Merge pull request #1238 from ToLeWeiss/exhaustive_fragmenter
fixed missing value
added exception to mitigate the indexing of the fragmentation result map to overflow
adjusted maximum tree depth in the comments to 27
Merge remote-tracking branch 'upstream/main' into exhaustive_fragmenter
I finally read the documentation of git cherry-pick and therefore applied the changes from BigInt back to int manually
cherry picked hopefully the right commit commit:d6f52e8 to revert to integers instead of BigInt
cherry picked commit: 71da5e6e4d to revert to integers instead of BigInt
switching to big integers to lift the upper limit of 31 splittable Bonds per fragmentation
made param comments more consistent and beginning with lowercase
added documentation for custom assertion
Merge branch 'main' into exhaustive_fragmenter
fixed typo
improved comments and code aesthetic for stereo information copying
fixed documentation links, added constructor for just saturation and added a setting for copying setereo information + tests
fixed regression in test and fixed wrong maximum tree depth
implemented stereo chemistry copying, a respective test and added missing documentation
implemented test for double bonds, cleaned up code and tests
improved comments and removed dependencies in the test class
removed useless comments and imports and added documentation if it is not possible to generate any fragments
Merge branch 'main' into exhaustive_fragmenter
added null check for input molecule and elaborated on comments
implemented comments and improved ring detection and aromaticity copying
WIP fixing issues in documentation and conformities in the code
fixed docs and corrected calculation for HashSet capacity
small test documentation fix
improved tests and documentation
refactored splitting function to improve saturation and improved documentation
Merge remote-tracking branch 'origin/main' into exhaustive_fragmenter
work in progress: added tests for pseudo-R-atom saturation and first tries to implement the pseudo-R-atom saturation
work in progress: first tries to add an saturation option with R-Atoms
Merge branch 'main' into exhaustive_fragmenter
work in progress: further improvements to comments and more private fields
work in progress: refining comments
work in progress: adding a setter for the maximum tree depth
work in progress: cleaning up imports
work in progress: cleaning up and adding documentation
work in progress: working reimplementation of the exhaustive fragmenter
work in progress: fixing for end atoms not in container
work in progress: figuring out why the end atoms are in a different container
work in progress: fix size limitation of the subset generation
work in progress: switch to primitive int matrices for subset generation and added more comprehensive comments for subset generation
work in progress: fixed indexing and fix of check for size constraint
switching to primitive integers because java Lists are indexed by these and therefore cant have more elements than 2^31 - 1
fix subset generation for case where initial capacity might be negative or too small because of truncation of BigInt to int
work in progress: first copy approaches to fix saturation problem and algorithm changes to use a power set to reduce computations
added tests for unsaturated setting and fixed unsaturated run
work in progress: implemented the option for the fragmenter to return saturated or unsaturated fragments, added a test and renamed some tests to prepare for the tests for unsaturated fragments
use thread safe collection in RoundRobinFormulaGenerator
Merge pull request #1226 from uli-f/doublebondstereochemistry-create-alignment-fix
Merge branch 'main' into doublebondstereochemistry-create-alignment-fix
Add JDK-25 to CI
Show how to use the SmirksTransform.
Update the documentation to make it clearer how to use the API
Merge pull request #3 from cdk/doublebondstereochemistry-create-alignment-fix
Some additional null checks and fix the test assertion now the order is corrected.
introduce processCarriers hook in AbstractStereo to allow subclasses to process carriers; move carrier swap logic from DoubleBondStereochemistry::create to overridden DoubleBondStereochemistry::processCarriers method
fix carrier consistency in DoubleBondStereochemistry::create; add tests for strict carrier swap behavior in DoubleBondStereochemistry::create; change DoubleBondStereochemistry::testMap_Map_Map to reflect changes in DoubleBondStereochemistry::create
Avoid regex DoS.
Move the InChI Readers to cdk-inchi module.
Merge pull request #1234 from JonasSchaub/descriptor-parallelisation
a bit of refactoring and linting in the ALogPDescriptor class
a bit of refactoring in the ALogPDescriptor class
a bit of code formatting and refactoring
makes SmallRingDescriptor threadsafe
makes ALogPDescriptor threadsafe
makes Wiener numbers descriptor threadsafe
Fix the terse SGroup test
Fix tests and cleanup CXSMILES output.
Allow terse SRU sgroups
Some extra test cases
Make sure when xbonds are specified they get added to the Sgroup
Fix an issue in SvgDrawVisitor which caused brackets not be drawn.
Fix a failing test, the force field configurator needs ring flags clearing.
Handle Link Nodes in CXSMILES
Simplify the SybylAtomTypeMatcherTest.
Move around some XML config files.
Remove XYZ reader from this test, we don't need to the full mol to isolate what is actually being asserted on.
Move the atom type aware saturation check to cdk-legacy, this is replaced by Kekulisation which is (a) the
Move the CDK AtomType Repos File test and fix the PDB one and the PubChem ASN reader.
Replace import of CMLReader with direct molecule creation. Bug 1014344 (https://sourceforge.net/p/cdk/bugs/231/) is testing CML round tripping so I've moved it there and switch it so CML depends on SMILES (test) rather than the other way around.
Avoid depending on the HinReader and create the molecule directly.
Merge pull request #1225 from JonasSchaub/sugar-detection-utility
adds the Sugar Detection Utility and fixes a few things in the Sugar Removal Utility
Do not sett rootAttachmentPoints if null.
Do not explicitly store the attachment atoms, instead we infer this from the explicit attachment points (wiggly crossed line).
When reading/writing RGroup queries we add on/trim explicit attachment points (wavy line across bond).
Add a utility method to AtomPlace which allows us to place a single atom which is sprouted off the structure.
Add a new utility to get the median bond length and fallback to a provided default if needed.
Move the AtomPlace to cdk-standard, we are going to need this. This does split the "layout" package but in future (CDK v3) they will all be collapsed together.
In CXSMILES we use an SGroup to store the attachment order. Rather than store a map in the 'RGroup' we can reuse this SGroup and get nicer rendering.
Improved stereochemistry invalidation when modifying molecules (Fixes #1221).
Grr DebugBong issues.
A constructor argument is also useful for refactoring code.
More replacements of IBond.Stereo with IBond.Display
A new addBond which accepts a display property as well so we can cleanup more usages of IBond.Stereo
Don't store crossed bonds as wavy.
Replace calls to IBond.getStereo() with IBond.getDisplay().
Replace usages of setStereo with setDisplay. We also add some alias names Up/Down for convenience.
Step 1 in cleaning up IBond.Stereo vs IBond.Display conflation. We deprecate IBond.Stereo get/set and implement them in terms of Display.