Activity for Egon Willighagen
-
Egon Willighagen committed [86f0d1] on Code
make RdfileReader class final; include usage example in RdfileReader class javadoc
-
add javadocs to public methods in RdfileRecord
-
Generally cheminf formats use ASCII and we should not be checking the entire Unicode table for digits since it is only a subset that we care about.
-
Make sure Atom/Bond's ged deref'd when going into a QueryAtomContainer.
-
Fix CCD/WebMolKit Sgroups that are missing the SBL.
-
Updated the location of my blog
-
Actually use the new smart "requires" flag.
-
Smarter SmartsPattern - only do the ring/arom finding if the query actually needs it.
-
Aromaticity is no longer needed here
-
Fix some code smells.
-
Make sure aromatic flags are always cleared.
-
Overhaul and optimise the aromaticity procedures in CDK.
-
Removed the redundant 'public' qualifier
-
Added more builder methods
-
Added more builder methods
-
Added a test class to check methods not tested otherwise (in this module)
-
Test the output of additional isotope info
-
Use the cleaner newAtom() API
-
Add atoms before bonds to hydrogens are created
-
Also create and add the hydrogen
-
Renamed so that the test is actually run (and it fails)
-
Cleaner code by newer API
-
Better assertion
-
There is no exception thrown
-
setUp is not a test
-
Remove explicit use of 'data' classes
-
Use interfaces instead of instances and use silent
-
Stabilise the CDK atom type based aromaticity model. This causes a small fraction of ChEMBL to change but overall is actually a bit close to the Daylight aromatic model.
-
The path based fingerprint should be identical with/without explicit hydrogens.
-
Removed a module that has been empty for a few years
-
Fix a bug in the MDLV2000Reader where the wrong "molecule" is used.
-
Do not try and fix/align layout when 0 or 1 atoms are fixed.
-
Only copy mapped bonds when deciding how to align the structure.
-
Removed more old comments/code
-
Removed old todo statements
-
Removed needless throws statements
-
Removed an unused import
-
Moved another class
-
Moved more test-extra IO test class to the Maven module where the tested classes are
-
Moving classes so that coverage testing can be calculated
-
Add some more generic templates for porphyrin like structures - Fixes #1079.
-
Watch out for isotopes when generating abbreviations. Fixes #1078
-
Bumped the action versions too
-
Only run JaCoCo once
-
Updated many maven plugin versions (checkstyle, compiler, dependency, gpg, jar, javadoc, source, surefire, and pmd) as well findbugs
-
Updated Jacoco, compatible with Java 21
-
Bumped the minimum Maven version to 3.6.3 (2019)
-
Link to the ChemPyFormatics 'book'
-
CMLXOM 4.9 and log4j 2.23.1
-
Updated ${version} in pom.xml
-
Ensure the ESSSR parameter is reported in the FP version info.
-
Increase the max fragment count when generating abbreviations.
-
Additional code smells - unused functions.
-
Options to write out the SD tags from the V2000/V3000 writer.
-
Clean up how option overiding works, when we add a single setting we should check if it already exists. This allows the base V2000/3000 writers to not emit SD tags by default, but when they are used in the SDWriter then they can emit the tags
-
Code Smell: Variable renaming.
-
Move the writing of non-structural data to the V2000 writer.
-
Cleaner Iterating SDF loop
-
Null check needed for reaction (RXN) readers - should probably also read the properties there as well.
-
Now the V2000/V3000 read the SD properties we can simplify the IteratingSdfReader.
-
Missed changes for test files.
-
Read in SDfile properties from single V3000 files to be consistent with V2000.
-
Fix a corner case when depicting cc(C)c
-
Updated JNA-InChI (updating the JNA dependency)
-
Add missing import
-
Added the catching and exception of the Intractable exception
-
added test case for initial cycles oom
-
The number of essential/relevant cycles can be exponential for some molecules - e.g. cyclo-phanes.
-
Fix test value
-
Address some SonarCloud issues
-
Tests are all now enabled.
-
Reinstate test which is fine but not perfect.
-
Warning message changed
-
Fix an error mesg in the AtomContainer implementation
-
Ensure correct error handling when not processing bonds.
-
Make sure we fail currently on bond order unknown.
-
Add in the option to recompute hydrogens.
-
Make sure the bonds are ordered correctly when setting DB stereo.
-
Preliminary work on some SMIRKS options and mimicking other implementations.
-
Allow reactions to be run backwards.
-
The warning message changed.
-
Add in SMIRKS options, currently just a PEDANTIC mode which does additional checks. The valence checking is a WIP but the basics are there.
-
Improved (but complicated) stereochemistry handling, there is likely a better abstraction to keep things more sane.
-
If we can't work out the bond order, make a guess (if possible). In strict/pedantic mode this will warn/fail.
-
Make sure bond position arrays grow correctly. Use ORDER=1 for UP/DOWN stereobonds, this makes it possible for SMIRKS to avoid the warning. Some tweaks are needed to SMARTS writing.
-
Track atom/bond indexes from input SMARTS allowing better warning/error messages.
-
Distinction between total vs implicit hydrogen count, fix default valences for new atoms.
-
Additional unit tests
-
Some extra test coverage.
-
Lazy transforms and the ability to limit the number of matches.
-
Make the Unique/All application of transforms lazy with an iterator.
-
Only call resyncStereo() once (after all ops have been applied) and make sure reactive centre gets marked correctly. These two tests are nice tests which should keep stereo but currently don't.
-
Improve op-code test coverage, normally the delete atom/bond gets optimized so we need specific cases to test those op-codes.
-
Various code cleanup
-
New op: PromoteH to handle the case where we need to possible move and create an explicit H in the product.
-
Correct these two aromatic SMIKRS patterns, we may want to keep the originals for compatibility mode testing.
-
Add a failing test to remind us to fix something.
-
Add in the charge test
-
When we optimize our op-codes (replace-atom) we need to remap the atom numbers in other op-codes.
-
AromaticBond implies aromatic atoms, aliphatic bond does not imply aliphatic atoms.
1 >