Universal File Access Functions

Universal file access functions provide local access APIs for files on a database server. Only files in the database directory and the log_directory directory can be accessed. A relative path is used for files in the database directory, and a path matching the log_directory configuration setting is used for log files. Only database initialization users can use these functions.

• pg_ls_dir(dirname text)

  Description: Lists files in a directory.

  Return type: setof text

  Note: pg_ls_dir returns all the names in the specified directory, except the special entries "." and "..".

  Example:

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

  gaussdb=# SELECT pg_ls_dir('./'); pg_ls_dir ---------------------- .gaussdb.conf.swp gaussdb.conf pg_tblspc PG_VERSION gs_ident.conf core server.crt pg_serial pg_twophase gaussdb.conf.lock pg_stat_tmp pg_notify pg_subtrans pg_ctl.lock pg_xlog pg_clog base pg_snapshots postmaster.opts postmaster.pid server.key.rand server.key.cipher pg_multixact pg_errorinfo server.key gs_hba.conf pg_replslot .gs_hba.conf.swp cacert.pem gs_hba.conf.lock global gaussdb.state (32 rows)

• pg_read_file(filename text, offset bigint, length bigint)

  Description: Returns the content of a text file.

  Return type: text

  Note: pg_read_file returns part of a text file. It can return a maximum of length bytes from offset. The actual size of fetched data is less than length if the end of the file is reached first. If offset is negative, it is the length rolled back from the file end. If offset and length are omitted, the entire file is returned.

  Example:

  1 2 3 4 5 6 7 8 9 10 11

  gaussdb=# SELECT pg_read_file('postmaster.pid',0,100); pg_read_file --------------------------------------- 53078 + /srv/BigData/hadoop/data1/dbnode+ 1500022474 + 8000 + /var/run/FusionInsight + localhost + 2 (1 row)

• pg_read_binary_file(filename text [, offset bigint, length bigint,missing_ok boolean])

  Description: Returns the content of a binary file.

  Return type: bytea

  Note: pg_read_binary_file is similar to pg_read_file, except that the result is a bytea value; accordingly, no encoding checks are performed. In combination with the convert_from function, this function can be used to read a file in a specified encoding.

  1	gaussdb=# SELECT convert_from(pg_read_binary_file('filename'), 'UTF8');

• pg_read_binary_file_blocks(filename text, blocknum bigint, blockcount bigint)

  Description: Returns the binary content of a specified page area of a compressed file.

  Return type: bytea

  Note: pg_read_binary_file_blocks is used only to read the binary content of a compressed file.

• pg_stat_file(filename text)

  Description: Returns status information about a file.

  Return type: record

  Note: pg_stat_file returns a record containing the file size, last access timestamp, last modification timestamp, last file status change timestamp, and a Boolean value indicating if it is a directory. Typical usages are as follows:

  1	gaussdb=# SELECT * FROM pg_stat_file('filename');

  1	gaussdb=# SELECT (pg_stat_file('filename')).modification;

  Example:

  1 2 3 4 5 6 7 8 9 10 11 12

  gaussdb=# SELECT convert_from(pg_read_binary_file('postmaster.pid'), 'UTF8'); convert_from -------------------------------------- 4881 + /srv/BigData/gaussdb/data1/dbnode+ 1496308688 + 25108 + /opt/user/Bigdata/gaussdb/gaussdb_tmp + * + 25108001 43352069 + (1 row)

  1 2 3 4 5 6 7 8 9

  gaussdb=# SELECT * FROM pg_stat_file('postmaster.pid'); size | access | modification | change | creation | isdir ------+------------------------+------------------------+------------------------ +----------+------- 117 | 2017-06-05 11:06:34+08 | 2017-06-01 17:18:08+08 | 2017-06-01 17:18:08+08 | | f (1 row)

  1 2 3 4 5

  gaussdb=# SELECT (pg_stat_file('postmaster.pid')).modification; modification ------------------------ 2017-06-01 17:18:08+08 (1 row)