Raivo Laanemets
Raivo Laanemets
I could see `::matchAll(regex)` be useful too in the case when multiple matching parts from the element/attribute/property text needs to be extracted. Maybe this goes into a separate issue?
@gajus > Do you have any example of when such would be the case? I propose to label feature proposals lacking an use case with the label `needs-use-case`. This helps...
https://github.com/jacobfriedman/p2p-prolog/issues/1#issuecomment-701250912 I will go over the build process again some time but the current instructions skip wasm zlib build.
@wouterbeek Yes, go ahead. More debugging output for packs is always good.
@tniessen, thank you! That's awesome. Despite these issues, it seems like porting scryer would be easier. - SWI-Prolog wasm version also does not support GMP; - Unless using threads, terminal...
@JanWielemaker where could I see the tests? They do not seem to be in swipl-devel nor in the pengines repo. The headless Chrome/Chromium approach will require Node as a dependency...
I have some basic automated test setup working using Puppeteer. After I have some actual tests, I will submit PR to swipl-devel.
Related: * Term representation (lazy JavaScript objects?). * String encoding (wchar_t does not seem to be UTF32). * Calling JavaScript from Prolog.
The new term representation should also be ported to . That interface also has string encoding issues.
I have looked at the JPL term representation and it has excessive (imho) wrapping of atoms and numbers, possibly as a requirement by the Java type system to represent term...