Re: [BlueObelisk-discuss] CML in Python

2025-07-31 Thread Peter Murray-Rust via Blueobelisk-discuss
That's great. (You can tell I have been elsewhere for several years!). I've now switched to Python for about 4-5 years and wondered if there was demand. At heart CML is a structured ontology/knowledge graph and it doesn't matter whether it's Java or Python (or possibly JS), and whether it's XML or

Re: [BlueObelisk-discuss] CML in Python

2025-07-31 Thread Geoffrey Hutchison
There’s already a chemical JSON, inspired by CML. It’s used by Avogadro and has support in OpenBabel. Several computational packages also support it. https://github.com/OpenChemistry/chemicaljson https://jcheminf.biomedcentral.com/articles/10.1186/s13321-017-0241-z If you have suggestions or ide

Re: [BlueObelisk-discuss] CML in Python

2025-07-31 Thread Peter Murray-Rust via Blueobelisk-discuss
Thx the content of simple CML documents can probably be translated/moved to JSON fairly easily. There are a few XML constructs which probably no-one uses such as or processing instructions and could be obsoleted. The ability to interact with other XML languages (MathML, SVG) is more serious. Can J

Re: [BlueObelisk-discuss] CML in Python

2025-07-31 Thread Egon Willighagen
I am fine with Java (and can use the CDK and other Java tools in Python). But CML (Core) in JSON and/or JSON-LD and/or YAML could be interesting. Egon On Tue, 29 Jul 2025 at 16:17, Peter Murray-Rust via Blueobelisk-discuss < blueobelisk-discuss@lists.sourceforge.net> wrote: > Is there a need (o

[BlueObelisk-discuss] CML in Python

2025-07-29 Thread Peter Murray-Rust via Blueobelisk-discuss
Is there a need (or desire) for a Python implementation of CML? (Or does it already exist?) . Given today's tools (such as Cursor) it should be fairly easy to implement the current schema and CML-XOM in Python. In particular if there are conformance tests in Java or other languages that's a strong