Skip to main content

Yaml data converter between DB and Repo

Project description

pg-data-yaml — Yaml interface for reference tables in PostgreSQL

Export, diff and sync rows of reference tables in PostgreSQL to YAML files in a repository.

installation

pip install pg-data-yaml

which tables are included

Table selection is configured with --comment-label or --table-list-predicate. One of these options is required for export, diff and sync (mutually exclusive).

--comment-label LABEL

Include tables whose comment contains LABEL. Any label can be used, for example global directory or env directory.

Optional clause in parentheses after the label is parsed from the comment (see marking tables below).

pg_data_yaml export -d mydb --out-dir /tmp/refs --comment-label "global directory"

--table-list-predicate PREDICATE

Include tables matching a SQL predicate. Table comments are ignored; export always uses select * from <schema>.<table> order by <primary key columns>.

The predicate is inserted into a query against pg_catalog.pg_class and pg_catalog.pg_namespace. Use n.nspname for schema name and c.relname for table name.

Example — tables without tenant columns:

pg_data_yaml export -d mydb --out-dir /tmp/refs --table-list-predicate "
not exists (
    select 1
      from information_schema.columns col
     where col.table_schema = n.nspname
       and col.table_name = c.relname
       and col.column_name in ('app_id', 'customer_id', 'customerid')
)"

Tables must have a primary key. Tables without a PK are skipped with a warning.

marking tables

When using --comment-label, add a comment on the table (via COMMENT ON TABLE):

synchronized directory

With a custom label:

global directory

Optional clause in parentheses:

  • if it contains the word select — a full export query, for example:
synchronized directory(select id, name from my_schema.my_table order by name)
  • otherwise — a WHERE filter inserted into the default template:
synchronized directory(not is_deleted)

select * from <schema>.<table> where not is_deleted order by <primary key columns>

If parentheses are omitted, export uses:

select * from <schema>.<table> order by <primary key columns>

Exported file layout: <out-dir>/<schema>/<table>.yaml

Each file is a YAML list of row mappings. Row and field order match the query result.

usage

export

usage: pg_data_yaml export [--help] [-d DBNAME] [-h HOST] [-p PORT] [-U USER] [-W PASSWORD]
                             (--comment-label LABEL | --table-list-predicate PREDICATE)
                             --out-dir OUT_DIR [--clean] [--rows-limit ROWS]

options:
  --help                show this help message and exit
  -d DBNAME, --dbname DBNAME
                        database name to connect to
  -h HOST, --host HOST  database server host or socket directory
  -p PORT, --port PORT  database server port
  -U USER, --user USER  database user name
  -W PASSWORD, --password PASSWORD
                        database user password
  --comment-label LABEL
                        include tables whose comment contains LABEL
  --table-list-predicate PREDICATE
                        include tables matching SQL PREDICATE (comments ignored)
  --out-dir OUT_DIR     directory for exporting files
  --clean               clean out_dir if not empty (env variable DATA_DIRECTORY_AUTOCLEAN=true)
  --rows-limit ROWS     skip tables with more than ROWS rows (default: 50000)

diff

usage: pg_data_yaml diff [--help] [-d DBNAME] [-h HOST] [-p PORT] [-U USER] [-W PASSWORD]
                           (--comment-label LABEL | --table-list-predicate PREDICATE)
                           --source SOURCE

options:
  --help                show this help message and exit
  -d DBNAME, --dbname DBNAME
  -h HOST, --host HOST  database server host or socket directory
  -p PORT, --port PORT  database server port
  -U USER, --user USER  database user name
  -W PASSWORD, --password PASSWORD
                        database user password
  --comment-label LABEL
                        include tables whose comment contains LABEL
  --table-list-predicate PREDICATE
                        include tables matching SQL PREDICATE (comments ignored)
  --source SOURCE       directory or yaml file to compare with the database

sync

usage: pg_data_yaml sync [--help] [-d DBNAME] [-h HOST] [-p PORT] [-U USER] [-W PASSWORD]
                           (--comment-label LABEL | --table-list-predicate PREDICATE)
                           --source SOURCE [--dry-run] [--echo-queries] [-y]
                           [--quiet] [--session-replication-role ROLE]

options:
  --help                show this help message and exit
  -d DBNAME, --dbname DBNAME
  -h HOST, --host HOST  database server host or socket directory
  -p PORT, --port PORT  database server port
  -U USER, --user USER  database user name
  -W PASSWORD, --password PASSWORD
                        database user password
  --comment-label LABEL
                        include tables whose comment contains LABEL
  --table-list-predicate PREDICATE
                        include tables matching SQL PREDICATE (comments ignored)
  --source SOURCE       directory or yaml file to sync to the database
  --dry-run             test run without real changes
  --echo-queries        echo commands sent to server
  -y, --yes             do not ask confirm
  --quiet               do not show yaml diff before applying changes
  --session-replication-role ROLE
                        set session_replication_role locally in transaction before DML

merge-envs

Move table files that are identical in all given environment directories into a shared base directory and remove them from the environment directories.

usage: pg_data_yaml merge-envs [--help] --source ENV_DIR [--source ENV_DIR ...] --out-dir OUT_DIR [--dry-run]

options:
  --help                show this help message and exit
  --source ENV_DIR      environment directory (repeat for each env)
  --out-dir OUT_DIR     base directory for common table files
  --dry-run             show actions without changing files

Example layout after merge:

refs/
  base/public/countries.yaml   # identical in all envs
  dev/public/special.yaml      # env-specific
  prod/public/special.yaml

examples

Comment label for synchronized reference data:

$ pg_data_yaml export -d my_database -h 127.0.0.1 -p 5432 -U postgres \
    --out-dir /tmp/refs/ --comment-label "synchronized directory"

Comment label for per-environment data:

$ pg_data_yaml export -d my_database --out-dir /tmp/refs/base --comment-label "global directory"
$ pg_data_yaml export -d my_database --out-dir /tmp/refs/dev --comment-label "env directory"

Predicate-based selection:

$ pg_data_yaml export -d my_database --out-dir /tmp/refs/ --table-list-predicate "
not exists (
    select 1 from information_schema.columns col
     where col.table_schema = n.nspname
       and col.table_name = c.relname
       and col.column_name in ('app_id', 'customer_id', 'customerid')
)"

Diff and sync use the same table selection options as export:

$ pg_data_yaml merge-envs --source /tmp/refs/dev --source /tmp/refs/prod --out-dir /tmp/refs/base
$ pg_data_yaml diff -d my_database -h 127.0.0.1 -p 5432 -U postgres --source /tmp/refs/ --comment-label "global directory"
$ pg_data_yaml sync -d my_database -h 127.0.0.1 -p 5432 -U postgres \
    --source /tmp/refs/ --comment-label "synchronized directory"
$ pg_data_yaml diff -d my_database -h 127.0.0.1 -p 5432 -U postgres --source /tmp/refs/public/countries.yaml

When syncing a directory, only tables that are both in the selected set and have a yaml file in --source are compared. A warning is printed and the table is skipped when the file exists but the table is not in the selection, or when the table is in the selection but the yaml file is missing. When syncing a single file, only that table is compared and updated if it is in the selection. Each table is synced in its own transaction, so a failed table does not leave other tables half-updated.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pg_data_yaml-0.1.0.tar.gz (22.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pg_data_yaml-0.1.0-py3-none-any.whl (24.5 kB view details)

Uploaded Python 3

File details

Details for the file pg_data_yaml-0.1.0.tar.gz.

File metadata

  • Download URL: pg_data_yaml-0.1.0.tar.gz
  • Upload date:
  • Size: 22.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.17

File hashes

Hashes for pg_data_yaml-0.1.0.tar.gz
Algorithm Hash digest
SHA256 7e5c7623afb8f0cca07d76aeed918d3c580ec5e1c6a9af56dee3d9f5258876f2
MD5 056a8d5ca435e322d07b6bf0a4b0158c
BLAKE2b-256 dc34fd35463bef10028d4900023c8fb07d23d1e4c1dc75cf258d211e1cadf602

See more details on using hashes here.

File details

Details for the file pg_data_yaml-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: pg_data_yaml-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 24.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.17

File hashes

Hashes for pg_data_yaml-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 158a16fb908c901304b9690067d7151bc38fa7eff10c264e9023ac7de55a36ad
MD5 aa47e7a8986ed91bd05efdc02f343f67
BLAKE2b-256 98717a43a90fa77217d246b24643082e065b103269eb4b96f815e766ea95956e

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page