taucmdr.cf.storage.local_file module

Local file backend for storage containers.

A persistent, transactional record storage system using TinyDB for both the database and the key/value store.

class taucmdr.cf.storage.local_file.LocalFileStorage(name, prefix, create_prefix=True)[source]

Bases: AbstractStorage

A persistent, transactional record storage system.

Uses TinyDB for both the database and the key/value store.

Record

alias of _JsonRecord

__enter__()[source]

Initiates the database transaction.

__exit__(ex_type, value, traceback)[source]

Finalizes the database transaction.

__str__()[source]

Human-readable identifier for this database.

_abc_impl = <_abc._abc_data object>
_make_prefix()[source]

Create the prefix if it’s missing and this storage may create it.

static _query(keys, match_any)[source]

Construct a TinyDB query object.

connect_database(*args, **kwargs)[source]

Open the database for reading and writing, or only for reading if it can’t be written.

connect_filesystem(*args, **kwargs)[source]

Prepares the store filesystem for reading and writing.

contains(keys, table_name=None, match_any=False)[source]

Check if the specified table contains at least one matching record.

The behavior depends on the type of keys:

  • self.Record.eid_type: check for the record with that element identifier.

  • dict: check for the record with attributes matching keys.

  • list or tuple: return the equivalent of map(contains, keys).

  • None: return False.

Parameters:

  • keys – Fields or element identifiers to match.

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • match_any (bool) – Only applies if keys is a dictionary. If True then any key in keys may match or if False then all keys in keys must match.

Returns:

True if the table contains at least one matching record, False otherwise.

Return type:

bool

Raises:

ValueError – Invalid value for keys.

count(table_name=None)[source]

Count the records in the database.

Parameters:

table_name (str) – Name of the table to operate on. See AbstractStorage.table.

Returns:

Number of records in the table.

Return type:

int

database_exists()[source]

Determine if the database file backing this Storage object exists.

Returns:

True if database exists; false otherwise.

Return type:

bool

property dbfile

Absolute path to the database file.

Type:

str

disconnect_database(*args, **kwargs)[source]

Close the database for reading and writing.

disconnect_filesystem(*args, **kwargs)[source]

Disconnects the store filesystem.

get(keys, table_name=None, match_any=False)[source]

Find a single record.

The behavior depends on the type of keys:

  • self.Record.eid_type: return the record with that element identifier.

  • dict: return the record with attributes matching keys.

  • list or tuple: return a list of records matching the elements of keys

  • None: return None.

Parameters:

  • keys – Fields or element identifiers to match.

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • match_any (bool) – Only applies if keys is a dictionary. If True then any key in keys may match or if False then all keys in keys must match.

Returns:

The matching data record if keys was a self.Record.eid_type or dict. list: All matching data records if keys was a list or tuple. None: No record found or bool(keys) == False.

Return type:

Record

Raises:

ValueError – Invalid value for keys.

insert(data, table_name=None)[source]

Create a new record.

If the table doesn’t exist it will be created.

Parameters:
Returns:

The new record.

Return type:

Record

is_writable()[source]

Check if the storage filesystem is writable.

Attempts to create and delete a file in prefix. See https://github.com/ParaToolsInc/taucmdr/issues/231.

A prefix that doesn’t exist is created first, unless this storage was told not to create it.

Returns:

True if a file could be created and deleted in prefix, False otherwise.

Return type:

bool

items()[source]

Iterate over items in the key/value store.

keys()[source]

Iterate over keys in the key/value store.

match(field, table_name=None, regex=None, test=None)[source]

Find records where field matches regex or test.

Either regex or test may be specified, not both. If regex is given, then all records with field matching the regular expression are returned. If test is given then all records with field set to a value that causes test to return True are returned. If neither is given, return all records where field is set to any value.

Parameters:

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • field (string) – Name of the data field to match.

  • regex (string) – Regular expression string.

  • test – Callable returning a boolean value.

Returns:

Matching data records.

Return type:

list

Raises:

ValueError – Invalid value for keys.

property prefix

Get the filesystem prefix for file storage.

The filesystem must be persistent and provide the usual POSIX filesystem calls. In particular, GNU software packages should be installable in the filesystem.

Returns:

Absolute path in the filesystem.

Return type:

str

purge(table_name=None)[source]

Delete all records.

Parameters:

table_name (str) – Name of the table to operate on. See AbstractStorage.table.

remove(keys, table_name=None, match_any=False)[source]

Delete records.

The behavior depends on the type of keys:

  • self.Record.eid_type: delete the record with that element identifier.

  • dict: delete all records with attributes matching keys.

  • list or tuple: delete all records matching the elements of keys.

Parameters:

  • keys – Fields or element identifiers to match.

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • match_any (bool) – Only applies if keys is a dictionary. If True then any key in keys may match or if False then all keys in keys must match.

Raises:

ValueError – bool(keys) == False or invalid value for keys.

search(keys=None, table_name=None, match_any=False)[source]

Find multiple records.

The behavior depends on the type of keys:

  • self.Record.eid_type: return the record with that element identifier.

  • dict: return all records with attributes matching keys.

  • list or tuple: return a list of records matching the elements of keys

  • None: return all records.

Parameters:

  • keys – Fields or element identifiers to match.

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • match_any (bool) – Only applies if keys is a dictionary. If True then any key in keys may match or if False then all keys in keys must match.

Returns:

Matching data records.

Return type:

list

Raises:

ValueError – Invalid value for keys.

table(table_name)[source]

Return a handle to a table.

Return a handle to the named table or, if table_name is None, return the default table.

Parameters:

table_name (str) – Name of the table or None.

Returns:

A database table object.

Return type:

object

unset(fields, keys, table_name=None, match_any=False)[source]

Update records by unsetting fields.

Update only allows you to update a record by adding new fields or overwriting existing fields. Use this method to remove a field from the record.

The behavior depends on the type of keys:

  • self.Record.eid_type: update the record with that element identifier.

  • dict: update all records with attributes matching keys.

  • list or tuple: apply update to all records matching the elements of keys.

Parameters:

  • fields (list) – Names of fields to remove from matching records.

  • keys – Fields or element identifiers to match.

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • match_any (bool) – Only applies if keys is a dictionary. If True then any key in keys may match or if False then all keys in keys must match.

Raises:

ValueError – bool(keys) == False or invalid value for keys.

update(fields, keys, table_name=None, match_any=False)[source]

Update records.

The behavior depends on the type of keys:

  • self.Record.eid_type: update the record with that element identifier.

  • dict: update all records with attributes matching keys.

  • list or tuple: apply update to all records matching the elements of keys.

Parameters:

  • fields (dict) – Data to record.

  • keys – Fields or element identifiers to match.

  • table_name (str) – Name of the table to operate on. See AbstractStorage.table.

  • match_any (bool) – Only applies if keys is a dictionary. If True then any key in keys may match or if False then all keys in keys must match.

Raises:

ValueError – bool(keys) == False or invalid value for keys.

values()[source]

Iterate over values in the key/value store.

class taucmdr.cf.storage.local_file._CachingMiddleware(storage_cls)[source]

Bases: CachingMiddleware

Write back only if something changed, so that closing a read-only database isn’t an error.

flush()[source]

Flush all unwritten data to disk.

class taucmdr.cf.storage.local_file._JsonFileStorage(path)[source]

Bases: JSONStorage

Allow read-only as well as read-write access to the JSON file.

TinyDB’s default storage (tinydb.JSONStorage) assumes write access to the JSON file. This isn’t the case for system-level storage and possibly others. A file that doesn’t exist and can’t be created is treated as an empty, read-only database.

_abc_impl = <_abc._abc_data object>
read()[source]

Read the last stored state.

Any kind of deserialization should go here. Raise ValueError here to indicate that the storage is empty.

Return type:

dict

write(*args, **kwargs)[source]

Write the current state of the database to the storage.

Any kind of serialization should go here.

Parameters:

data (dict) – The current state of the database.

class taucmdr.cf.storage.local_file._JsonRecord(database, element, eid=None)[source]

Bases: StorageRecord

eid_type

alias of int