I updated all the dependencies to the latest versions and removed the replaced dependency.

@auxten Please let me know if this is not OK and if you based the project on certain dependencies. Then we can pin those dependencies. Also, I am thinking about various ways to contribute to this project. Please let me know of the next steps or create tickets, so we can talk about various aspects of it.

This PR removes most of the unused/dead Makefile code relevant to the CockroachDB project. This PR also fixes some small issues inside the build processes such as the import paths defined inside sql.y and the non-alphanumeric build constraint.

All generated files could be generated using make generate. Make generate does not yet ensure that all the necessary tools are installed. This will be included in a future PR.

NOTE: the vendor directory inside the project is currently outdated. I would like to suggest removing the vendor directory since all import issues have been resolved inside this PR.

(echo "// Code generated by goyacc. DO NOT EDIT.";
echo "// GENERATED FILE DO NOT EDIT";
cat pkg/sql/parser/gen/sql.go.tmp |
sed -E 's/^const ([A-Z][_A-Z0-9]) =.$/const \1 = lex.\1/g') > pkg/sql/parser/sql.go.tmp || rm pkg/sql/parser/sql.go.tmp mv -f pkg/sql/parser/sql.go.tmp pkg/sql/parser/sql.go goimports -w pkg/sql/parser/sql.go (echo "// Code generated by make. DO NOT EDIT.";
echo "// GENERATED FILE DO NOT EDIT";
echo;
echo "package lex";
echo;
grep '^const [A-Z][_A-Z0-9]* ' pkg/sql/parser/gen/sql.go.tmp) > pkg/sql/lex/tokens.go.tmp || rm pkg/sql/lex/tokens.go.tmp mv -f pkg/sql/lex/tokens.go.tmp pkg/sql/lex/tokens.go go install -v ./pkg/cmd/prereqs can't load package: package ./pkg/cmd/prereqs: cannot find package "." in:

I am getting a high severity security warning, can we update this dependency?

Improper Input Validation in GoGo Protobuf

An issue was discovered in GoGo Protobuf before 1.3.2. plugin/unmarshal/unmarshal.go lacks certain index validation, aka the "skippy peanut butter" issue.

解析此种类型的sql失败:

SELECT COUNT ( * ) FROM pg_class C LEFT JOIN pg_namespace n ON n.oid = C.relnamespace WHERE C.relkind = ANY ( '{r,v,m}' :: CHAR [] ) UNION SELECT COUNT ( * ) FROM pg_attribute A JOIN pg_class C ON A.attrelid = C.oid JOIN pg_namespace n ON C.relnamespace = n.oid JOIN pg_type tp ON tp.typelem = A.atttypid WHERE A.attnum > 0 UNION SELECT COUNT ( * ) FROM information_schema.routines

出错的位置在 "::" 操作符

github.com/auxten/postgresql-parser/pkg/util/hlc ../../auxten/postgresql-parser/pkg/util/hlc/hlc_clock_device_linux.go:51:11: cannot use ctx (type context.Context) as type string in argument to logrus.Infof

Using arch. Commented out that single log write on line 51 and 'log' import and the example worked.

Thanks!

Desc

Upgrade github.com/gogo/protobuf version close #17

Note

I just upgrade package version and i don't know how to test it. I only tried example to be executable.

CockroachDB uses "STRING" (but accepts "TEXT" for compatibility) and PostgreSQL uses "TEXT" always.

Fixes https://github.com/auxten/postgresql-parser/issues/15

I am not sure if you are planning to use this code for future features. I would like to suggest to remove the storage package if there are no plans made for this package.

go vet currently throws the error that main is redeclared inside the example directory. This PR fixes this issue by moving the examples into separate directories.

Hi,

I'm experimenting with postgresql-parser to intercept and rewrite some queries to a postgresql database.

One problem I have encountered is that the formatter will output "STRING" by default for string data type. PostgreSQL does not support this, instead the data type name is "TEXT". CockroachDB supports "TEXT" as an alias for "STRING" [1]

In order to make query-rewriting work, I've patched postgresql-parser to output "TEXT" data type by default instead of "STRING". I think this could be a reasonable default since CockroachDB supports both.

[1] https://www.cockroachlabs.com/docs/stable/string.html

Hi,

I'm experimenting with postgresql-parser to intercept and rewrite some queries to a postgresql database.

One problem I have encountered is that the formatter will output "STRING" by default for string data type. PostgreSQL does not support this, instead the data type name is "TEXT". CockroachDB supports "TEXT" as an alias for "STRING" [1]

In order to make query-rewriting work, I've patched postgresql-parser to output "TEXT" data type by default instead of "STRING". I think this could be a reasonable default since CockroachDB supports both.

[1] https://www.cockroachlabs.com/docs/stable/string.html