A different exception class is used when a name is simply unknown to a custom webservice compared to when the connection to the webservice failed.

Some names containing "ylidene" were converted to an invalid structure.

More CAS registry numbers are supported.

Name to Structure could use a large amount of memory at startup if it was used simultaneously in a large number of threads.

Isotope descriptors were missing for the carbon atom of some characteristic groups (carboxylic acid, carboxamide, ...) on a cyclic parent.

Some names containing "epoxy" were converted to an invalid structure.

Polyalenes (for instance pentalene, octalene, ...) are now supported.

Charges were missing on some metals: nickel, zinc, ...

Longer common names are now supported.

CAS Registry Numbers with invalid check digits are now reported as errors with a specific error message. This happens in particular for CAS Numbers which have been entered with a typo.

String of the form X-XX-X were wrongly recognized as CAS Registry Numbers.

The common name dictionary is extended to include INCI names (International Nomenclature of Cosmetic Ingredients).

The new nameField format option can be used to set which field/property stores the original name (instead of the molecule title).

Nucleoside substituents such as adenosyl are now supported.

Some nucleosides and nucleotides had incorrect stereochemistry.

CAS number conversion was failing since version 17.9 in .NET 32 bit applications.

The homology group haloalkyl is now represented more accurately.

The conversion of some sugars has been improved.

Names with a numbered C locant before a homology group, such as "1-C-acyl..." were not converted.

When converting CAS Registry Numbers, a more advanced detection of network proxy settings is now in place. This includes supporting the proxy auto-configuration (PAC) and the Web Proxy Autodiscovery Protocol (WPAD) protocols.

More ASCII syntaxes are supported for superscript numbers in bridged ring system names.

The name format was not detected for names contain the special "soft-hyphen" character.

CAS registry number resolution failed when the ChemAxon user home directory did not exist. This could happen exceptionally, such as running under the NetworkService system account in Windows.

Generic structures now get assigned the type 'generic' even when imported from a common name, such as "ester", "fatty acid", ...

All character entities are now recognized in the USPTO XML format.

CAS number resolution using the remote webservice works again in Java versions lower than 8 and in .NET applications such as JChem for Office.

The webservice call for CAS number resolution never uses an HTTP redirection, resulting in faster response time.

More OCR error correction is performed in alkyl ranges.

Improvements when processing names with added spaces.

Ring fusions involving multiple bonds are now supported.

The traditional steroid numbering of cyclopenta[a]phenanthrene is now supported in addition to the IUPAC numbering.

The numbering of complex fused ring systems is now supported.

Capital S- is now understood as sym- where appropriate.

Names with added space characters are better recognized.

More fused names are understood.

Metals and some other element names were imported with an unneeded zero valence property.

More disconnected structures are now supported, for instance pyridine acetate.

More complex ethers are now supported.

Non-strictly IUPAC forms of Hantzsch-Widman names are now recognized.

Names separated by ';' were not detected from arbitrary text formats. This affected in particular JChem for Excel.

Some salts and esters were not recognized.

The N^{...} syntax for superscript locants is now supported.

More OCR errors are corrected.

Some names with multiplied suffixes were not converted.

The common name dictionary received a significant update.

The homology group heterocycloalkyl is now supported.

Some names containing bond modifiers with non-consecutive locants were not supported, for instance: 1,3,5(10)-Estratriene.

The iso form of chains, such as isodecyl, is now better supported.

Some element names were wrongly detected and represented as structures.

Some names containing "ano" lead to corrupt structures.

Higher order fusions with omitted primed locants are now supported.

Fused ring systems with higher-order attached components are now supported.

Fused ring systems with multiple first-order attached components are now supported.

"Fatty acid" had an overly broad representation.

HTTP proxy configuration is now described in the logging system and in the stack trace when an HTTP connection fails. This is useful in particular when investigating why a CAS number resolution failed.

Some incomplete names used to be converted to impossible aromatic structures.

The webservice call for CAS number resolution was using an HTTP redirection, resulting in slower response time.

The conversion of Japanese element names was enabled by disabling common name conversion.

Some group types (for example ethoxy, propargyl, C1-C8-alkoxycarbonyl, ...) that were generated as complete structures are now generated with an attachment point.

Names containing a substituent to a formohydrazido group were not converted.

The support for Japanese chemical names found in Japanese legal documents has been vastly improved.

Some syntaxes for benzofused ring systems were not supported.

The support for functional groups containing Selenium and Tellurium atoms has been improved.

Chain ester names used as substituents are now recognized.

All locants are now supported in tryptamine.

Support has been added for more cases of accidental spaces in name (often caused by OCR), such as "m ethyl".

Support has been improved for parent names starting with a symbol, such as s- and as-indacene.

Additional OCR error correction allows to convert more names from scanned documents and patents.

Spiro names that mistakenly lack a prime on the second spiro atom locant are now automatically fixed.

The accuracy in the conversion of Chinese names to structures is improved.

The numbering of flavanone has been fixed.

Multiple custom dictionaries can be specified using the 'dict' format option: dict=<dictionary_path_1>;<dictionary_path_2>

The 'systematic' name format option does not disable the conversion of other name types anymore.

Names containing the keyword "substituted" are now supported, for instance: "2,4-disubstituted pyridine".

More syntaxes for carbon count in homology groups are recognized, for instance "(C1-6) alkyl".

The "Heterocyclyl" homology group is now supported.

Fix new OCR error in Japanese.

CAS Registry Numbers can now be converted when entered with extra whitespace before or after the number. This is especially useful when using copy-paste in end-user applications.

Several improvements increase the conversion of systematic names.

Names with small typographical errors, such as added spaces or missing dashes, are better supported. This especially improves the extraction of names from lower quality sources, such as patent texts generated by OCR.

Esters of nitric acid were imported with a radical instead of the C-O bond.

Some additional OCR error correction is performed in Chinese.

Some lists of names improperly separated lead to extremely long processing time.

β-alanine wrongly had a chiral stereocenter.

The file size of the naming component has been reduced by 10%, leading to reduced startup time in the applet.

Japanese names with fullwidth full stop characters were not supported.

Some E/Z and R/S stereodescriptors were not honored correctly.

Some more rarer acids are now supported, including those based on an Antimony atom.

Names containing the "sulfonato" prefix were converted incorrectly.

Japanese names containing the '=' separator character (for instance in esters) were not supported.

Support was added for acetimidamides and the imidamido suffix.

Failures from a custom naming webservice are now reported as an exception, to make it easier to solve misconfiguration of the service.

Conversion of systematic names is improved by supporting more functional groups.

Added support for the olato suffix in inorganic nomenclature.

Parent homology groups such as alkyl or hydroxyalkyl are now generated with an attachment point.

Custom name to structure webservice and CAS number resolution now try to use the system HTTP proxies, when available.

Naming webservice URLs can now contain a [DOMAIN] parameter that get replaced by the Windows domain name of the machine, or by the network domain name on non-Windows computers. This information can be used by the webservice, for instance for simple access control or user-based customization of the result.

Japanese and Chinese names with full-width curly braces were not supported.

Corporate IDs (registry numbers) can be converted using the API and class chemaxon.naming. n2s .DatabaseNameConverter.

Support for organo-metallic names was extended.

Support for CAS names was improved.

Halonium ions are imported correctly.

Substituted ferrocene and analogues are now supported.

CAS number conversion and naming webservice calls could run for a long time or even not return if the webservice was not responsive.

OCR error correction is improved.

Lowercase d- and l- stereo-descriptors are now interpreted as D- and L-, since names are sometimes (wrongly) converted to all lowercase. This is correct most of the time since the old usage of d- and l- for optical activity (dextrorotatory and levorotatory) is out of use and replaced by (+) and (-).

Names with (r) and (s) pseudo-asymmetric stereo centers are now fully supported.

Chinese name to structure coverage was improved by about 3 percentage points.

Additional OCR error correction is performed.

Some names designating multiplied fused systems failed to be converted.

Some names referring to the locant of spiro atoms were failing to be converted to structure or were missing stereochemistry on those atoms.

Some partially saturated polyacenes and polyaphenes were not supported.

Some Chinese and Japanese chemical names with punctuation characters were were not supported by the automatic format recognizer.

Double bonds in large rings that are part of fused systems had unknown stereochemistry. They now have cis stereochemistry.

Some substituents separated by a space from an acid parent were wrongly interpreted as esters.

The naming webservice was not working in the applet.