Enum value parsing: do not parse by whitespace

[shlomi-noach]

Description

Addressing #15492, this is an initial simple fix that allows whitespace in enum value. A more formal fix to follow, as this fix is incomplete in that it still requires more escaping.

A complete fix is likely to benefit from some sqlparser capabilities, though the use case goes beyond parsing CREATE TABLE statements and thus we might still take a lesser-than-yacc approach.

Related Issue(s)

#15492

Checklist

• "Backport to:" labels have been added if this change should be back-ported to release branches
• If this change is to be back-ported to previous releases, a justification is included in the PR description
• Tests were added or are not required
• Did the new or modified tests pass consistently locally and on CI?
• Documentation was added or is not required

Deployment Notes

[vitess-bot]

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

• Ensure that the Pull Request has a descriptive title.
• Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

• Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

• Apply the release notes (needs details) label if users need to know about this change.
• New features should be documented.
• There should be some code comments as to why things are implemented the way they are.
• There should be a comment at the top of each new or modified test to explain what the test does.

New flags

• Is this flag really necessary?
• Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

• Each item in Jobs should be named in order to mark it as required.
• If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

• Protobuf changes should be wire-compatible.
• Changes to _vt tables and RPCs need to be backward compatible.
• RPC changes should be compatible with vitess-operator
• If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
• vtctl command output order should be stable and awk-able.

[codecov]

Codecov Report

Attention: Patch coverage is 85.71429% with 9 lines in your changes are missing coverage. Please review.

Project coverage is 65.67%. Comparing base (696fe0e) to head (2b4a10c).
Report is 116 commits behind head on main.

Files	Patch %	Lines
go/vt/schema/parser.go	82.75%	5 Missing ⚠️
go/sqltypes/value.go	88.23%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #15493      +/-   ##
==========================================
- Coverage   67.41%   65.67%   -1.75%     
==========================================
  Files        1560     1563       +3     
  Lines      192752   194476    +1724     
==========================================
- Hits       129952   127714    -2238     
- Misses      62800    66762    +3962     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mattlord]

Any reason(s) not to use the csv encoding packing in the stdlib? https://pkg.go.dev/encoding/csv

[dbussink]

The main issue would be that quoting works differently there with " and not ' like in SQL. Also escaping of quotes is different, so we'd have to write something specific for this or leverage instead sqlparser as @shlomi-noach mentioned.

[mattlord]

Yeah, I did notice that the csv encoder simply removes any commas in the values which surprised me. I do think that we should take this opportunity to address this more fully though as MySQL happily accepts enum string values with commas:

mysql> create table t1 (id int not null auto_increment primary key, ev enum('hi there', 'with, comma'));
Query OK, 0 rows affected (0.02 sec)

mysql> show create table t1;
+-------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Table | Create Table                                                                                                                                                                                        |
+-------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| t1    | CREATE TABLE `t1` (
  `id` int NOT NULL AUTO_INCREMENT,
  `ev` enum('hi there','with, comma') DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci |
+-------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.01 sec)

[dbussink]

Right, there's a bunch of things that still fail here. I agree we should fix this properly.

[dbussink]

@shlomi-noach I pushed up a little specific parser for this in f90e797 that deals also with quotes inside the values as well.

I also added some low level test there with stuff like newlines, I think we should also add those to the end-to-end test too to try all the exotic things possible here.

[dbussink]

Realized my parser was not enough and also added a proper decoder for string values so we handle things line newlines correctly.

[dbussink]

@shlomi-noach it looks like MySQL in neither information_schema.columns or when using SHOW CREATE TABLE includes a space after the ,. The new parser doesn't accept that, do you know if we depend anywhere outside of tests on allowing that space?

[shlomi-noach]

Nothing depends on existence of spaces. Part of the fix here is to actually assume they never exist, as you say.

[shlomi-noach]

Additional changes in BufDecodeStringSQL and parseEnumOrSetTokens look good to me.

[shlomi-noach]

Proposed followup in #17133