file_store

This is the oldest code in Antelope (and among my first ever pieces of python code). Please don't judge.

Written to access ILCD and ecospold archives in either zipped or unzipped or 7z format and GaBi web databases equally

Unfortunately, the 7z format has some kind of bug where time per read grows linearly with the number of reads. I should report it.

FileStore

Bases: object

A strictly local archive of files to be used as a repository

__init__(path, internal_prefix=None, query_string=None, cache=True)

Create a FileStore object from a path. Basically encapsulates the compression algorithm and presents a common interface to client code: self.listfiles() self.countfiles() self.readfile(filename)

writing providers is not presently supported.

By default remote archives create a local cache to store downloaded files- this cache is mounted as an ordinary archive and is checked first when readfile() is called. You can force a re-download by specifying .readfile(..., force_download=True)

Parameters:

Name	Type	Description	Default
path			required
internal_prefix		if present, silently absorb / conceal the specified prefix (prepend to requests, trim from responses).	None
query_string		for remote repositories, append the supplied string after '?' in the URL	None
cache		(True) for remote repositories, cache downloaded files locally and use first	True

Returns:

Type	Description
	an archive object

listfiles(in_prefix='')

generate files in the archive, removing internal prefix, optionally filtering to specified prefix.

Parameters:

Name	Type	Description	Default
in_prefix		optional prefix to limit	''

Returns:

Type	Description

readfile(fname, force_download=False)

Have to decide what this does. I think it should return the raw data- since there's no way to get a pointer to a file in a generic archive

Parameters:

Name	Type	Description	Default
fname			required
force_download		for remote caching archives:	False

Returns:

Type	Description