On Sat, Oct 28, 2023 at 09:39:53PM -0400, Tom Lane wrote:
> Bruce Momjian <br...@momjian.us> writes:
> > My apologies, wrong patch attached, right one attached now.
>
> I think this one is fine as-is:
>
> /* Only single-byte delimiter strings are supported. */
> if (strlen(opts_out->delim) != 1)
> ereport(ERROR,
> - (errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
> + (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
> errmsg("COPY delimiter must be a single
> one-byte character")));
>
> While we have good implementation reasons for this restriction,
> there's nothing illogical about wanting the delimiter to be more
> general. It's particularly silly, from an end-user's standpoint,
> that for example 'é' is an allowed delimiter in LATIN1 encoding
> but not when the server is using UTF8. So I don't see how the
> distinction you presented justifies this change.
Agreed, my mistake.
> + if (opts_out->freeze && !is_from)
> + ereport(ERROR,
> + (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
> + errmsg("COPY freeze only available using COPY
> FROM")));
>
> Not thrilled by the wording here. I don't like the fact that the
> keyword FREEZE isn't capitalized, and I think you omitted too many
> words for intelligibility to be preserved. Notably, all the adjacent
> examples use "must" or "must not", and this decides that that can be
> omitted.
I think it is modeled after:
errmsg("COPY force null only available using COPY FROM")));
> I realize that you probably modeled the non-capitalization on nearby
> messages like "COPY delimiter", but there's a difference IMO:
> "delimiter" can be read as an English noun, but it's hard to read
> "freeze" as a noun.
>
> How about, say,
>
> errmsg("COPY FREEZE must not be used in COPY TO")));
>
> or perhaps that's redundant and we could write
>
> errmsg("FREEZE option must not be used in COPY TO")));
I now have:
errmsg("COPY FREEZE mode only available using COPY FROM")));
Updated patch attached.
--
Bruce Momjian <br...@momjian.us> https://momjian.us
EDB https://enterprisedb.com
Only you can decide what is important to you.
diff --git a/doc/src/sgml/ref/copy.sgml b/doc/src/sgml/ref/copy.sgml
index d12ba96497..82b65543c3 100644
--- a/doc/src/sgml/ref/copy.sgml
+++ b/doc/src/sgml/ref/copy.sgml
@@ -224,6 +224,7 @@ COPY { <replaceable class="parameter">table_name</replaceable> [ ( <replaceable
open and there are no older snapshots held by this transaction. It is
currently not possible to perform a <command>COPY FREEZE</command> on
a partitioned table.
+ This option is allowed only in <command>COPY FROM</command>.
</para>
<para>
Note that all other sessions will immediately be able to see the data
diff --git a/src/backend/commands/copy.c b/src/backend/commands/copy.c
index c5d7d78645..92558b6bb0 100644
--- a/src/backend/commands/copy.c
+++ b/src/backend/commands/copy.c
@@ -728,16 +728,22 @@ ProcessCopyOptions(ParseState *pstate,
/* Don't allow the delimiter to appear in the null string. */
if (strchr(opts_out->null_print, opts_out->delim[0]) != NULL)
ereport(ERROR,
- (errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("COPY delimiter must not appear in the NULL specification")));
/* Don't allow the CSV quote char to appear in the null string. */
if (opts_out->csv_mode &&
strchr(opts_out->null_print, opts_out->quote[0]) != NULL)
ereport(ERROR,
- (errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("CSV quote character must not appear in the NULL specification")));
+ /* Check freeze */
+ if (opts_out->freeze && !is_from)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("COPY FREEZE mode only available using COPY FROM")));
+
if (opts_out->default_print)
{
if (!is_from)
