New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Move SQL TRIGGERs into server #3273

Closed

Gerold103 opened this issue Mar 22, 2018 · 1 comment

Assignees

Labels

ddl refactoring sql

Milestone

Collaborator

Gerold103 commented Mar 22, 2018

Move SQL triggers into Tarantool's struct space.

Gerold103 added refactoring sql ddl sql dd integration labels

Gerold103 self-assigned this

Collaborator Author

Gerold103 commented Mar 22, 2018

Part of #2217.

kyukhin added this to the 2.1.0 milestone

Gerold103 assigned kshcherbatov and unassigned Gerold103

kshcherbatov added a commit that referenced this issue


          box: move db->pShchema init to sql_init

80e32b9

As we are going to call parser on box.cfg() to recreate triggers
from SQL, we should initialize Schema as it used in sqlite3BeginTrigger.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: fix sql len in tarantoolSqlite3RenameTrigger

ddf2fbf

Part of #3273.

kshcherbatov added a commit that referenced this issue


          box: introduce box_space_id_by_name

95c32aa

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

e3842ff

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert builded triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          box: move db->pShchema init to sql_init

9c68415

As we are going to call parser on box.cfg() to recreate triggers
from SQL, we should initialize Schema as it used in sqlite3BeginTrigger.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: fix sql len in tarantoolSqlite3RenameTrigger

9e6efc9

Part of #3273.

kshcherbatov added a commit that referenced this issue


          box: introduce box_space_id_by_name

6fa87e6

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

df86483

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert builded triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

20e7cde

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert builded triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          box: move db->pShchema init to sql_init

c9ddca0

As we are going to call parser on box.cfg() to recreate triggers
from SQL, we should initialize Schema as it used in sqlite3BeginTrigger.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: fix sql len in tarantoolSqlite3RenameTrigger

9293e74

Part of #3273.

kshcherbatov added a commit that referenced this issue


          box: introduce box_space_id_by_name

eacc773

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move sqlite3DeleteTrigger to sql.h

1f0cb06

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

068cdf9

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

c077580

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

8e42899

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          box: move db->pShchema init to sql_init

f35cab0

As we are going to call parser on box.cfg() to recreate triggers
from SQL, we should initialize Schema as it used in sqlite3BeginTrigger.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: fix sql len in tarantoolSqlite3RenameTrigger

56b2020

Part of #3273.

kshcherbatov added a commit that referenced this issue


          box: introduce box_space_id_by_name

7e75129

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move sqlite3DeleteTrigger to sql.h

fff9d84

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

c0750f9

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: remove global sql_trigger hash

1d6bcf8

As new _trigger format contain space_id, we can do not
store AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move sqlite3DeleteTrigger to sql.h

5f865ca

As we are going to port triggers to server, we need
an instrument to release allocated memory in alter.cc.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: new _trigger space format with space_id

b9627e0

As we would like to lookup triggers by space_id in future
on space deletion to delete associated _trigger tuples we
need to introduce new field space_id as secondary key.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

f520759

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.
Global DB hash has been kept as we still ned to get Trigger AST
by name on trigger deletion.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: refactor AST trigger object name

c854303

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: VDBE tests for trigger existence

cff35c8

Trigger presence in system should be tested on each VDBE
execution attempt, not on Parser iteration as system state
could be changed between opcode run.
With this patch, such checks are the part of VDBE program
looking to the _trigger space and raise error message if
tuple with specified key already exists.

Part of #3435, #3273

kshcherbatov added a commit that referenced this issue


          sql: remove global sql_trigger hash

799c3e4

As new _trigger format contain space_id, we can do not
store AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: refactor trigger API functions

6da6af1

Changed Trigger structure name to sql trigger to avoid
confusing similarity with Tarantool trigger structure.
Refactored related code to match tarantool codestyle.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: VDBE tests for trigger existence

ec2d373

Trigger presence in system should be tested on each VDBE
execution attempt, not on Parser iteration as system state
could be changed between opcode run.
With this patch, such checks are the part of VDBE program
looking to the _trigger space and raise error message if
tuple with specified key already exists.

Part of #3435, #3273

kshcherbatov added a commit that referenced this issue


          sql: remove global sql_trigger hash

1c0b25b

As new _trigger format contain space_id, we can do not
store AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: refactor trigger API functions

965b25a

Changed Trigger structure name to sql trigger to avoid
confusing similarity with Tarantool trigger structure.
Refactored related code to match tarantool codestyle.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: VDBE tests for trigger existence

85b9878

Trigger presence in system should be tested on each VDBE
execution attempt, not on Parser iteration as system state
could be changed between opcode run.
With this patch, such checks are the part of VDBE program
looking to the _trigger space and raise error message if
tuple with specified key already exists.

Part of #3435, #3273

kshcherbatov added a commit that referenced this issue


          sql: remove global sql_trigger hash

297a7b0

As new _trigger format contains space_id, we can avoid
storing AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

8ec9bca

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.
Global DB hash has been kept as we still ned to get Trigger AST
by name on trigger deletion.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: refactor trigger API functions

3ef0a1f

Changed Trigger structure name to sql trigger to avoid
confusing similarity with Tarantool trigger structure.
Refactored related code to match tarantool codestyle.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: VDBE tests for trigger existence

8dd4f75

Trigger presence in system should be tested on each VDBE
execution attempt, not on Parser iteration as system state
could be changed between opcode run.
With this patch, such checks are the part of VDBE program
looking to the _trigger space and raise error message if
tuple with specified key already exists.

Part of #3435, #3273

kshcherbatov added a commit that referenced this issue


          sql: remove global sql_trigger hash

da3be9d

As new _trigger format contains space_id, we can avoid
storing AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move sqlite3DeleteTrigger to sql.h

54789d7

As we are going to port triggers to server, we need
an instrument to release allocated memory in alter.cc.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: new _trigger space format with space_id

8443b95

As we would like to lookup triggers by space_id in future
on space deletion to delete associated _trigger tuples we
need to introduce new field space_id as secondary key.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: move Triggers to server

3ea6336

Introduced sql_triggers field in space structure.
Changed parser logic to do not insert built triggers, just only
to do parsing. All triggers insertions and deletions are operated
via on_replace_dd_trigger now.
Global DB hash has been kept as we still ned to get Trigger AST
by name on trigger deletion.

Resolves #3273.

kshcherbatov added a commit that referenced this issue


          sql: refactor trigger API functions

a0444f4

Changed Trigger structure name to sql trigger to avoid
confusing similarity with Tarantool trigger structure.
Refactored related code to match tarantool codestyle.

Part of #3273.

kshcherbatov added a commit that referenced this issue


          sql: VDBE tests for trigger existence

0d52f71

Trigger presence in system should be tested on each VDBE
execution attempt, not on Parser iteration as system state
could be changed between opcode run.
With this patch, such checks are the part of VDBE program
looking to the _trigger space and raise error message if
tuple with specified key already exists.

Part of #3435, #3273

kshcherbatov added a commit that referenced this issue


          sql: remove global sql_trigger hash

a19f342

As new _trigger format contains space_id, we can avoid
storing AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kyukhin pushed a commit that referenced this issue


          sql: move sqlite3DeleteTrigger to sql.h

93867c5

As we are going to port triggers to server, we need
an instrument to release allocated memory in alter.cc.

Part of #3273.

kyukhin pushed a commit that referenced this issue


          sql: new _trigger space format with space_id

7cb3d2c

As we would like to lookup triggers by space_id in future
on space deletion to delete associated _trigger tuples we
need to introduce new field space_id as secondary key.

Part of #3273.

kyukhin closed this as completed in

2c6945c

kyukhin pushed a commit that referenced this issue


          sql: refactor trigger API functions

d2e755b

Changed Trigger structure name to sql trigger to avoid
confusing similarity with Tarantool trigger structure.
Refactored related code to match tarantool codestyle.

Part of #3273.

kyukhin pushed a commit that referenced this issue


          sql: VDBE tests for trigger existence

5b7970b

Trigger presence in system should be tested on each VDBE
execution attempt, not on Parser iteration as system state
could be changed between opcode run.
With this patch, such checks are the part of VDBE program
looking to the _trigger space and raise error message if
tuple with specified key already exists.

Part of #3435, #3273

kyukhin pushed a commit that referenced this issue


          sql: remove global sql_trigger hash

3e16e5d

As new _trigger format contains space_id, we can avoid
storing AST pointers in HASH. Requested AST could be found
by name in appropriate space.

Part of #3273.

kyukhin mentioned this issue

Work with 'IF NOT EXISTS' in VDBE #3435

Closed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment