artifact_id;status_id;status_name;priority;submitter_id;submitter_name;assigned_to_id;assigned_to_name;open_date;close_date;last_modified_date;summary;details;"Category";"Group";"Category";"Group"
1011149;1;"Open";3;438241;"Paul Hardy";100;"Nobody";"2012-01-23 18:28";"";"2012-09-24 02:14";"Hex and Quote characters allowed in control file?";"I have no problem loading a table using pg_bulkload. Eg.
 
OUTPUT = schema.tablename                                               # [<schema_name>.]table_name
INPUT = /filelocation/table.csv                                   # Input data location (absolute path)
TYPE = CSV                                                        # Input file type
QUOTE = ""\""""                                                      # Quoting character
ESCAPE = \                                                        # Escape character for Quoting
DELIMITER = ~   
 
However, we have two problems I cannot seem to overcome or get the correct setup… that is, our schema names contain a # sign in them therefore if you want to reference this you need double quotes.
 
“schema#1”.tablename
 
I cannot seem to get the ctl file to accept this name because of the quotes.
 
The other problem is we use hex 7F as our delemeter for all our files and I cannot seem to overcome that problem as well.
 
Note that I can load this file no problem with \copy command, but we would like to test the difference between the pg_loader and \copy for large volume files. 
 
Eg. Of \copy command….
#!/bin/ksh
. /source statement
 
hex7f=""E'\x7F'""
schema='""schema#1""'
table=""tablename""
database=""db""
 
 
starttime=$(psql -d $database -tc ""select now();"")
echo $starttime
psql -c ""truncate $schema.$table;"" -d $database
psql -c ""\copy $schema.$table FROM /filelocation/$table.csv DELIMITER AS $hex7f NULL AS ''"" -d $database
 
this will successfully load a table with schema name of # in it with hex 7F as the delimeter.

Anyone out there run across this? have a solution?";"None";"None";"None";"None"
1011150;1;"Open";3;438370;"christophe CERQUEIRA";118405;"Mitsuru Hasegawa";"2012-01-24 18:29";"";"2012-09-24 02:14";"I can't compil pg_bulkload 3.1.1 on AIX Systems 5.3";"I can't compil the .so in lib Directories.

in lib subdirectories the files .o are good but i c'ant link correctly because i have an error

My command : 

/usr/bin/ld -o pg_bulkload.so  pg_bulkload.o  logger.o binary.o parser_csv.o parser_binary.o parser_tuple.o parser_tuple.o parser_function.o pg_btree.o reader.o pg_strutil.o writer.o source.o writer_direct.o writer_buffered.o writer_binary.o writer_parallel.o libbsd.a pgut/pgut-ipc.o pgut/pgut-be.o -L/usr/lib -lbsd  -lcrypt -lpthread -lpthreads -bnoentry -bexpall -bnoglink -L../../../src/interfaces/libpq -lpq -bI:../../../src/backend/postgres.imp  -bnoquiet -bM:SRE -lc

the result : 

ld: 0711-318 ERROR: Undefined symbols were found.
        The following symbols are in error:
 Symbol                    Inpndx  TY CL Source-File(Object-File) OR Import-File{Shared-object}
                              RLD: Address  Section  Rld-type Referencing Symbol
 ----------------------------------------------------------------------------------------------
 .flock                    [36]    ER PR logger.c(logger.o)
                                   0000062c .text    R_RBR    [74]    .LoggerLog
                                   000007fc .text    R_RBR    [74]    .LoggerLog
ER: The return code is 8.


i don't understand why the flock librairies ar not present because i put the directive -lbsd on the link command.
Thank you for you quicly response...
Chris




 ";"Install Problem (example)";"None";"Install Problem (example)";"None"