aboutsummaryrefslogtreecommitdiff
path: root/libcrystfel/src/image.c
AgeCommit message (Collapse)Author
2020-02-11Don't use gzbuffer if it's not availableThomas White
Enables compatability with zlib < 1.2.3.5
2020-01-10Mass update of copyright datesThomas White
2019-09-10sort_peaks(): Handle NULL peak listThomas White
2019-07-17Merge branch 'tom/zmq'Thomas White
2019-06-05Fix obvious bug in image_add_feature()Thomas White
Introduced by aa1676f35317df92840b27ba78f13c13308bc7d4
2019-05-14Remove "valid" item from "struct imagefeature"Thomas White
2019-05-07Update all source code comments to Doxygen formatThomas White
2019-04-16Recognise (if not handle) all CBF formatsThomas White
2019-04-16Basic CBF reader with byte offset decompressionThomas White
2019-04-15Strip ouf CBFlib referencesThomas White
2019-04-03Consider pixel as bad if its value is NaN or infinityThomas White
2019-03-14Clean up MessagePack stuffThomas White
2019-01-03Use gzclose() instead of gzclose_r()Thomas White
The only downside to this is that the zlib library will be twice the size if statically linked into libcrystfel. Since we don't anticipate much use for static linking, and since the extra space is unlikely to be a serious problem anyway, this seems fine.
2018-11-02Fix possibly uninitialised dataf in read_cbf_data()Thomas White
2018-09-20Allow cbf.gz buffer to grow if the file is largeThomas White
2018-09-20Handle gzipped CBF filesThomas White
2018-09-20Stop reading "simply" if file type is unknownThomas White
2018-09-20Factorise CBF data readThomas White
2018-09-19Stop opening file if it's not recognisedThomas White
2018-09-19Handle 16bpp CBFs (as well as 32bpp)Thomas White
2017-09-27remove_flagged_crystals(): Remove multiple bad crystalsThomas White
A silly bug
2017-09-21Improve error message when using CBF files and a multi-event geometryThomas White
2017-09-07remove_flagged_crystals(): Return the number of crystals removedThomas White
2017-06-23Accept either cbf/cbf.h or cbflib/cbf.hThomas White
You get cbflib/cbf.h if you install CBFlib from source, and cbf/cbf.h if using the package manager in Fedora, probably other systems as well. We definitely don't want to include the cbf or cbflib folder in the include path, because CBFlib may bring its own HDF5 headers, which we don't want to touch with a bargepole.
2017-06-21Don't compile CBF stuff without CBFlibThomas White
2017-05-09Allow reading CBF without geometry (only for simple viewing)Thomas White
2017-05-09Add missing checksThomas White
2017-05-05Show filename of CBF file if it can't be readThomas White
2017-05-05Fill in photon energy, clen and adu for CBFsThomas White
This needed a bit of reorganisation and clarification of who is repsonsible for loading what. The low-level file loaders, e.g. hdf5_read and hdf5_read2 in hdf5-file.c or cbf_read in image.c, are responsible. There is a helper function adjust_centering_for_rail in detector.h which they can use. It seems like this could be done more cleanly at the imagefile layer. However, we need to make sure that the "old" hdfile API still works on its own, even when not accessed via the new imagefile API.
2017-05-05Read CBF dataThomas White
2017-05-03Detect CBF files, interface bitsThomas White
2017-05-03HDF5 reading under new APIThomas White
2017-05-02Skeleton image file APIThomas White
2017-05-02Initial CBF stuffThomas White
2017-03-13Ask for C99 in configure.ac, remove weird C99 declarationsThomas White
2016-07-01No more slab-relative coordinates in libcrystfelThomas White
2016-01-29Perform prediction refinement straight after indexingThomas White
This allows indexing to be attempted again (either a new method or with "retry") if the prediction refinement fails, increasing overall indexing rate. Side-effect: there are some hoops which would need to be jumped through to store the profile radius before refinement and hence enable scripts/plot-predict-refine to work. For now, we'll ignore this as it's clear that the prediction refinement is working.
2015-12-18"Retry" mechanism for indexingThomas White
This increases the indexing rate a bit in situations where there are lots of weak peaks which, although they may be real, don't help indexing. The weakest 10% of peaks get cut out and the indexing re-run. This also allows multiple hits to be indexed, using the "inelegant peak subtraction method", by retrying indexing in the same way after deleting peaks which are accounted for by the lattice just found.
2015-01-27FormattingValerio Mariani
2015-01-27Add image_reflection_closest()Valerio Mariani
2014-09-05Multi-event modeValerio Mariani
2014-07-24Fixes closest feature detection bugValerio Mariani
2014-01-21FussinessThomas White
2014-01-21Fixed memory leaks in free_all_crystals functionKenneth Beyerlein
2013-02-18Read GrainSpotter's output properlyThomas White
2013-02-05WIP on bringing programs up to dateThomas White
2012-03-12Update copyright notices to show actual contribution yearsThomas White
2012-03-09Update licence notices in line with recommendations from TTThomas White
2012-02-22Fix bounds of image feature number (spotted by Chuck)Thomas White
2012-02-22GPLv3 boilerplate commentsThomas White