Clarify distinction between 'source' and 'compiled' template - Githubissues

accordproject / template-archive

Smart Legal Contracts & Templating System

https://accordproject.org/projects/cicero/

Apache License 2.0

281 stars 119 forks source link

Clarify distinction between 'source' and 'compiled' template #258

Closed jeromesimeon closed 5 years ago

jeromesimeon commented 5 years ago

Currently, cicero support both JavaScript and Ergo templates, and archives.

We should consolidate and clarify that distinction so that:

Source archives are always using Ergo as logic
Compiled archives can use JavaScript, but can also be targeting other backends (e.g., JVM, WASM, distributed ledgers)

Some work items could include:

[ ] Decide on naming conventions which distinguish source archives from compiled ones (e.g., .cta for source, .ctjs for JavaScript archive, .ctjava for Java archive, etc.
[ ] Distinguish between cicero archive (always create a source archive) and cicero compile --language XXX to create an artifact/archive for a target XXX
[ ] Distinguish calls cicero parse and cicero execute on source or target, maybe through and option
[ ] Maintain a source link in compiled archives

jeromesimeon commented 5 years ago

This issue is probably moot. The current status answers those questions as follows:

Whether source or target is identified by the language field inside the package.json for the archive (ergo for source, e.g.,javascript as target)
cicero archive allows for a specific language (default is source archive, --language can be used to create a compiled archive)
cicero parse and execute work on either ergo or compiled javascript
unclear we need to create source links

jeromesimeon commented 5 years ago

One last possible item would be to provide a way to check the integrity of a compiled archive. This might be better done through a digital signature or checksum. See Issue #262