Horton

Thread

Commits

  1. Doc: explain that the string types can't store \0 (ASCII NUL).

  1. Mention invalid null byte sequence

    The Post Office <noreply@postgresql.org> — 2020-12-05T21:58:49Z

    The following documentation comment has been logged on the website:

Page: https://www.postgresql.org/docs/13/datatype-character.html
Description:

I discovered accidentally that PostgreSQL doesn't accept null byte in text
type. It seems that Oracle does (see
https://www.postgresql.org/message-id/de752e01-f36c-821e-9181-cfba78c0fbc8%40propaas.com)
and SQLite does it too.

So it should written in the character type that null byte are not accepted,
it would make like easier to migrate to PostgreSQL :)

  2. Re: Mention invalid null byte sequence

    Laurenz Albe <laurenz.albe@cybertec.at> — 2020-12-07T09:02:43Z

    On Sat, 2020-12-05 at 21:58 +0000, PG Doc comments form wrote:
> The following documentation comment has been logged on the website:
> 
> Page: https://www.postgresql.org/docs/13/datatype-character.html
> Description:
> 
> I discovered accidentally that PostgreSQL doesn't accept null byte in text
> type. It seems that Oracle does (see
> https://www.postgresql.org/message-id/de752e01-f36c-821e-9181-cfba78c0fbc8%40propaas.com)
> and SQLite does it too.
> 
> So it should written in the character type that null byte are not accepted,
> it would make like easier to migrate to PostgreSQL :)

+1; how about the attached patch?

Yours,
Laurenz Albe

  3. Re: Mention invalid null byte sequence

    Adrien CLERC <bugs-postgresql@antipoul.fr> — 2020-12-07T09:57:47Z

    Le 07/12/2020 à 10:02, Laurenz Albe a écrit :
> On Sat, 2020-12-05 at 21:58 +0000, PG Doc comments form wrote:
>> The following documentation comment has been logged on the website:
>>
>> Page: https://www.postgresql.org/docs/13/datatype-character.html
>> Description:
>>
>> I discovered accidentally that PostgreSQL doesn't accept null byte in text
>> type. It seems that Oracle does (see
>> https://www.postgresql.org/message-id/de752e01-f36c-821e-9181-cfba78c0fbc8%40propaas.com)
>> and SQLite does it too.
>>
>> So it should written in the character type that null byte are not accepted,
>> it would make like easier to migrate to PostgreSQL :)
> +1; how about the attached patch?

That would be a good start indeed.

I don't know the policy for documentation redundancy in PostgreSQL, but 
it should be good to mention that also in 
https://www.postgresql.org/docs/current/sql-syntax-lexical.html#SQL-SYNTAX-STRINGS-ESCAPE 
since the basic "SELECT E'la\x00la';" will fail while "SELECT 
E'la\x01la';" will not.

And, as a lazy person, I also would like to see it in the general 
datatype page, since it's a common behavior.

Anyway, merging the first patch will enable the search for "NUL 
character" to return a result, and that will be definitively a nice 
improvement!

Have a nice day!

Adrien

  4. Re: Mention invalid null byte sequence

    Tom Lane <tgl@sss.pgh.pa.us> — 2020-12-07T20:27:22Z

    Laurenz Albe <laurenz.albe@cybertec.at> writes:
> On Sat, 2020-12-05 at 21:58 +0000, PG Doc comments form wrote:
>> The following documentation comment has been logged on the website:
>> Page: https://www.postgresql.org/docs/13/datatype-character.html
>> Description:
>> 
>> So it should written in the character type that null byte are not accepted,
>> it would make like easier to migrate to PostgreSQL :)

> +1; how about the attached patch?

I had thought that this was already documented, but after digging around
I can only find it mentioned in the contexts of saying that literal
strings and quoted identifiers can't contain \0.  So yeah, we need to
improve that.

I agree with the submitter that the place one would expect to read about
this is in datatype-character.html.  So I'd propose the attached.
Maybe there's reason to repeat the info in charset.sgml, but it seems
like more of a datatype limitation than a character set issue.

			regards, tom lane

  5. Re: Mention invalid null byte sequence

    Laurenz Albe <laurenz.albe@cybertec.at> — 2020-12-08T10:34:15Z

    On Mon, 2020-12-07 at 15:27 -0500, Tom Lane wrote:
> Laurenz Albe <laurenz.albe@cybertec.at> writes:
> > On Sat, 2020-12-05 at 21:58 +0000, PG Doc comments form wrote:
> > > The following documentation comment has been logged on the website:
> > > Page: https://www.postgresql.org/docs/13/datatype-character.html
> > > Description:
> > > 
> > > So it should written in the character type that null byte are not accepted,
> > > it would make like easier to migrate to PostgreSQL :)
> >
> > +1; how about the attached patch?
> 
> I had thought that this was already documented, but after digging around
> I can only find it mentioned in the contexts of saying that literal
> strings and quoted identifiers can't contain \0.  So yeah, we need to
> improve that.
> 
> I agree with the submitter that the place one would expect to read about
> this is in datatype-character.html.  So I'd propose the attached.
> Maybe there's reason to repeat the info in charset.sgml, but it seems
> like more of a datatype limitation than a character set issue.

+1 on your patch.

Yours,
Laurenz Albe

  6. Re: Mention invalid null byte sequence

    Tom Lane <tgl@sss.pgh.pa.us> — 2020-12-08T17:07:25Z

    Laurenz Albe <laurenz.albe@cybertec.at> writes:
> On Mon, 2020-12-07 at 15:27 -0500, Tom Lane wrote:
>> I agree with the submitter that the place one would expect to read about
>> this is in datatype-character.html.  So I'd propose the attached.
>> Maybe there's reason to repeat the info in charset.sgml, but it seems
>> like more of a datatype limitation than a character set issue.

> +1 on your patch.

Pushed, thanks for looking it over.

			regards, tom lane