Re: PG_DUMP producing incorrect SQL

looks like the tabs in your dump file were replaced by spaces

HTH

WBL

On Thu, May 17, 2012 at 7:33 PM, <lindsay(dot)braine(at)emc(dot)com> wrote:

> I am sure I am doing something wrong – because I cannot find anything
> about this on Google or in the Community Archives – but here goes:****
>
> ** **
>
> I am running postgres-9.1.3****
>
> I have one database, two schemas.****
>
> ** **
>
> When using pg_dump –format plain to create an export file, the result is a
> file that contains SQL errors, preventing psql from parsing and executing.
> ****
>
> ** **
>
> Here is the problem:****
>
> ** **
>
> Here is the ouput for one of my tables from pg_dump :****
>
> ** **
>
> --****
>
> -- Name: attributevalue; Type: TABLE; Schema: SATURN; Owner: SATURNuser;
> Tablespace: ****
>
> --****
>
> CREATE TABLE attributevalue (****
>
> dtype character varying(31) NOT NULL,****
>
> f_id uuid NOT NULL,****
>
> f_lastmodified timestamp without time zone NOT NULL,****
>
> f_credentials_id uuid,****
>
> f_value double precision,****
>
> f_flag_value boolean,****
>
> f_int_value integer,****
>
> f_family character varying(255),****
>
> f_processor_type character varying(255),****
>
> f_version character varying(255),****
>
> f_string_value character varying(255)****
>
> );****
>
> ** **
>
> --****
>
> -- Data for Name: attributevalue; Type: TABLE DATA; Schema: SATURN; Owner:
> SATURNuser****
>
> --****
>
> COPY attributevalue (dtype, f_id, f_lastmodified, f_credentials_id,
> f_value, f_flag_value, f_int_value, f_family, f_processor_type, f_version,
> f_string_value) FROM stdin;****
>
> OSAttributeValue fdf0dba2-7683-421e-a58a-1478be78edcc 2012-05-17
> 09:30:04.565 \N \N \N \N WINDOWS amd64 5.2 \N****
>
> OSAttributeValue 26d4f4f9-0d97-4532-9f88-acd31d6e2474 2012-05-17
> 09:30:45.315 \N \N \N \N WINDOWS \N \N \N****
>
> \.****
>
> ** **
>
> Note the presence of the default \N placeholder used for empty/null values.
> ****
>
> I have another table****
>
> ** **
>
> --****
>
> -- Name: config_sysvars; Type: TABLE; Schema: XXX; Owner: SATURNuser;
> Tablespace: ****
>
> --****
>
> CREATE TABLE config_sysvars (****
>
> f_id uuid NOT NULL,****
>
> f_lastmodified timestamp without time zone NOT NULL,****
>
> f_name character varying(255),****
>
> f_type character varying(255),****
>
> f_value character varying(255)****
>
> );****
>
> ** **
>
> --****
>
> -- Data for Name: config_sysvars; Type: TABLE DATA; Schema: XXX; Owner:
> SATURNuser****
>
> --****
>
> COPY config_sysvars (f_id, f_lastmodified, f_name, f_type, f_value) FROM
> stdin;****
>
> 9fbfb915-c42e-4b5e-93ca-6a23abc0a661 2012-05-17 09:27:54.284
> schemaVersion String 6.0.0****
>
> 14c3bf58-e230-4353-9500-46fcb9632d7f 2012-05-17 09:29:52.909 mailHost
> String ****
>
> a1c7824a-ef9c-4546-871a-637748407b5e 2012-05-17 09:29:52.924 mailFrom
> String ****
>
> e8c338d4-be1f-4ad7-830b-905f7942223d 2012-05-17 09:30:04.612
> XXXServerHostUUID UUID 1929b62f-e150-4ea2-aee4-35f8443bc368****
>
> \.****
>
> ** **
>
> Note the missing \N after the mainHost String and mailFrom String. ****
>
> This causes the COPY statement to ERROR with :****
>
> ** **
>
> ERROR: missing data for column "f_value"****
>
> CONTEXT: COPY config_sysvars, line 2:
> "1649fe21-e806-4078-ab14-e7cf045c9265 2012-05-16
> 17:12:23.105 mailHost String"****
>
> ** **
>
> What am I doing wrong/misunderstood about using pg_dump.****
>
> ** **
>
> All advice gratefully received.****
>
> ** **
>
> ** **
>
> ** **
>
> ** **
>
> ** **
>

--
"Quality comes from focus and clarity of purpose" -- Mark Shuttleworth