Open Hub
Open Hub

 Settings |  Report Duplicate

2
I Use This!
Low Activity

Commits : Listings

Analyzed about 14 hours ago. based on code collected about 14 hours ago.
Oct 20, 2024 — Oct 20, 2025
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
fixed typo
Maxim Yurkin
More... about 4 years ago
Fixes issue 156 and other improvements of `misc/pip`: - the inside-test is performed separately for each material; tighter bounds for winding number (normalized solid angle) around 0 or +-1 are enforced - adds recognition of duplicate materials in `obkj_read` and `wrml_read` - ignores material fields in `wrml_read` completely (were not used anyway); default material is also now not defined there - improves robustness of handling a lot of materials (> material_max) in `oogl_read` and `smf_read` - adds option to ignore material information (at compile time) - removes duplication of some output (and source file `limits_cor3.f90`) - removes a lot of unused variables and function arguments - makes uniform indentation - adds `implicit none` everywhere - fixed handling of default materials in `data_check` and `data_init` - removes subroutine `interact`
Maxim Yurkin
More... about 4 years ago
Added support for quadrilateral surface patches in `misc/pip`. Related changes: - `order_max` (or `face_order_max`) and `face_order` are now passed into (and returned from) `ivread_wr` - the maximum order of face polygons (in all code) is now controlled by `face_order_max` in `FEM-Geo-Wr.f90`
Maxim Yurkin
More... about 4 years ago
The interface routine for `misc/pip` has been reworked. - previously the required box sizes for two smaller dimensions were determined incorrectly (due to incorrect integer rounding), but this was compensated (for grids>10) by inflating the dipole grid by a factor 1.1. Effectively, this implied a smaller grid (by 10%) but the overall shape was determined correctly. This is now fixed, and the grid is centered inside the particle (its exact bounding box) - the same as is done in ADDA. - also a lot of unused variables has been removed. - the output has been made more robust (potentially ready for rectangular dipoles), and more in flavor of ADDA. Now the dipole integer coordinates go from 0 to nx-1 instead of from -mx (or 1-mx) to mx. But the format is still that of DDSCAT. - `shape_test.dat` has been replaced by the updated output (contains significantly more dipoles)
Maxim Yurkin
More... about 4 years ago
Superellipsoid (#298)
Jerome Fung
More... about 4 years ago
Minor style corrections
Maxim Yurkin
More... over 4 years ago
Updates tests to account for new '-shape coated2'
Maxim Yurkin
More... over 4 years ago
new shape coated2 (#291)
Mercede-hash
More... over 4 years ago
- adds a workaround for #286 - the workaround for #146 is now operational only for Intel compilers v.11.x - fixes typos in iw_compile.bat - removes some debugging output for FFTW when using Microsoft (Visual Studio) linker
Maxim Yurkin
More... over 4 years ago
- updates intel compilation flag to the recent version (based on tests on linux) - adds a section intelX for new Intel oneAPI compilers - gcc full optimization flag slightly updated - iw_compile.bat can now switch between debug and fully optimized regimes - minor changes to remove several compiler warnings
Maxim Yurkin
More... over 4 years ago
- enables compilation of ADDA (seq mode) with latest Intel compilers (both Classic and oneAPI) on Windows (with separate iw_compile.bat) - updates detection of compilers in param.c - fixes several compiler warnings (mainly related to silent conversion between doublecomplex and double) - slightly changes the simulation result for combination of `-scat so -surf ... -rect_dip ...` (not detected by existing tests) - updates gcc warnings in main Makefile - restructures description of files in src/README.md - now mostly in alphabetical order (as in the file list on GitHub)
Maxim Yurkin
More... over 4 years ago
Added couple of calls to clFinish, should help with #288
Maxim Yurkin
More... over 4 years ago
Adds debug calls to oclunload() - tries to help with #288
Maxim Yurkin
More... over 4 years ago
Updates comments with respect to sh_form_str1(2). In response to #287
Maxim Yurkin
More... over 4 years ago
- function pointers are now declared as extern in interaction.h (main definition is now in interactio.c). - removed duplicate declaration of Xmatrix in vars.c/h This fixes #284 - compilation errors with gcc version 10 and higher.
Maxim Yurkin
More... over 4 years ago
Tests updated to compare against v1.4.0. In particular, odd grid values added
Maxim Yurkin
More... almost 5 years ago
Fixes compiler warnings with clang
Maxim Yurkin
More... almost 5 years ago
Add includes to fix compilation errors with PRECISE_TIMING on Unix
Maxim Yurkin
More... almost 5 years ago
Merge branch 'master' of github.com:adda-team/adda
Maxim Yurkin
More... almost 5 years ago
- Updated release_sequence.txt and related scripts - Fixed a few typos and -x attributes
Maxim Yurkin
More... almost 5 years ago
Finally removed mention of migration from Google Code
Maxim Yurkin
More... almost 5 years ago
A lot of changes to build and packaging scripts. Fixes #210 (finalizes).
Maxim Yurkin
More... almost 5 years ago
Preparing files for release v1.4.0: Windows binaries and sample output
Maxim Yurkin
More... almost 5 years ago
fixes possibility to compile from remote folder
Maxim Yurkin
More... almost 5 years ago
Preparing file for release v1.4.0: docs and const.h
Maxim Yurkin
More... almost 5 years ago
minor changes to fix compiler warnings
Maxim Yurkin
More... almost 5 years ago
- New command line option: -so_buf {no|line|full} for manually specifying the buffering of stdout. Fixes #277. - On Widows line buffering and default POSIX behavior (when command line is not used) is emulated. Most printf calls are replaced with macros. - Removed 6 dots in stdout after the "Fourier transform of Dmatrix" (or Rmatrix). They were not properly buffered in many cases. - Removed redundant flushing of stderr. Instead, fflush(NULL) was added before output to stderr (to better position it with respect to other output. - 'include <windows.h>' was moved from os.h (to make the latter lightweight) into param.c and timing.h (only there it is really needed).
Maxim Yurkin
More... almost 5 years ago
- Updated help strings according to existing incompatibilities of -rect_dip - Ignore patterns in comp2exec updated accordingly
Maxim Yurkin
More... almost 5 years ago
minor change in comments
Maxim Yurkin
More... almost 5 years ago
Enhancement of rectangular dipoles. Fixes #255. - gridSpaceX,gridSpaceY,gridSpaceZ refactored into dsX,dsY,dsZ - introduced kdX,kdY,kdZ - gridspace and kD explicitly marked as deprecated, they equal DBL_MAX when rect_dip is invoked - dpl is now defined with respect to the largest dipole dimension (e.g. lambda/dsMax instead of lambda/dsX) - a few help strings updated (to mention rect_dip) and typos fixed - added explicit test for incompatibility of rect_dip with '-scat so' - added several more incompatibility tests deeper in the code (should never be reached) - added output of dipole size to log; it is different for rectangular and cubical dipoles - removed mention of "along the greatest dimension" for information of IGT limit, when -rect_dip is used (assume that dipole size is always the largest) - ZsumShift is now in real units (was in units of kdZ) - slightly changed the text for "lattice spacing" and "coordinates of the zero dipole" in headers of DDSCAT shape files - updated logic to get default values of dpl, size, etc. when they are not given in the command line. This fixes a couple of rare bugs when -rect_dip is used - fixed a bug in '-init_field wkb' in combination with -rect_dip - interaction.c: - functions no more accept unitsGrid argument; UnitsGridToCoord and UnitsGridToCoordShift now transform integer vectors to real distance - ReflParams removed, its calls replaced by InterParams. The latter no more calculates rn - code for invoking IGT routines simplified - updated instructions for adding new interaction or reflection formulations - updated '-scat igt_so' to work fine with -rect_dip. Although, it is now not always perfectly accurate for Cext - RD_TRICKY test updated: now -size 5 is automatically added to most of the tests. This allowed many tests in suite_rd to be reverted to the default one (removed some special cases for RD_TRICKY)
Maxim Yurkin
More... almost 5 years ago