diffractem.stream_parser module

class diffractem.stream_parser.StreamParser(filename, parse_now=True, serial_offset=- 1, new_folder=None)

Bases: object

property cell

cell section as dictionary

Type

return

change_path(new_folder=None, old_pattern=None, new_pattern=None)

property files

property geometry

geometry section as dictionary

Type

return

get_cxi_format(what='peaks', shots=None, half_pixel_shift=True)

property indexed

property input_file

property num_crystals

property num_shots

property options

crystfel call options (ONLY – ones) as dict

Type

return

parse(new_folder)

property peaks

property shots

write(filename, include_peaks=True, include_indexed=True, include_geom=True, include_cell=True)

diffractem.stream_parser.augment_stream(streamname, outfile, new_fields, where='chunk')

Add new fields to chunk headers in the stream file, which can then be used for chopping or filtering. Somewhat similar to indexamajig’s “include-hdf5-field” option, just after the fact.

Parameters

• streamname (str) – Name of stream file

• new_fields (pd.DataFrame) – pandas DataFrame with index matching the file and Event of the stream file and columns matching the additional fields to be added

diffractem.stream_parser.chop_stream(streamname, id_list, id_field='hdf5/%/shots/frame', id_suffix='frame', fn_contains=None)

Chops a stream file into sub-streams containing only shots with a specific value of a defined field, which must be in the chunk header. Useful e.g. for chopping into aggregation frames, different sample grids, runs with different rotation angles etc.

If you just want to select a sub-set of a stream file instead of chopping it up into many parts, consider using the stream_grep script included with CrystFEL, which is way faster and more flexible.

Parameters

• streamname (str) – Stream file name

• id_list (str) – List of values of the ID variable which you want to have in the final files.

• id_field (str) – Field in chunk data to select by. Defaults to ‘hdf5/%/shots/frame’.

• id_appendix (str) – Appendix to be applied to the output stream file names. Defaults to ‘frame’.

diffractem.stream_parser.make_substream(stream, Ncryst, seed=None, filename=None, query=None)

Write a stream file containing a sub-set of events to a new stream file.

Parameters

• stream (StreamParser) – StreamParser object holding the original stream

• Ncryst (int) – Number of events to sample

• seed (Optional[int], optional) – Seed of the random generator. Defaults to None.

• filename (Optional[str], optional) – Output stream filename. Defaults to filename of original stream, with ‘-N_{Ncryst}’ appended, where {Ncryst} is the number of sampled crystals.

• query (str, optional) – query to pre-select events. A sensible choice might be to only pick indexed events by setting query=’indexed_by != “none”’. Defaults to None.

Returns

[description]

Return type

[type]

diffractem.stream_parser.parse_str_val(input)