2.2 Release Notes
John Mayfield edited this page Oct 31, 2018
·
26 revisions
Download complete JAR: cdk-2.2.jar
(26MB)
- As with all releases, code that has been deprecated (in this release this includes a lot of the existing SMARTS stack) has been moved to the
cdk-legacy
module. Including this module in your project will allow your code to compile. Please note that usage of deprecated classes is not supported and your code should update to use newer APIs. - The "AtomContainer2" API is now the default implementation. To disable this set the environment or system property
CdkUseLegacyAtomContainer=true
. Potential problems with existing code bases are described here. - SMARTS now treats
[D2]
as the number of explicit neighbours rather than number of heavy neighbours. You can match the old behaviour with theSmarts.FLAVOR_CDK_LEGACY
flag. -
IElement
implementations no longer stores the symbol explicitly and only the atomic number. Setting the symbol of atom to something other than one defined in the periodic table (e.g. "X") will default to 'R'. TheIPseudoAtom
is still the correct method to store atoms with custom labels,atom.setLabel("Lys")
for example. -
MurckoFragmenter
now will not calculate 'frameworks' by default and only the main scaffold.
- New APIs for reading, writing, and manipulating SMARTS expressions. SMARTS is read/written with the
Smarts
API:
IQueryAtomContainer query = new QueryAtomContainer();
if (!Smarts.parse(query, "[n,o]1cccc1"))
System.err.println(Smarts.getLastErrorMesg());
QueryAtom qatom = (QueryAtom) query.getAtom(0);
qatom.getExpression()
.and(new Expr(Expr.Type.TOTAL_H_COUNT, 0).negate());
String sma = Smarts.write(query); // sma=[n,o;!H0]1cccc1
The parser allows a flavor to be parsed in that enables support for syntax or behaviour extensions from various toolkits.
// #X means a hetero atom
Smarts.parse(query, "[#X]", Smarts.FLAVOR_MOE);
// z<num> = num hetero nbrs
Smarts.parse(query, "[z2]", Smarts.FLAVOR_CACTVS);
// R2 in OEChem is the same as x2
Smarts.parse(query, "[R2]", Smarts.FLAVOR_OECHEM);
// SSSR based R2, Z<num> = in ring of size <num>
Smarts.parse(query, "[R2]", Smarts.FLAVOR_DAYLIGHT);
// old CDK interpretation was num non-H neighbours
Smarts.parse(query, "[D2]", Smarts.FLAVOR_CDK_LEGACY);
- SMARTS atom/bond expression trees are now represented by the single
Expr
class. Previously each expression was defined as a separate class making it very difficult to inspect and modify values.
Expr e = new Expr(ELEMENT, 8); // #8
e.and(DEGREE, 2); // [#8D2]
e.or(ELEMENT, 7); // [#8D2,#7]
- Creating a query from 'real' molecule is now easier with
QueryAtomContainer.create()
, see QueryAtomContainer. - New subgraph search algorithm available,
DfPattern
. Note this algorithm will only withCdkUseLegacyAtomContainer=false
(now the default). - Cleaned up logging, quieter builds, set level from API
ILoggingTool.setLevel()
. Default logging to standard error can now be configured with system property:-Dcdk.logging.level=ERROR
. - Custom comparator in Canonical labelling,
Canon.label()
, Atom-Maps in SMILES can now define ordering - #465
- New fingerprint
AtomPairs2DFingerprinter
#417 - New descriptor
FractionalCSP3Descriptor
for the fraction of Sp3 carbons. #487 -
SubstructureFingerprinter
can now generate a count fingerprints. - Faster Murcko scaffold algorithm, requesting just main scaffold is now O(N) - MurckoFragmenter. #466
- Improved hydrogen bond acceptor count #495
- Support EPS output from Depiction - #433
- Render delocalised bonds as dashes - #403
- Better abbreviations of fragments/salts - #491
- Improved reaction layout when aligning mapped atoms.
- SVG units - can now render SVG in px as well as mm - #414
- New highlight style that adds a white outline around text - #258
- Improved
IsotopePatternGenerator
speed and output now provides the isotopic compositions of the generated masses (most abundant MF). - Round trip charge and explicit isotopes from/to molecular formulas.
[[13]CCH5O]-
- Improved mass calculations, missing isotopes are not considered to be the most abundant (major) but rather the natural mass. Setting the isotopes explicitly allows the old behaviour.
- Smaller V2000 molfile by truncating default property blocks, behaviour matches current BIOVIA tools.
MDLV2000Writer mdlw = new MDLV2000Writer();
mdlw.getSetting(MDLV2000Writer.OptWriteDefaultProperties)
.setSetting("false");
Example:
CDK 0425181143
5 4 0 0 1 0 0 0 0 0999 V2000
0.0000 0.0000 0.0000 C 0 0 1 0 0 0
0.0000 0.0000 0.0000 C 0 0
0.0000 0.0000 0.0000 C 0 0
0.0000 0.0000 0.0000 O 0 0
0.0000 0.0000 0.0000 C 0 0
1 2 1 0
2 3 1 0
1 4 1 0
1 5 1 0
M END
- Correctly read implicit D/T when reading InChIs
213 John Mayfield
35 Saulius Gražulis
17 Kazuya Ujihara
9 bach
8 Egon Willighagen
6 michaelwenk
4 Rajarshi Guha
4 Lyle D. Burgoon
2 Katrin Leinweber
1 potatostodie
1 meier-rene
1 balaji
1 Sundaram
1 Raven
1 Matt Swain
1 Martin Gütlein
43 Egon Willighagen
26 John Mayfield
- add highlight mode with outer-glow and white-edges around symbols to improve readability Martin Gütlein on 2017-01-05
- Back to dev version (2.2-SNAPSHOT) John Mayfield on 2017-12-28
- Fix existing unit test, no change to result but inputs were wrong. John Mayfield on 2017-12-29
- Adjustment required to symbol visibility, otherwise carbons will be displayed as '>CH' in the delocalised rings. John Mayfield on 2017-12-29
- Preliminary handling of delocalised bonds, we put a dash on one side. Some small rings will use a circle/donut/life buoy in future but this does not handle the general case (e.g. porphyrin). John Mayfield on 2017-12-29
- Option to force rendering of delocalised bonds. John Mayfield on 2017-12-29
- Check for molecular descriptors modifying mol/atom/bond flags. John Mayfield on 2018-01-03
- Descriptors that need a clone to avoid modifying the input. John Mayfield on 2018-01-03
- Polymer clone is broken in the BioPolymer sub-class (case in point of why cloning is bad) - fix the clone and use it in this descriptor. John Mayfield on 2018-01-03
- removing a for loop for setting bond order single in constructtautomers balaji on 2018-01-04
- refactoring formatfactory.java removing duplicate code and converted into methods Sundaram on 2018-01-05
- Ignore lp: designations. John Mayfield on 2018-01-09
- Sgroup manipulator to help with copying/cloning. John Mayfield on 2018-01-13
- Ensure sgroups get cloned. John Mayfield on 2018-01-13
- Refined hydrogen suppression, watch out for crossing bonds. John Mayfield on 2018-01-13
- Hydrogens can still be deleted from Sgroups, so we need to update the sgroups accordingly. John Mayfield on 2018-01-13
- Correct assertion John Mayfield on 2018-01-13
- Isolated tests for SgroupManipulator John Mayfield on 2018-01-14
- Improve doc and generify replace maps. John Mayfield on 2018-01-14
- Remove stray character from prev commits. John Mayfield on 2018-01-14
- Improved placement of positional variation bonds. John Mayfield on 2018-01-14
- Updated the copyright line Egon Willighagen on 2018-01-14
- Fix to read MDL RXN file by ReaderFactory Kazuya Ujihara on 2018-01-17
- Synchronized with a line counts of drugs.smi Kazuya Ujihara on 2018-01-28
- Allow units to be specified for SVG John Mayfield on 2018-01-30
- Update unit test John Mayfield on 2018-01-30
- Additional test option, avoid window popping up when running build on OS X. John Mayfield on 2018-01-31
- Remove calls to print in test cases. John Mayfield on 2018-02-01
- Ignore non-test (no assertion) that is printing to STDOUT John Mayfield on 2018-02-01
- Remove calls to print in main code base, replacing when required with logging John Mayfield on 2018-02-01
- JENA/RDF wants and SLF4J implementation, complains if not. John Mayfield on 2018-02-01
- Remove stack trace dumping, the IP code is doing odd calls catching, re-reporting errors. Better now. John Mayfield on 2018-02-01
- Underlying JAXPSAX parser is complaining, bit of a silly test but we can dupe it with an empty file. John Mayfield on 2018-02-01
- Only report errors in travis. John Mayfield on 2018-02-01
- Suppress Log4J unconfigured warning in cdk-iordf tests. John Mayfield on 2018-02-01
- Added PaDEL AtomPairs2D fingerprint. Lyle D. Burgoon on 2018-02-06
- .DS_Store files removed Lyle D. Burgoon on 2018-02-08
- made minor fixes as requested. Lyle D. Burgoon on 2018-02-08
- generate new DOIs against preferred resolver Katrin Leinweber on 2018-02-12
- link DOIs to preferred resolver Katrin Leinweber on 2018-02-12
- corrected test case Lyle D. Burgoon on 2018-02-13
- AtomRef/BondRef should provide a toString. John Mayfield on 2018-02-15
- Correct condition check for create 'QueryBondRef' in AtomContainer2 John Mayfield on 2018-02-15
- Recursive SMARTS atom should use the Pattern API, not sure why I used Ullmann here. Also remove redundant logger. John Mayfield on 2018-02-15
- Contributor 100 (Katrin Leinweber) and 101 (Lyle D. Burgoon) Egon Willighagen on 2018-02-16
- Add a new method atom::getBond(), that allows testing as to whether two atoms are connected outside of the AtomContainer. John Mayfield on 2018-02-17
- Revert "Add a new method atom::getBond(), that allows testing as to whether two atoms are connected outside of the AtomContainer." Egon Willighagen on 2018-02-18
- Required method John Mayfield on 2018-02-18
- Revert "Revert "Add a new method atom::getBond(), that allows testing as to whether two atoms are connected outside of the AtomContainer."" John Mayfield on 2018-02-18
- Another Atom implementation John Mayfield on 2018-02-18
- Doc updated. John Mayfield on 2018-02-18
- CXSMILES relative stereo can omit the component ids. John Mayfield on 2018-02-20
- Update README.md John Mayfield on 2018-02-28
- Update README.md John Mayfield on 2018-02-28
- Remove direct usage of 'AtomContainer' John Mayfield on 2018-02-28
- Allow the 'matches' method to be overridden so down-stream implementations can provide a more efficient algorithm (recursive vs itterable). John Mayfield on 2018-02-28
- Fixing read-in of the CIF unit cell. Saulius Gražulis on 2018-02-28
- Adding a test for the CIF cell constant read. Saulius Gražulis on 2018-02-28
- BUG FIX: correcting CIFReader atom loop processing. Saulius Gražulis on 2018-02-28
- Removing unused code. Saulius Gražulis on 2018-02-28
- QueryAtom/Bond need hashCode/equals to check Atom/BondRefs John Mayfield on 2018-02-28
- Now use AtomContainer2 by default. John Mayfield on 2018-03-03
- Compact representation of logical predicates over atoms/bonds. John Mayfield on 2018-03-03
- Can't test the recursive expressions yet.... John Mayfield on 2018-03-03
- Ignore not enough... John Mayfield on 2018-03-03
- Update SmiFlavor.java John Mayfield on 2018-03-06
- Trying to revive the EPS file output, with bounding box. The bounding box seems wrong, though. Saulius Gražulis on 2018-03-08
- Managing to fit the molecule image into a bounding box using the regexp replacement on the final string. Saulius Gražulis on 2018-03-08
- Making replacements more compact. Saulius Gražulis on 2018-03-08
- Using safer 'replaceFirst' function. Saulius Gražulis on 2018-03-08
- Removing empty line. Saulius Gražulis on 2018-03-08
- Making my code OS-generic by using 'System.getProperty("line.separator")'. Saulius Gražulis on 2018-03-08
- Adding %%BoundingBox at the beginning of an EPS stream. Saulius Gražulis on 2018-03-09
- Adding references to documentation and the "... EPSF-3.0" comment. Saulius Gražulis on 2018-03-09
- Extending comments. Saulius Gražulis on 2018-03-09
- Adding test case for depiction.toEpsStr(). Saulius Gražulis on 2018-03-09
- Fixing typo in comments. Saulius Gražulis on 2018-03-09
- Adding specific EPS format. Saulius Gražulis on 2018-03-09
- Adding EPS_FMT to supported format list. Saulius Gražulis on 2018-03-09
- Fixing scaling for EPS output. Adding test for PS. Saulius Gražulis on 2018-03-09
- Adding missing @Test marker. Saulius Gražulis on 2018-03-09
- Adding one more EPS output test. Saulius Gražulis on 2018-03-09
- Optimising imports. Saulius Gražulis on 2018-03-09
- Shortening test code. Saulius Gražulis on 2018-03-09
- Removing unused imports. Saulius Gražulis on 2018-03-09
- Correcting format name (EPS) in comments. Saulius Gražulis on 2018-03-09
- Determining new-line separator from file contents, not from OS type. Saulius Gražulis on 2018-03-09
- Adding (CC) header. Saulius Gražulis on 2018-03-09
- Using else-if as suggested by Egon. Saulius Gražulis on 2018-03-09
- Fixing spelling of 'Encapsulated' in comments. Saulius Gražulis on 2018-03-09
- Reverting: uncommenting the '[INFO] Using the new AtomContainer...' mesage. Saulius Gražulis on 2018-03-09
- Moving citations to JavaDoc comments. Saulius Gražulis on 2018-03-09
- Adding URI reference to PLDS92. Saulius Gražulis on 2018-03-09
- Updated the working directory info for the c06d7abfeaa3ffc781af1b5e7063780e8c9e2948 path Egon Willighagen on 2018-03-11
- Added Saulius Gražulis to the author lists Egon Willighagen on 2018-03-11
- Fixing DOS new-line escape sequences. Saulius Gražulis on 2018-03-11
- Adding my name to (C) record in DepictionTest.java. Saulius Gražulis on 2018-03-11
- Fixing bibliography references. Saulius Gražulis on 2018-03-11
- Updating cross-reference keys. Saulius Gražulis on 2018-03-11
- Remove assertion to fail Kazuya Ujihara on 2018-03-14
- Provide access to subquery from Expr John Mayfield on 2018-03-16
- Update documentation John Mayfield on 2018-03-16
- Make QueryAtom and QueryBond non-abstract, add an Expr field. John Mayfield on 2018-03-16
- Update component grouping to also use the component id's from the query getProperty(). Previously queries/targets were handled differently incoming patches allow us to treat this more the same leading cleaner code - notice both the query and target now go into 'determineComponents' in the constructor. John Mayfield on 2018-03-16
- Slight improvement on ordering of expression trees when using 'AND', without this '[C@H]' in SMARTS comes out as '[HC@]' valid but ugly. John Mayfield on 2018-03-18
- New SMARTS parser and generator. John Mayfield on 2018-03-18
- Duplicate existing parser tests from the JAVACC parse showing everything is passed okay. John Mayfield on 2018-03-18
- Improved logging API, each logger can now have it's level set explicitly. For Log4J we forward this call to the underlying implementation. John Mayfield on 2018-03-18
- logger is never set and so would cause a NPE if the exception is ever thrown. John Mayfield on 2018-03-18
- Logging should go to stderr not stdout! John Mayfield on 2018-03-18
- Avoid memory allocation for info, error, warn Object array methods John Mayfield on 2018-03-18
- More sensible default level, we actually now see there are possible problems in the PDB tests! John Mayfield on 2018-03-18
- Actually configure a logger for the log4j tests. John Mayfield on 2018-03-18
- Log4J inherits it's level from the root logger. John Mayfield on 2018-03-18
- Now using stderr John Mayfield on 2018-03-18
- Null safe appends to StringBuilder John Mayfield on 2018-03-18
- Don't allow all double bonds to move when use aromatic flag is set. John Mayfield on 2018-03-18
- Template for propellane John Mayfield on 2018-03-18
- Updated README on compiling. John Mayfield on 2018-03-21
- Update README.md John Mayfield on 2018-03-21
- Move warning inside else condition. John Mayfield on 2018-03-25
- Documentation new Smarts documentation. John Mayfield on 2018-03-25
- a and A don't need square brackets. John Mayfield on 2018-03-25
- affect not effect John Mayfield on 2018-03-25
- Added support to generate molecular formula that includes mass number of non-major isotopes are present. This can be useful in formula generation Rajarshi Guha on 2018-04-16
- Ensure we check for unset mass number Propery chain appends Rajarshi Guha on 2018-04-16
- Updated handling of mass nunber in formula string generation Added old function signature Rajarshi Guha on 2018-04-16
- More efficient and clean MF generation. John Mayfield on 2018-04-18
- Prevent loss of isotopes in Molfiles. John Mayfield on 2018-04-25
- Make it easier to change output options using the API without having to build a property list etc. John Mayfield on 2018-04-25
- Use StringBuilder John Mayfield on 2018-04-25
- Clean up generation of atom properties. John Mayfield on 2018-04-25
- When possible put the charge and mass in the atom block as well as using M ISO, M RAD, and M CHG John Mayfield on 2018-04-25
- Add option to skip writing all the zeros in molfiles (matches BIOVIA default). John Mayfield on 2018-04-25
- More efficient number formatting and truncate bond properties as well. John Mayfield on 2018-04-25
- Suppress hydrogens on extended tetrahedrals. John Mayfield on 2018-05-03
- Accessing self-bond should be null. John Mayfield on 2018-05-03
- fixup (suppress H on ex tet) John Mayfield on 2018-05-03
- For consistency, if dimensions is 3D then we have a Z even if all 0. John Mayfield on 2018-05-03
- Representation of extended CisTrans John Mayfield on 2018-05-03
- Round tripping of extended Cis/Trans through SMILES John Mayfield on 2018-05-03
- Perceive extended Cis/Trans from 2D and 3D coordinates. John Mayfield on 2018-05-03
- Extended Cis/Trans layout generation. John Mayfield on 2018-05-03
- We assign two wedge bonds when there are four neighbors, however if we've assigned one and the second choice is a possible stereo-centre don't assign it. John Mayfield on 2018-05-03
- Improved detection of 'potential' tetrahedrals, we want to avoid putting fat end of wedges to these. John Mayfield on 2018-05-03
- Handle extended Cis/Trans when suppressing hydrogens. John Mayfield on 2018-05-03
- Support longer extended tetrahedral systems. John Mayfield on 2018-05-09
- Test InChI is still okay, limited to 2/3 cumulenes. John Mayfield on 2018-05-09
- Improved small ring test for stereo perception. John Mayfield on 2018-05-09
- Make sure extended Cis/Trans atoms are consistently stored. John Mayfield on 2018-05-09
- Improved Aryl-Aryl atropisomer test. John Mayfield on 2018-05-09
- Further adjustments to accommodate longer extended tetrahedrals John Mayfield on 2018-05-09
- Resolve problem with testing wrong atom when assigning wedges. John Mayfield on 2018-05-09
- Additional 2D layout templates. John Mayfield on 2018-05-09
- Improved stereo-detection, do not perceive stereo on fused trivalent nitrogens. John Mayfield on 2018-05-09
- Missed constraint, reject double bonds in rings <= 7. John Mayfield on 2018-05-09
- Resolve test-failure, we set the hasZ if the file is marked as '3D'. John Mayfield on 2018-05-09
- Some restructuring requiring to resolve failing unit tests. John Mayfield on 2018-05-09
- Read implicit D/T from InChIs. John Mayfield on 2018-05-15
- Read correct number of implicit D/T from InChIs. meier-rene on 2018-05-15
- Doc updates John Mayfield on 2018-05-16
- Remove dodgey template, skewed ring. John Mayfield on 2018-05-18
- Ignore test, waiting on Beam patch. John Mayfield on 2018-05-18
- Not warning worthy John Mayfield on 2018-05-18
- Clean up CML stereo handling and check for empty string... John Mayfield on 2018-05-18
- atomNumberMap should be cleared when a new model is loaded, serial numbers reset John Mayfield on 2018-05-18
- Not a warning John Mayfield on 2018-05-18
- StAX parser could be more written more robustly as a "recursive descent parser" but looks like all the elements being warned on here do need further action. John Mayfield on 2018-05-18
- Mols were wrong way round... wonder if this would have change results n years ago... (in CDK Legacy) John Mayfield on 2018-05-18
- Corrected JavaDoc Raven on 2018-05-24
- Utility to methods to clear any mass numbers that are determined to the be the major isotope. John Mayfield on 2018-05-29
- Parse formulas with explicit mass indication. John Mayfield on 2018-05-29
- Fix wrong class names Kazuya Ujihara on 2018-06-09
- Parameters[0] of ElementRule can be null Kazuya Ujihara on 2018-06-09
- Some rules are not stateless, e.g. ElementRule Kazuya Ujihara on 2018-06-09
- COUNTS is followed by space and digits Kazuya Ujihara on 2018-06-09
- Update JavaDoc with superscript John Mayfield on 2018-06-10
- Ensure formulas are generated consistently. John Mayfield on 2018-06-19
- Warn on usage of old AC now. John Mayfield on 2018-06-19
- Not a warning, common to have truncated lines. John Mayfield on 2018-06-19
- Confusing but now correct, query is in the target and so the qIndex is from the tAtom which is in the reactant. John Mayfield on 2018-06-19
- Remove base tests, it looks like there was some discussion on whether params should be non-null by default. For IsotopePatternRule and ElementRule they need initilaization to have non-null params. Additionally we rename the IsotopePatternRuleTest for validate to override the subclass (that fails due to null params). John Mayfield on 2018-06-19
- CDK is now Java 1.7+ so we can always validate in the same way. John Mayfield on 2018-06-19
- Skip install step in travis, it is done automatically when tests are run. John Mayfield on 2018-06-19
- Far too many warnings from the tests, often over zealous logging but for now lower the logging level to only show ERRORs. John Mayfield on 2018-06-19
- Add bpy to abbreviation label symbol list Matt Swain on 2018-06-20
- Added a new contributor (two more, but without full names) Egon Willighagen on 2018-06-21
- Add list of SMARTS convering the countable MACCS keys bach on 2018-06-24
- Extend standard-substructure for countable MACCS bach on 2018-06-24
- Update SMARTS pattern for countable MACCS fps: aromatic 6-ring bach on 2018-06-24
- Make SubstructureFingerprinter compatible with countable maccs fps bach on 2018-06-24
- Add tests for substrcture fingerprints binary bach on 2018-06-24
- Implement counting fps for 'SubstructureFingerprints' bach on 2018-06-24
- Update substructure key SMARTS: aromatic 6-ring bach on 2018-06-24
- Fix broken test: KlekotaRothFingerprinterTest bach on 2018-06-24
- Minor cleanup, assertEquals instead of assertTrue, remove new enum type and just use the constructor to set the patterns from an array. bach on 2018-06-24
- Use VF instead of Ullman. John Mayfield on 2018-06-24
- Include OvalElements in Bounds calculation. John Mayfield on 2018-06-26
- Anonymise should copy coordinates from the input. John Mayfield on 2018-06-26
- Spiro Atom Count Descriptor Rajarshi Guha on 2018-06-27
- Ensure documentation is consistent with the actual implementation. John Mayfield on 2018-06-27
- Outer glow with white highlight. John Mayfield on 2018-06-27
- Minor modification to the canonical labelling allows a custom atom comparator to be used. John Mayfield on 2018-06-27
- Use comparator in SMILES generation, this will be expanded in future. By default it produces the exact same ordering as the current bit packing used in Canon. However it is much simpler to add optional invariants (e.g. AtomAtomMaps). John Mayfield on 2018-06-27
- Additional SMILES flavor that renumbers atom-atom maps during creation. John Mayfield on 2018-06-27
- Linear time algorithm for computing the Murcko scaffold. Used if only the scaffold is requested. John Mayfield on 2018-06-27
- Test case for CHEMBL529226 - previously reported as being 17 hours, now runs in ~ 2 ms (singleFragmentOnly). John Mayfield on 2018-06-27
- Only store the atomic number of an atom and not it's redundant symbol. John Mayfield on 2018-06-27
- Overloaded type in Java 9 plus cause problems with method lookup here. Fixed by casting to the super class. John Mayfield on 2018-06-27
- Only load a dependant mechanism if one exists. John Mayfield on 2018-06-28
- Remove cdk-log4j dependency, defaults to DEBUG logging whilst running tests. John Mayfield on 2018-06-28
- Catch bonds that are not handled at this point. John Mayfield on 2018-06-28
- Remove redundant dictRef from CML amino acids file John Mayfield on 2018-06-28
- Better layout when spiro degree > 2 John Mayfield on 2018-06-28
- Missing if statement John Mayfield on 2018-06-29
- Fix JavaDoc parameter name John Mayfield on 2018-06-30
- Correct typo and change default constructor to only calculate the scaffold and not all possible frameworks. John Mayfield on 2018-06-30
- Added another recent author Egon Willighagen on 2018-07-01
- Updated the version Egon Willighagen on 2018-07-01
- IsPrecedingPathTo(int) don't throw exception Kazuya Ujihara on 2018-07-02
- Improve alignment of mapped reaction layout (don't crash). John Mayfield on 2018-07-20
- Cache Smiles in AtomTypeTools Kazuya Ujihara on 2018-07-22
- Failure tests for new Atom(String) Kazuya Ujihara on 2018-07-29
- Fix setting fields on new Atom(String) Kazuya Ujihara on 2018-07-29
-
- NMRSHIFT constant field added for sulfur - NMRSPECTYPE constant fields added as spectrum type identifiers for e.g. 1D DEPT90 or 2D HSQC experiments michaelwenk on 2018-07-31
- Set default implicit hydrongen count at null Kazuya Ujihara on 2018-08-02
- Remove Lo4J from cdk-bundle. John Mayfield on 2018-08-06
- changed the @see path to specType class member in Spectrum class michaelwenk on 2018-08-06
- Correct API call. John Mayfield on 2018-08-13
- Update issue management. John Mayfield on 2018-09-17
- changed "NMRSPECTYPE_1D_ARBITRARY" to "NMRSPECTYPE_1D" michaelwenk on 2018-09-18 @see links removed michaelwenk on 2018-10-05
- Correct corner case of parsing SMARTS with atom-maps (in the new parser). John Mayfield on 2018-10-07
- Be strict when placing partial rings, check bond count in addition to atom count. John Mayfield on 2018-10-07
- Units should be parsed in to the SVG Draw Visitor. John Mayfield on 2018-10-08
- Improved writing of SMARTS element expressions. John Mayfield on 2018-10-09
- Access error messages from SMARTS parser. John Mayfield on 2018-10-09
- Adapt stereo queries to use new (and old) API points. John Mayfield on 2018-10-10
- Use new SMARTS API in SmartsPattern. Tests are no longer needed as package-private method was removed. John Mayfield on 2018-10-10
- Still to be optimised (as comment suggests) but working without stereo as provide (need to work out dependency juggling to get the stereo working). John Mayfield on 2018-10-10
- Replace direct usages of matcher classes (now all unified under the Expr API) John Mayfield on 2018-10-10
- Replace usage of SMARTSQueryTool with SmartsPattern. Some efficiency improvements as SMARTS are actually "compiled now". John Mayfield on 2018-10-10
- Old SMARTS had wrong semantics for 'D' with respect to explicit H atoms. We can replicate this behaviour if needed but since the SMARTS are from elsewhere it's more correct to suppress H for the match. John Mayfield on 2018-10-10
- Duplicate some class ready for the mass of other classes to cdk-legacy. Duplicate so we can have the nice clean ".cdk.smarts" package rather than the current ".cdk.smarts", ".cdk.smiles.smarts", ".cdk.isomorphism.matchers.smarts". Duplicating maintains backwards compatibility if the legacy module is included. John Mayfield on 2018-10-10
- Move to legacy the 86 classes of the old SMARTS API, now replaced with 2 public ones (Expr+SmartsPattern). John Mayfield on 2018-10-10
- Documentation updates and deprecation marking. John Mayfield on 2018-10-11
- Round trip MF charge. John Mayfield on 2018-10-11
- Some corrections. John Mayfield on 2018-10-11
- New constructors for QueryAtom/Bond. John Mayfield on 2018-10-12
- Fix bitwise 'and' operators to logical ones Kazuya Ujihara on 2018-10-13
- Add FractionalCSP3Descriptor Kazuya Ujihara on 2018-10-14
- Move more one-of matches to cdk-legacy and replace with the new Expr APIs. Some backwards compatibility was needed for classes expecting Query objects to have a symbol/bond order (bad behaviour). John Mayfield on 2018-10-15
- Hide away post match query filters, users should not have to worry about these and their previous locations made recursive SMARTS difficult. John Mayfield on 2018-10-15
- Fix to avoid using clone method Kazuya Ujihara on 2018-10-16
- Add contributor Kazuya Ujihara on 2018-10-16
- Correct element in "MMElementRule" John Mayfield on 2018-10-16
- Both null and 0 mean unspecified isotope. John Mayfield on 2018-10-16
- Add FractionalCSP3Descriptor to services info Kazuya Ujihara on 2018-10-17
- Do not align atoms if their geometry doesn't match, for now this means (collinear vs bent). John Mayfield on 2018-10-18
- Better handling of combined abbreviations. Previously a set of connected components was split and individually abbreviated (then combined). Now we perform a leave one out test of multiple components, allowing us to capture the Pd(dppf)Cl2 (3 components) and DCM (1 component) in this example. John Mayfield on 2018-10-18
- Null check needed. John Mayfield on 2018-10-18
- index potatostodie on 2018-10-18
- Faster and cleaner duplicate filtering. A map with floating point keys doesn't really work so it had to do two passes, C10000 test is now 3x faster. John Mayfield on 2018-10-21
- Rename parameter, note min abundance vs intensity. John Mayfield on 2018-10-21
- Generate the molecular formula for each isotope container. John Mayfield on 2018-10-21
- More robust total exact mass calculation. John Mayfield on 2018-10-21
- Moving avg. mass when merging. John Mayfield on 2018-10-23
- Adjustable resolution. John Mayfield on 2018-10-23
- Utilities for cleaner formula handling. John Mayfield on 2018-10-23
- Clean up method further, localising variables and for-each loop. Fair speed improvement. John Mayfield on 2018-10-23
- We can move the conditions out of the n x n loop, also check if it is essentially 0 rather than equal to 0. John Mayfield on 2018-10-23
- We can create the list of isotopes once for each atom and then re-use it. John Mayfield on 2018-10-23
- Allow an isotope container to have multiple formulas. John Mayfield on 2018-10-23
- Generate and merge multiple MF in isotope generator when requested. John Mayfield on 2018-10-23
- Remove min abundance options (this change set) - the value isn't actually minimum abundance but rather a filter as we go along. John Mayfield on 2018-10-23
- Weighted moving average. John Mayfield on 2018-10-24
- Resolve #495 John Mayfield on 2018-10-25
- Some adjustment needed to match tetrahedrals correctly. John Mayfield on 2018-10-25
- Support CACTVS style ranges expressions, we normalise these to a simple OR list. John Mayfield on 2018-10-25
- Better ordering of OR lists, already and logic for AND lists. John Mayfield on 2018-10-25
- Minor test cleanup John Mayfield on 2018-10-25
- Don't allow 'Z' in loose SMARTS. RDKit now match CACTVS so 2:1. John Mayfield on 2018-10-25
- Need to store more info to get stereo correct. John Mayfield on 2018-10-25
- New very flexible way of creating Query structures. John Mayfield on 2018-10-25
- The majority of these methods can not be replaced by the new generic one. John Mayfield on 2018-10-25
- Warn about expr type not supported fully in SMARTS. John Mayfield on 2018-10-25
- Documentation on new create method. John Mayfield on 2018-10-26
- Update documenation on QueryAtomContainerCreator. John Mayfield on 2018-10-26
- Additional documentation and typo fix. John Mayfield on 2018-10-26
- Cleaner H adjustment. John Mayfield on 2018-10-26
-
- constant field for NMR signal intensity added michaelwenk on 2018-10-26
-
- constant field for NMR signal multiplicity added michaelwenk on 2018-10-26
- Now fixed? John Mayfield on 2018-10-26
- Update beam, not changes from the existing SNAPSHOT just a release version. John Mayfield on 2018-10-27
- Deprecate subgraph matching is UIT, saw someone using this the other day. John Mayfield on 2018-10-27
- Custom JavaDoc, make code examples more obvious. John Mayfield on 2018-10-27
- Resolve JavaDoc errors. John Mayfield on 2018-10-27
- Large clean up README, lots of links were dead and there were very verbose details that could be moved elsewhere. John Mayfield on 2018-10-27
- Renamed BUGS => ISSUES and update info John Mayfield on 2018-10-28
- More standardisation. John Mayfield on 2018-10-28
- Update copyright. John Mayfield on 2018-10-29
- Depth-first (DF) substructure search algorithm. John Mayfield on 2018-10-29
- Use the new matchRoot method for recursive SMARTS, much more efficient. John Mayfield on 2018-10-29
- Use reflection to get a IChemObjectBuilder if none was provided. John Mayfield on 2018-10-29
- Release 2.2 John Mayfield on 2018-10-30