quote_ident loses encoding

[ged]

Original report by Matt Daw (Bitbucket: mattdaw, GitHub: mattdaw).

---

Calling quote_ident on a UTF8 encoded string "shot_sg_assigned_to___dělá_kdo___connections" produces ASCII encoded string ""shot_sg_assigned_to___d\xC4\x9Bl\xC3\xA1_kdo___connections"".

I'm modifying the result using force_encoding, which is a reasonable workaround... but it would be nice to have it fixed in the gem.

[ged]

Original comment by Michael Granger (Bitbucket: ged, GitHub: ged).

---

Thanks for reporting this. This should be a pretty small fix, so we'll likely wrap it up into the next release.

[ged]

Original comment by Matt Daw (Bitbucket: mattdaw, GitHub: mattdaw).

---

Great, thanks!

Matt

[ged]

Original comment by Lars Kanis (Bitbucket: larskanis, GitHub: larskanis).

---

I can add this it.

Connection#quote_ident is the home-brewed version of #escape_identifier which was introduced in postgresql-9.0. So this is duplicated code. Both are almost equally fast in a simple test I just did (quote_ident has a little less overhead for each call but escape_identifier is faster with more than approximately 7 characters) and they do exactly the same. Both don't set the encoding so far. Any thoughts about the future of them? Should we deprecate quote_ident or alias it to escape_identifier or leave it like it is and add a comment only?

[ged]

Original comment by Lars Kanis (Bitbucket: larskanis, GitHub: larskanis).

---

Should be fixed with 72acd792 .

[ged]

Original comment by Michael Granger (Bitbucket: ged, GitHub: ged).

---

Set proper encoding on the string returned by quote_ident,

From 72acd792eb92004655226fd31307dc821109f2e2 Mon Sep 17 00:00:00 2001
escape_literal and escape_identifier.
This fixes issue #163 : #163 .