write, including append from live messages and full rewrite
  from export, so two profanity processes can't interleave bytes
  on the same log
- DoS / abuse guards:
    * FF_MAX_LINE_LEN = 10 MB \u2014 lines longer than this are rejected
      at read with a warning; the parser will not allocate
      unbounded memory for a single record
    * FF_MAX_LMC_DEPTH = 100 \u2014 `corrects:` chain walk stops at this
      depth and emits a warning, preventing a malicious correction
      cycle from spinning the apply pass
    * FF_VERSION_SCAN_MAX = 16 \u2014 header version probe never reads
      past 16 leading comment lines, even on garbage input
    * Empty / inverted byte-range early-return in page-up read path
      so a malformed time filter cannot cause an unbounded scan
    * Zero-entry index guard so a file whose every line failed to
      parse cannot cause a NULL deref on later page-up
- LMC sender validation: an incoming correction whose sender does
  not match the original message's sender is rejected at write
  time and surfaced via cons_show_error; a cycle in the apply pass
  is broken via a visited-set
- jid_create_from_bare_and_resource treats NULL, empty string, and
  the literal "(null)" as no resource and returns a bare jid;
  similar normalisation for barejid eliminates the legacy
  "user@host/(null)" artefact that leaked into stored fulljids
  whenever g_strdup_printf("%s", NULL) ran inside create_fulljid

Commands
- `/history switch sqlite|flatfile` \u2014 runtime backend swap, closes
  the old backend and opens the new one without reconnecting
- `/history export [<jid>]` \u2014 SQLite -> flat-file, merging with any
  existing flatlog (dedup keyed on a SHA-256 hash mixing stanza_id,
  timestamp, from_jid, body \u2014 robust against id reuse by older
  clients)
- `/history import [<jid>]` \u2014 flat-file -> SQLite, same merge
  semantics, runs inside a single SQLite transaction with rollback
  on per-contact failure
- `/history verify [<jid>]` \u2014 integrity check; emits a structured
  list of issues (ERROR / WARNING / INFO) per file:
    * file-level: missing log, wrong permissions (\u2260 0600), UTF-8
      BOM present, CRLF line endings, empty file
    * line-level: invalid UTF-8 (with byte offset), embedded
      control characters, unparsable lines, timestamps out of
      order, duplicate `id:` and `aid:` (tracked separately so a
      stanza/archive id collision isn't double-reported)
    * cross-line: broken `corrects:` references whose target id is
      not present in the file
- `/history backend` \u2014 show currently active backend
- Active backend indicator `[sqlite]` / `[flatfile]` in the status
  bar next to the JID
- Roster-JID autocomplete for verify / export / import
- export and import open a SQLite handle on demand when the
  flatfile backend is currently active, so migration works
  regardless of which backend is live

Tests
- Unit: database_export (parser round-trip, escape/unescape, dedup
  key stability, JID normalisation), database_stress (14 cases
  exercising rapid writes, large messages, deep LMC chains, MAM
  dedup, concurrent contacts)
- Functional: history persistence across reconnects, export /
  import round-trip with content equality, MUC migration,
  timestamp normalisation across timezones
- Bench harness P1\u2013P5 (synthetic load: bulk insert, time-range
  read, page-up scroll, MAM ingest, mixed workload) and failure
  modes F1\u2013F17 (page-up cursor and forward-iteration symmetry,
  oversized lines, MAM dedup, LMC depth and cycles, BOM/CRLF,
  missing log, empty file, mtime+inode flip, broken corrects, etc.)
- All bench tests integrate with the existing make targets and
  emit CSV rows for baseline comparison

Author: jabber.developer2 <jabber.developer2@jabber.space>
Reviewed-by: jabber.developer <jabber.developer@jabber.space>
2026-05-05 19:26:07 +00:00

708 lines
25 KiB
C

// SPDX-License-Identifier: GPL-3.0-or-later
//
// This file is part of CProof.
// See LICENSE for the full GPLv3 text and the special OpenSSL linking exception.
/*
* bench_export_import.c
* vim: expandtab:ts=4:sts=4:sw=4
*
* S7/S8 export + import bench harness.
*
* Subcommands:
* seed --db=PATH --rows=N [--profile=mixed] [--lmc-rate=PCT]
* export --account=JID --csv=PATH [--label=TAG]
* import --account=JID --csv=PATH [--label=TAG]
* roundtrip --rows=N --csv=PATH # seed → export → fresh DB → import → full diff
* verify --db-a=PATH --db-b=PATH # full byte-by-byte content diff of two SQLite DBs
*
* Layout (under $BENCH_DATA_DIR, default /tmp/cproof-bench-export):
* database/<account_jid_dir>/chatlog.db
* flatlog/<account_jid_dir>/<contact_jid_dir>/history.log
*
* Seeding writes directly to SQLite via a separate sqlite3 handle (faster than
* dispatching through the backend's add_incoming for each row). Export/import
* uses the *real* log_database_export_to_flatfile / _import_from_flatfile.
*/
#include "config.h"
#include <errno.h>
#include <fcntl.h>
#include <inttypes.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/stat.h>
#include <time.h>
#include <unistd.h>
#include <glib.h>
#include <glib/gstdio.h>
#include <sqlite3.h>
#include "bench_common.h"
#include "bench_csv.h"
#include "config/account.h"
#include "database.h"
#include "database_flatfile.h"
// ---------------------------------------------------------------------------
// Tiny RNG / body helpers — duplicated from gen_history (kept local so this
// binary is independent of gen_history layout).
static uint64_t
xrng(uint64_t* s)
{
uint64_t x = *s;
x ^= x << 13;
x ^= x >> 7;
x ^= x << 17;
*s = x;
return x;
}
static char*
make_uuid(uint64_t* rng)
{
uint64_t a = xrng(rng), b = xrng(rng);
return g_strdup_printf("%08x-%04x-4%03x-%04x-%012lx",
(unsigned)(a >> 32),
(unsigned)((a >> 16) & 0xffff),
(unsigned)(a & 0xfff),
(unsigned)((b >> 48) & 0xffff) | 0x8000,
(unsigned long)(b & 0xffffffffffffUL));
}
static char*
make_body(uint64_t* rng, size_t target)
{
char* buf = g_malloc(target + 1);
static const char alpha[] = "abcdefghijklmnopqrstuvwxyz0123456789 ";
static const int alpha_n = sizeof(alpha) - 1;
for (size_t i = 0; i < target; i++)
buf[i] = alpha[xrng(rng) % alpha_n];
buf[target] = '\0';
return buf;
}
// ---------------------------------------------------------------------------
// Account dir helpers
static char*
db_path_for(const char* account_jid)
{
const char* base = getenv("BENCH_DATA_DIR");
if (!base || !base[0])
base = "/tmp/cproof-bench-export";
auto_gchar gchar* jid_dir = ff_jid_to_dir(account_jid);
auto_gchar gchar* parent = g_strdup_printf("%s/database/%s", base, jid_dir);
g_mkdir_with_parents(parent, 0755);
return g_strdup_printf("%s/chatlog.db", parent);
}
// ---------------------------------------------------------------------------
// Schema (lifted from database_sqlite.c so the seeder can populate without
// going through _sqlite_init's full machinery)
static const char* SCHEMA_DDL = "CREATE TABLE IF NOT EXISTS `ChatLogs` ("
"`id` INTEGER PRIMARY KEY AUTOINCREMENT, "
"`from_jid` TEXT NOT NULL, "
"`to_jid` TEXT NOT NULL, "
"`from_resource` TEXT, `to_resource` TEXT, "
"`message` TEXT, `timestamp` TEXT, `type` TEXT, "
"`stanza_id` TEXT, `archive_id` TEXT, "
"`encryption` TEXT, `marked_read` INTEGER, "
"`replace_id` TEXT, "
"`replaces_db_id` INTEGER, `replaced_by_db_id` INTEGER);"
"CREATE TABLE IF NOT EXISTS `DbVersion` ("
"`dv_id` INTEGER PRIMARY KEY, `version` INTEGER UNIQUE);"
"INSERT OR IGNORE INTO `DbVersion` (`version`) VALUES ('2');"
"CREATE INDEX IF NOT EXISTS ChatLogs_timestamp_IDX ON `ChatLogs` (`timestamp`);"
"CREATE INDEX IF NOT EXISTS ChatLogs_to_from_jid_IDX ON `ChatLogs` (`to_jid`, `from_jid`);"
"CREATE TRIGGER IF NOT EXISTS update_corrected_message "
"AFTER INSERT ON ChatLogs FOR EACH ROW WHEN NEW.replaces_db_id IS NOT NULL "
"BEGIN UPDATE ChatLogs SET replaced_by_db_id = NEW.id "
"WHERE id = NEW.replaces_db_id; END;";
// ---------------------------------------------------------------------------
// Seed: open sqlite3 directly, build schema, batched-INSERT N rows.
typedef struct
{
const char* db_path;
int64_t rows;
int contacts;
int lmc_pct;
size_t avg_body;
uint64_t seed;
const char* account_jid;
} seed_opts_t;
static int
do_seed(const seed_opts_t* o)
{
sqlite3* db = NULL;
if (sqlite3_open(o->db_path, &db) != SQLITE_OK) {
fprintf(stderr, "seed: cannot open %s: %s\n", o->db_path, sqlite3_errmsg(db));
return 2;
}
char* err = NULL;
if (sqlite3_exec(db, SCHEMA_DDL, NULL, NULL, &err) != SQLITE_OK) {
fprintf(stderr, "seed: schema failed: %s\n", err);
sqlite3_free(err);
sqlite3_close(db);
return 2;
}
sqlite3_exec(db, "PRAGMA journal_mode=WAL;", NULL, NULL, NULL);
sqlite3_exec(db, "PRAGMA synchronous=NORMAL;", NULL, NULL, NULL);
if (sqlite3_exec(db, "BEGIN TRANSACTION;", NULL, NULL, &err) != SQLITE_OK) {
fprintf(stderr, "seed: BEGIN failed: %s\n", err);
sqlite3_free(err);
sqlite3_close(db);
return 2;
}
const char* INSERT_SQL = "INSERT INTO ChatLogs ("
"from_jid, to_jid, from_resource, to_resource, message, timestamp, type, "
"stanza_id, archive_id, encryption, marked_read, replace_id"
") VALUES (?, ?, ?, ?, ?, ?, 'chat', ?, ?, 'none', -1, ?);";
sqlite3_stmt* stmt = NULL;
if (sqlite3_prepare_v2(db, INSERT_SQL, -1, &stmt, NULL) != SQLITE_OK) {
fprintf(stderr, "seed: prepare failed: %s\n", sqlite3_errmsg(db));
sqlite3_close(db);
return 2;
}
uint64_t rng = o->seed ? o->seed : 1;
GDateTime* base = g_date_time_new_utc(2020, 1, 1, 0, 0, 0.0);
int64_t span_secs = (int64_t)5 * 365 * 24 * 3600; // 5 years
char* prev_sid_per_contact[1024] = { 0 };
int n_contacts = o->contacts > 0 ? o->contacts : 1;
if (n_contacts > 1024)
n_contacts = 1024;
double t0 = bench_now_ms();
for (int64_t i = 0; i < o->rows; i++) {
int ci = (int)(i % n_contacts);
auto_gchar gchar* contact_jid = g_strdup_printf("buddy%03d@bench.example", ci);
gboolean is_lmc = prev_sid_per_contact[ci] != NULL && o->lmc_pct > 0
&& (int)(xrng(&rng) % 100) < o->lmc_pct;
int64_t off_secs = (int64_t)((double)i / (double)o->rows * (double)span_secs);
GDateTime* ts = g_date_time_add_seconds(base, off_secs);
auto_gchar gchar* iso = g_date_time_format_iso8601(ts);
g_date_time_unref(ts);
auto_gchar gchar* sid = make_uuid(&rng);
auto_gchar gchar* aid = (xrng(&rng) & 1) ? make_uuid(&rng) : NULL;
size_t body_len = o->avg_body > 0 ? o->avg_body : 50 + (xrng(&rng) % 200);
auto_gchar gchar* body = make_body(&rng, body_len);
auto_gchar gchar* res = g_strdup_printf("res-%d", (int)(xrng(&rng) % 3));
sqlite3_bind_text(stmt, 1, contact_jid, -1, SQLITE_TRANSIENT);
sqlite3_bind_text(stmt, 2, o->account_jid, -1, SQLITE_TRANSIENT);
sqlite3_bind_text(stmt, 3, res, -1, SQLITE_TRANSIENT);
sqlite3_bind_null(stmt, 4); // to_resource
sqlite3_bind_text(stmt, 5, body, -1, SQLITE_TRANSIENT);
sqlite3_bind_text(stmt, 6, iso, -1, SQLITE_TRANSIENT);
sqlite3_bind_text(stmt, 7, sid, -1, SQLITE_TRANSIENT);
if (aid)
sqlite3_bind_text(stmt, 8, aid, -1, SQLITE_TRANSIENT);
else
sqlite3_bind_null(stmt, 8);
if (is_lmc)
sqlite3_bind_text(stmt, 9, prev_sid_per_contact[ci], -1, SQLITE_TRANSIENT);
else
sqlite3_bind_null(stmt, 9);
if (sqlite3_step(stmt) != SQLITE_DONE) {
fprintf(stderr, "seed: insert failed at row %" PRId64 ": %s\n",
i, sqlite3_errmsg(db));
sqlite3_finalize(stmt);
sqlite3_close(db);
return 2;
}
sqlite3_reset(stmt);
if (!is_lmc) {
g_free(prev_sid_per_contact[ci]);
prev_sid_per_contact[ci] = g_strdup(sid);
}
if (((i + 1) % 100000) == 0) {
double dt = bench_now_ms() - t0;
fprintf(stderr, " seeded %" PRId64 " / %" PRId64 " (%.1fs, %.0f rows/s)\n",
i + 1, o->rows, dt / 1000.0, (double)(i + 1) / (dt / 1000.0 + 1e-9));
}
}
sqlite3_finalize(stmt);
if (sqlite3_exec(db, "COMMIT;", NULL, NULL, &err) != SQLITE_OK) {
fprintf(stderr, "seed: COMMIT failed: %s\n", err);
sqlite3_free(err);
}
sqlite3_close(db);
g_date_time_unref(base);
for (int i = 0; i < n_contacts; i++)
g_free(prev_sid_per_contact[i]);
double dt = bench_now_ms() - t0;
fprintf(stderr, "seed: done %" PRId64 " rows in %.2fs\n", o->rows, dt / 1000.0);
return 0;
}
// ---------------------------------------------------------------------------
// Init dispatcher (sets active_db_backend = sqlite, runs _sqlite_init)
static gboolean
init_sqlite_backend(const char* account_jid)
{
setenv("BENCH_ACCOUNT_JID", account_jid, 1);
db_backend_t* be = db_backend_sqlite();
if (!be) {
fprintf(stderr, "init: db_backend_sqlite() returned NULL\n");
return FALSE;
}
active_db_backend = be;
ProfAccount fake = { 0 };
fake.jid = (gchar*)account_jid;
if (!be->init(&fake)) {
fprintf(stderr, "init: backend init failed\n");
return FALSE;
}
return TRUE;
}
static void
close_sqlite_backend(void)
{
if (active_db_backend && active_db_backend->close)
active_db_backend->close();
active_db_backend = NULL;
}
// ---------------------------------------------------------------------------
// SQLite row count
static int64_t
db_row_count(const char* db_path)
{
sqlite3* db = NULL;
if (sqlite3_open(db_path, &db) != SQLITE_OK)
return -1;
int64_t n = -1;
sqlite3_stmt* stmt = NULL;
if (sqlite3_prepare_v2(db, "SELECT COUNT(*) FROM ChatLogs;", -1, &stmt, NULL) == SQLITE_OK) {
if (sqlite3_step(stmt) == SQLITE_ROW)
n = sqlite3_column_int64(stmt, 0);
sqlite3_finalize(stmt);
}
sqlite3_close(db);
return n;
}
// Full content diff: dump every ChatLogs row from both DBs ordered by
// (timestamp, stanza_id), compare tuple-wise. Returns count of mismatches.
// Tuples compared: (from_jid, to_jid, message, timestamp, type, stanza_id,
// archive_id, encryption, replace_id).
// marked_read intentionally excluded — it's not faithfully roundtripped
// (export reads -1/null from SQLite, flatfile may emit it differently).
static int64_t
db_diff_full(const char* a_path, const char* b_path, int64_t* total_a, int64_t* total_b)
{
sqlite3* da = NULL;
sqlite3* db = NULL;
if (sqlite3_open(a_path, &da) != SQLITE_OK)
return -1;
if (sqlite3_open(b_path, &db) != SQLITE_OK) {
sqlite3_close(da);
return -1;
}
const char* SQL = "SELECT IFNULL(from_jid,''), IFNULL(to_jid,''), IFNULL(message,''), "
"IFNULL(timestamp,''), IFNULL(type,''), IFNULL(stanza_id,''), "
"IFNULL(archive_id,''), IFNULL(encryption,''), IFNULL(replace_id,'') "
"FROM ChatLogs ORDER BY timestamp ASC, stanza_id ASC;";
sqlite3_stmt* sa = NULL;
sqlite3_stmt* sb = NULL;
if (sqlite3_prepare_v2(da, SQL, -1, &sa, NULL) != SQLITE_OK
|| sqlite3_prepare_v2(db, SQL, -1, &sb, NULL) != SQLITE_OK) {
if (sa)
sqlite3_finalize(sa);
if (sb)
sqlite3_finalize(sb);
sqlite3_close(da);
sqlite3_close(db);
return -1;
}
int64_t na = 0, nb = 0, mismatches = 0;
int shown = 0;
while (1) {
int ra = sqlite3_step(sa);
int rb = sqlite3_step(sb);
if (ra != SQLITE_ROW && rb != SQLITE_ROW)
break;
if (ra == SQLITE_ROW)
na++;
if (rb == SQLITE_ROW)
nb++;
if (ra != rb) {
mismatches++;
if (shown < 5) {
fprintf(stderr, " diff: row count drift at ra_row=%" PRId64 " rb_row=%" PRId64 " ra=%d rb=%d\n", na, nb, ra, rb);
shown++;
}
continue;
}
for (int c = 0; c < 9; c++) {
const unsigned char* va = sqlite3_column_text(sa, c);
const unsigned char* vb = sqlite3_column_text(sb, c);
const char* sva = (const char*)(va ? va : (const unsigned char*)"");
const char* svb = (const char*)(vb ? vb : (const unsigned char*)"");
if (g_strcmp0(sva, svb) != 0) {
mismatches++;
if (shown < 5) {
fprintf(stderr, " diff: row %" PRId64 " col=%d a=\"%s\" b=\"%s\"\n",
na, c, sva, svb);
shown++;
}
break;
}
}
}
sqlite3_finalize(sa);
sqlite3_finalize(sb);
sqlite3_close(da);
sqlite3_close(db);
if (total_a)
*total_a = na;
if (total_b)
*total_b = nb;
return mismatches;
}
// ---------------------------------------------------------------------------
// CLI dispatch
static void
usage(void)
{
fputs(
"Usage: bench_export_import SUBCMD [options]\n\n"
"Subcommands:\n"
" seed --db=PATH --rows=N [--contacts=K] [--lmc=PCT] [--body=BYTES] [--seed=S] [--account=JID]\n"
" export --account=JID --csv=PATH [--label=TAG]\n"
" import --account=JID --csv=PATH [--label=TAG]\n"
" roundtrip --rows=N --csv=PATH [--label=TAG] [--lmc=PCT] [--body=BYTES] [--full-diff]\n"
" verify --db-a=PATH --db-b=PATH\n",
stderr);
}
static int
cmd_seed(int argc, char** argv)
{
seed_opts_t o = { 0 };
o.account_jid = "bench@bench.example";
o.rows = 1000;
o.contacts = 1;
o.lmc_pct = 0;
o.avg_body = 0;
o.seed = 42;
char* db_path = NULL;
for (int i = 0; i < argc; i++) {
const char* a = argv[i];
if (strncmp(a, "--db=", 5) == 0)
db_path = g_strdup(a + 5);
else if (strncmp(a, "--rows=", 7) == 0)
o.rows = strtoll(a + 7, NULL, 10);
else if (strncmp(a, "--contacts=", 11) == 0)
o.contacts = atoi(a + 11);
else if (strncmp(a, "--lmc=", 6) == 0)
o.lmc_pct = atoi(a + 6);
else if (strncmp(a, "--body=", 7) == 0)
o.avg_body = strtoull(a + 7, NULL, 10);
else if (strncmp(a, "--seed=", 7) == 0)
o.seed = strtoull(a + 7, NULL, 10);
else if (strncmp(a, "--account=", 10) == 0)
o.account_jid = a + 10;
}
if (!db_path)
db_path = db_path_for(o.account_jid);
o.db_path = db_path;
int r = do_seed(&o);
g_free(db_path);
return r;
}
static int
cmd_export(int argc, char** argv)
{
const char* account = "bench@bench.example";
const char* csv = NULL;
const char* label = "S7_export";
const char* volume = "export";
for (int i = 0; i < argc; i++) {
const char* a = argv[i];
if (strncmp(a, "--account=", 10) == 0)
account = a + 10;
else if (strncmp(a, "--csv=", 6) == 0)
csv = a + 6;
else if (strncmp(a, "--label=", 8) == 0)
label = a + 8;
else if (strncmp(a, "--volume=", 9) == 0)
volume = a + 9;
}
if (!init_sqlite_backend(account))
return 2;
auto_gchar gchar* db_path = db_path_for(account);
int64_t rows_before = db_row_count(db_path);
double t0 = bench_now_ms();
int exported = log_database_export_to_flatfile(NULL);
double dt = bench_now_ms() - t0;
long rss = bench_peak_rss_kb();
fprintf(stderr, " %s: %d rows exported in %.2fs (db_rows=%" PRId64 ")\n",
label, exported, dt / 1000.0, rows_before);
if (csv) {
struct stat st;
uint64_t db_size = (stat(db_path, &st) == 0) ? (uint64_t)st.st_size : 0;
auto_gchar gchar* note = g_strdup_printf(
"exported=%d db_rows=%" PRId64, exported, rows_before);
bench_csv_append(csv, label, volume, db_size, (uint64_t)rows_before, dt, rss, note);
}
close_sqlite_backend();
return exported >= 0 ? 0 : 1;
}
static int
cmd_import(int argc, char** argv)
{
const char* account = "bench@bench.example";
const char* csv = NULL;
const char* label = "S8_import";
const char* volume = "import";
for (int i = 0; i < argc; i++) {
const char* a = argv[i];
if (strncmp(a, "--account=", 10) == 0)
account = a + 10;
else if (strncmp(a, "--csv=", 6) == 0)
csv = a + 6;
else if (strncmp(a, "--label=", 8) == 0)
label = a + 8;
else if (strncmp(a, "--volume=", 9) == 0)
volume = a + 9;
}
if (!init_sqlite_backend(account))
return 2;
auto_gchar gchar* db_path = db_path_for(account);
int64_t rows_before = db_row_count(db_path);
double t0 = bench_now_ms();
int imported = log_database_import_from_flatfile(NULL);
double dt = bench_now_ms() - t0;
long rss = bench_peak_rss_kb();
int64_t rows_after = db_row_count(db_path);
fprintf(stderr, " %s: %d imported, db rows %" PRId64 " -> %" PRId64 " in %.2fs\n",
label, imported, rows_before, rows_after, dt / 1000.0);
if (csv) {
struct stat st;
uint64_t db_size = (stat(db_path, &st) == 0) ? (uint64_t)st.st_size : 0;
auto_gchar gchar* note = g_strdup_printf(
"imported=%d rows_before=%" PRId64 " rows_after=%" PRId64,
imported, rows_before, rows_after);
bench_csv_append(csv, label, volume, db_size,
(uint64_t)(rows_after - rows_before), dt, rss, note);
}
close_sqlite_backend();
return imported >= 0 ? 0 : 1;
}
// Roundtrip: seed_A → export → import to fresh DB_B → diff(A, B)
static int
cmd_roundtrip(int argc, char** argv)
{
int64_t rows = 10000;
const char* csv = NULL;
const char* label = "S8e_roundtrip";
const char* volume = "roundtrip";
int lmc = 0;
size_t body = 0;
int do_diff = 0;
for (int i = 0; i < argc; i++) {
const char* a = argv[i];
if (strncmp(a, "--rows=", 7) == 0)
rows = strtoll(a + 7, NULL, 10);
else if (strncmp(a, "--csv=", 6) == 0)
csv = a + 6;
else if (strncmp(a, "--label=", 8) == 0)
label = a + 8;
else if (strncmp(a, "--volume=", 9) == 0)
volume = a + 9;
else if (strncmp(a, "--lmc=", 6) == 0)
lmc = atoi(a + 6);
else if (strncmp(a, "--body=", 7) == 0)
body = strtoull(a + 7, NULL, 10);
else if (strcmp(a, "--full-diff") == 0)
do_diff = 1;
}
const char* account_a = "rt-a@bench.example";
const char* account_b = "rt-b@bench.example";
auto_gchar gchar* db_a = db_path_for(account_a);
auto_gchar gchar* db_b = db_path_for(account_b);
unlink(db_a);
unlink(db_b);
auto_gchar gchar* wal_a = g_strdup_printf("%s-wal", db_a);
unlink(wal_a);
auto_gchar gchar* shm_a = g_strdup_printf("%s-shm", db_a);
unlink(shm_a);
auto_gchar gchar* wal_b = g_strdup_printf("%s-wal", db_b);
unlink(wal_b);
auto_gchar gchar* shm_b = g_strdup_printf("%s-shm", db_b);
unlink(shm_b);
fprintf(stderr, "===== roundtrip: rows=%" PRId64 " account_a=%s account_b=%s =====\n",
rows, account_a, account_b);
// 1) Seed DB_A
seed_opts_t so = { 0 };
so.db_path = db_a;
so.rows = rows;
so.contacts = 1;
so.lmc_pct = lmc;
so.avg_body = body;
so.seed = 42;
so.account_jid = account_a;
double seed_ms_t0 = bench_now_ms();
if (do_seed(&so) != 0)
return 2;
double seed_ms = bench_now_ms() - seed_ms_t0;
// 2) Export DB_A → flatlog under account_a
if (!init_sqlite_backend(account_a))
return 2;
double export_t0 = bench_now_ms();
int exported = log_database_export_to_flatfile(NULL);
double export_ms = bench_now_ms() - export_t0;
close_sqlite_backend();
fprintf(stderr, " exported %d rows in %.2fs\n", exported, export_ms / 1000.0);
// 3) Cross-mount the flatlog tree at account_b's expected location
const char* base = getenv("BENCH_DATA_DIR");
if (!base || !base[0])
base = "/tmp/cproof-bench-export";
auto_gchar gchar* dir_a = g_strdup_printf("%s/flatlog/%s", base, ff_jid_to_dir(account_a));
auto_gchar gchar* dir_b = g_strdup_printf("%s/flatlog/%s", base, ff_jid_to_dir(account_b));
auto_gchar gchar* parent_b = g_path_get_dirname(dir_b);
g_mkdir_with_parents(parent_b, 0755);
// Symlink dir_a as dir_b so import on account_b reads the same files.
unlink(dir_b);
if (symlink(dir_a, dir_b) != 0 && errno != EEXIST) {
fprintf(stderr, " symlink %s -> %s failed: %s\n", dir_b, dir_a, strerror(errno));
return 2;
}
// 4) Import flatlog → DB_B
if (!init_sqlite_backend(account_b))
return 2;
double import_t0 = bench_now_ms();
int imported = log_database_import_from_flatfile(NULL);
double import_ms = bench_now_ms() - import_t0;
close_sqlite_backend();
fprintf(stderr, " imported %d rows in %.2fs\n", imported, import_ms / 1000.0);
int64_t rows_a = db_row_count(db_a);
int64_t rows_b = db_row_count(db_b);
fprintf(stderr, " row counts: A=%" PRId64 " B=%" PRId64 "\n", rows_a, rows_b);
int64_t mismatches = 0;
double diff_ms = 0;
if (do_diff) {
double diff_t0 = bench_now_ms();
int64_t na = 0, nb = 0;
mismatches = db_diff_full(db_a, db_b, &na, &nb);
diff_ms = bench_now_ms() - diff_t0;
fprintf(stderr, " full content diff: a_rows=%" PRId64 " b_rows=%" PRId64 " mismatches=%" PRId64 " (%.2fs)\n",
na, nb, mismatches, diff_ms / 1000.0);
}
long rss = bench_peak_rss_kb();
double total_ms = seed_ms + export_ms + import_ms + diff_ms;
if (csv) {
struct stat st;
uint64_t db_size = (stat(db_a, &st) == 0) ? (uint64_t)st.st_size : 0;
auto_gchar gchar* note = g_strdup_printf(
"rows=%" PRId64 " seed=%.0fms export=%.0fms import=%.0fms diff=%.0fms "
"exported=%d imported=%d rows_a=%" PRId64 " rows_b=%" PRId64
" mismatches=%" PRId64 " body=%zu lmc=%d",
rows, seed_ms, export_ms, import_ms, diff_ms,
exported, imported, rows_a, rows_b, mismatches, body, lmc);
bench_csv_append(csv, label, volume, db_size, (uint64_t)rows,
total_ms, rss, note);
}
int ok = (rows_a == rows_b)
&& (!do_diff || mismatches == 0)
&& exported >= 0 && imported >= 0;
fprintf(stderr, " %s\n", ok ? "PASS" : "FAIL");
return ok ? 0 : 1;
}
static int
cmd_verify(int argc, char** argv)
{
const char* a = NULL;
const char* b = NULL;
for (int i = 0; i < argc; i++) {
const char* x = argv[i];
if (strncmp(x, "--db-a=", 7) == 0)
a = x + 7;
else if (strncmp(x, "--db-b=", 7) == 0)
b = x + 7;
}
if (!a || !b) {
usage();
return 2;
}
int64_t na = 0, nb = 0;
int64_t m = db_diff_full(a, b, &na, &nb);
fprintf(stderr, "verify: a_rows=%" PRId64 " b_rows=%" PRId64 " mismatches=%" PRId64 "\n",
na, nb, m);
return (m == 0 && na == nb) ? 0 : 1;
}
int
main(int argc, char** argv)
{
if (argc < 2) {
usage();
return 2;
}
const char* sub = argv[1];
int sub_argc = argc - 2;
char** sub_argv = argv + 2;
if (g_strcmp0(sub, "seed") == 0)
return cmd_seed(sub_argc, sub_argv);
if (g_strcmp0(sub, "export") == 0)
return cmd_export(sub_argc, sub_argv);
if (g_strcmp0(sub, "import") == 0)
return cmd_import(sub_argc, sub_argv);
if (g_strcmp0(sub, "roundtrip") == 0)
return cmd_roundtrip(sub_argc, sub_argv);
if (g_strcmp0(sub, "verify") == 0)
return cmd_verify(sub_argc, sub_argv);
usage();
return 2;
}