Age | Commit message (Collapse) | Author | |
---|---|---|---|
2020-07-29 | Add crystfel_has_peakfinder9() | Thomas White | |
separate config.h file. | |||
2020-07-29 | Create detgeom and data arrays when reading from stream | Thomas White | |
2020-07-29 | Don't double-count last chunk | Thomas White | |
2020-07-29 | (Temporarily) disable pinkIndexer | Thomas White | |
2020-07-29 | Fix peak list loading from stream | Thomas White | |
I don't think this functionality has ever actually been used. It was completely broken. | |||
2020-07-29 | data_template_free: Allow NULL | Thomas White | |
2020-07-29 | Random access Stream API | Thomas White | |
2020-07-29 | Update comments | Thomas White | |
2020-07-29 | Set a default data location | Thomas White | |
2020-07-29 | stream_close: Handle NULL | Thomas White | |
2020-07-29 | Add missing prototype | Thomas White | |
2020-07-29 | Avoid segfault if there are no panels in DataTemplate | Thomas White | |
Obviously, this should never happen. | |||
2020-07-29 | Reject geometry file if it contains unrecognised lines | Thomas White | |
2020-07-29 | Recognise .cxi extension as HDF5 | Thomas White | |
2020-07-29 | Improve error message | Thomas White | |
2020-07-29 | is_hdf5_file: Only check extension | Thomas White | |
It seems that H5Fis_hdf5 doesn't close the file. So, we rapidly run out of file descriptors. Also, opening each file just to enumerate single-frame files seems like a waste of time. | |||
2020-07-29 | Avoid event enumeration if only one event is possible | Thomas White | |
The enumeration works in this case, of course, but is a waste of a not insignificant amount of time. | |||
2020-07-29 | Factorise num_path_placeholders | Thomas White | |
2020-07-29 | Improve error message when peak list is not given in DataTemplate | Thomas White | |
2020-07-29 | Make sure all HDF5 resources are freed | Thomas White | |
2020-07-29 | Improve error message | Thomas White | |
2020-07-29 | Set default dimensions (0=ss, 1=fs) | Thomas White | |
2020-07-29 | Handle event expansion even for single-event files | Thomas White | |
2020-07-29 | render_scale -> colscale_lookup | Thomas White | |
2020-07-29 | Break render.c down and move to geoptimiser and CrystFELImageView | Thomas White | |
These two functions will diverge in the future. Also, this change removes the last remaining dependency of libcrystfel on Gdk/Gtk-anything. | |||
2020-07-29 | Remove dtempl->defaults | Thomas White | |
This is 100% private to the parser | |||
2020-07-29 | Add crystal_copy_deep() | Thomas White | |
2020-07-29 | Implement dimension part of event expansion, and clean up tests | Thomas White | |
2020-07-29 | Path part of event expansion | Thomas White | |
2020-07-29 | Require mask/satmap paths to have fewer placeholders than data | Thomas White | |
2020-07-29 | Check that paths for all panels have the same number of placeholders | Thomas White | |
2020-07-29 | Fussiness | Thomas White | |
2020-07-29 | Fix image_hdf5_read_peaks_cxi | Thomas White | |
2020-07-29 | image_hdf5_get_value without 'struct event' | Thomas White | |
2020-07-29 | substitute_path: Special case for no placeholders | Thomas White | |
2020-07-29 | Remove events.c, part 2 | Thomas White | |
New event parsing, done at the last minute in image-hdf5.c | |||
2020-07-29 | Remove peak_info_location | Thomas White | |
peak_list already exists | |||
2020-07-29 | Remove events.c, part 1 | Thomas White | |
2020-07-29 | Set up adu_per_photon correctly | Thomas White | |
2020-07-29 | Units for wavelength/photon energy | Thomas White | |
2020-07-29 | Fix some memory bugs | Thomas White | |
2020-07-29 | Remove unused includes | Thomas White | |
2020-07-29 | Finish converting stream.c | Thomas White | |
2020-07-29 | Implement camera length units in data_template_to_detgeom | Thomas White | |
This is a special reduced version which cannot load metadata from an file, because no file is available in this case. | |||
2020-07-29 | Convert image-msgpack.c to DataTemplate/detgeom | Thomas White | |
2020-07-29 | Tidy up | Thomas White | |
Regarding check_path_existence: This was a very complicated procedure, for not very much gain. The attempt to open a dataset for real, which invariably follows, can still fail after this check. | |||
2020-07-29 | Convert predict-refine to detgeom | Thomas White | |
2020-07-29 | Update stream to new API | Thomas White | |
2020-07-29 | Const-cleaning | Thomas White | |
2020-07-29 | Add data_template_panel_name_to_number and rename data_template_panel_name | Thomas White | |