* better messages
* fold sql_prepare()
* rewrite the message decode

Add iproto_execute command to get not only sql rows as the result,
but also sql meta data about each result column.

Make SQLITE_LIMIT_VARIABLE_NUMBER be static

SQLITE_LIMIT_VARIABLE_NUMBER is the run time limit of the
variable count in the query. The Tarantool doesn't support dynamic
limits for such things. Let's make this limit be enum instead
of value in the limits array (struct sqlite3.aLimit).

Part of #2285.

Raise syntax error in a case of empty requests. Update sql-tokenizer
to support exising tests.

Test alias.test.lua is fully disabled, see #2582

Part of #2381

Iconv is a library to convert a sequence of characters in one
character encoding to a sequence of characters in another character
encoding. Example below converts utf-16 big endian string into utf-8
string:

    convertor = require('iconv').new('UTF-16BE', 'UTF-8')
    converted_string = convertor(source_string)

Closes #2587

It belongs there, really.

Before this patch, the new space, created by alter specification,
would be put into space cache only after successful WAL write.

This behaviour is not linearizable: on a replica, the WAL is
played sequentially, and the order of events could differ from the
master.

Besides, it could crash, as demonstrated in gh-2074 test case.

Since we use a cascading rollback for all transactions on WAL
write error, it's OK to put a space into space cache
before WAL write, so that the new transactions apply to the new
space.

This patch does exactly that.

All subsequent requests are executed against the new space.

This patch also removes on_replace trigger in the old space, since
all checks against the new tuple format are performed using the new
space.

Fixes #2074.

* UTF-16 support removed
* Tests with UTF-16* cases removed

Closes #2295

closes gh-2576

* update comments
* add a test case for altering a primary key on the fly
* rename AddIndex to CreateIndex
* factor out common code into a function

Non destructive swap_index_def function sould be used because memtx_tree
stores a pointer to an index_def used while tree creation. Fixed #2570

Use `net.box` instead of `console`.

LLVM is more eager for frame size than GCC
Current maximal call chain length may corrupt stack then.
Reduce maximal call chain length to 30.

Related to #2550

If after recovery first client to invoke SQL differs from
ADMIN, then initialization of SQL subsystem will fail due to
insufficient access rights. Problem is that SQL invokes
create_from_tuple() to fill space_def structure, which in turn
verifies access right.
As far as SQL is a static part of Tarantool, there's no sense
to initialize the system by demand. Change performs data dictionary
initialization upon config completion.
Regression test added as well.

Closes #2483

Needed for #2285

* Right now Tarantool runs compilation of a query on top of fiber's stack
  so, so number of triggeres compilation of which was triggered by chain
  should be limited. Introduce dedicated global variable which decreases
  each time new compilation in back track is initiated. Error if zero is
  reached. Set it to limit @ parser
* Add test for verious cases
* Fix out of order build for maintainer mode
* Regenerate parser

Closes #2550

Right now SQL query compiler is run on top of fiber's stack,
which is limited to 64KB by default, so maximum
number of entities in compound SELECT statement should be less than
50 (verified by experiment) or stack guard will be triggered.

In future we'll introduce heuristic which should understand that
query is 'complex' and run compilation in separate thread with larger
stack. This will also allow us not to block Tarantool while compiling
the query.

Closes #2548

This is a first implentation of box.internal.sql_create_function
It works properly on limited set of cases.
The main purpose for this feature by now is tests.

Some test cases relaying on this feature were enabled.

Part of #2233

If a vylog record doesn't get flushed to disk due to an error, objects
it refers to (index->key_def, range->begin and range->end) may get
destroyed, resulting in a crash. To avoid that, we must copy those
objects to vylog buffer.

Closes #2532

 - Remove tx and cursor latencies as they are useless - they actually
   account how long a tx/cursor was open, not latencies.
 - Remove vy_stat->get_latency as it doesn't account latency of select,
   besides we now have per index read latency.
 - Remove vy_stat->dumped_statements and dump_total as these statistics
   are reported per index as well.
 - VY_STAT_TX_OPS is currently unused (always 0). Let's use it for
   accounting the total number of statements committed in tx instead of
   vy_stat->write_count.

index.info() is supposed to show index stats, not options.
box.space.<space_name>.index.<index_name>.options looks like
a better place for reporting index options.

Needed for #1662

Replace box.info.vinyl().performance.upsert_{squashed,applied} with per
index index.info().upsert.{squashed,applied}.

Needed for #1662

closes gh-2516

* Changes to sqltester:
  - Change lsearch to be closer to tcl implementation:
      return index of first match or return -1
  - Make execsql2 work with multistatements
* Converted tests include cases for large (64b) integers

Part of #2381

Replace box.info.vinyl().performance.iterator.{run,mem} global counters
with the following per index counters:

  memory
    iterator
      lookup            # number of lookups in the memory tree
      get               # number of statements returned by mem iterator
        rows
        bytes

  disk
    iterator
      lookup            # number of lookups in the page index
      get               # number of statements returned by run iterator
        rows
        bytes

      bloom             # number of times bloom filter
        hit             #   allowed to avoid a disk read
        miss            #   failed to prevent a disk read

      read              # number of statements actually read from disk
        rows
        bytes
        bytes_compressed
        pages

Needed for #1662

It's useless - it only checks that certain counters are present, but
doesn't give a damn about what they show. At the same time, I have to
update its output every time I modify vinyl statistics, which I'm doing
pretty often these days. So let's ditch it - I'll rewrite it from
scratch after I'm done reworking vinyl statistics.

gh-2520: add comments explaining the test case for gh-2520 (upsert
caching).

Use a tuple cache to turn an UPSERT in the PREPARE, if the cache
contains a tuple with the exact same key. The UPSERT can be applied
to the cached tuples, because
1. The cache always contains only REPLACE statements;
2. The cache always contains only newest statements.

Closes #2520

When an UPSERT is prepared, it can use a cached statement to be
turned into REPLACE. But now it doesn't use the cache.

Test for #2520

Malformed MessagePack can cause `int k` counter overflow
inside mp_check()/mp_next().

See https://github.com/tarantool/nginx_upstream_module/issues/79
See https://github.com/rtsisyk/msgpuck/issues/16
Closes #2540

Follow up 15a9af26

exp2(64) <= UINT64_MAX is true due to type conversion.
Use exp2(64) instead of UINT64_MAX (optimized out even on -O0).

Fixes buggy box/indices_any_type.test.lua on x86_64

Closes #2472

ffi.new('long', x) and ffi.new('unsigned long', x) uses arbitratry
user-defined ctypeid instead of built-in CTID_INT32 and CTID_UINT32.
Our tonumber64() implementation is not ready for that. Remove these
two test cases and move forward. Nobody cares.

Fixes box/misc.test.lua on i386 and armhf

Follow up #2459

It is unsafe, because system spaces use triggers to keep records in sync
with internal objects while space truncation doesn't invoke triggers.

Closes #2523

Currently, space.create must insert an entry into 'truncate' system
space - if it does not, space.truncate won't work. This is incontinent,
as it makes it impossible to create spaces by simply inserting a tuple
into the 'space' system space via iproto API. So let's insert entries
into truncate space lazily, on the first space truncation.

Closes #2524