donz0r's blog

The class DatabaseEntry now has a new classmethod: from_fits_filepath. It receives a path to a FITS filename and parses its header to make a new instance of DatabaseEntry. All key-value pairs can be found in the attribute fits_header_entries, which is a list of FitsHeaderEntry instances. Additionally, this classmethod searches for the keys INSTRUME, WAVELNTH, DATE-OBS / DATE_OBS, and DATE-END / DATE_END. Their values are used to set the attribute instrument, wavemin and wavemax, observation_time_start or observation_time_end, respectively.

>>> from sunpy.database import DatabaseEntry
>>> entry = DatabaseEntry.from_fits_filepath('sunpy/data/efz20010101.143610.fits')
>>> entry.instrument
'EIT'
>>> entry.wavemin, entry.wavemax
(195.0, 195.0)
>>> entry.observation_time_start, entry.observation_time_end
(datetime.datetime(2001, 1, 1, 14, 36, 10, 983000), None)
>>> for header_entry in entry.fits_header_entries:
...     if header_entry.key in ('INSTRUME', 'WAVELNTH', 'DATE-OBS'):
...         print header_entry
...
<FitsHeaderEntry(id None, key 'DATE-OBS', value '2001-01-01T14:36:10.983')>
<FitsHeaderEntry(id None, key 'INSTRUME', value 'EIT')>
<FitsHeaderEntry(id None, key 'WAVELNTH', value 195)>

There is also a new utility function called entries_from_path to get an iterator of DatabaseEntry instances by passing a directory path. This function also accepts the optional arguments recursive and pattern. The former argument is boolean and determines whether to search the directory recursively or not. The latter one is a string and defines how a FITS path is detected by its filename. This value is passed to fnmatch.filter and its default is *.fits. Again, the values of the attributes instrument, wavemin, wavemax, observation_time_start, and observation_time_end are set, if possible.

>>> from sunpy.database import entries_from_path
>>> from sunpy.data.test import rootdir as sampledir
>>> entries = list(entries_from_path(os.path.join(sampledir, 'EIT')))
>>> len(entries)
13
>>> first_entry, path = entries[0]
>>> first_entry.instrument
'EIT'
>>> first_entry.wavemin, first_entry.wavemax
(195.0, 195.0)
>>> first_entry.observation_time_start, first_entry.observation_time_end
(datetime.datetime(2004, 3, 1, 2, 0, 10, 642000), None)