| | 31 | * Sphinx and doc strings |
| | 32 | - Zach suggests we use the Sphinx napolean extension included with Sphinx instead of numpydoc. |
| | 33 | - Napolean handles numpy and google doc string formats. |
| | 34 | - Zach will test it and show us good examples of how to format our doc strings. |
| | 35 | |
| | 36 | * Python type hints |
| | 37 | - Discussed the advantages and disadvantages of Python type hints in function arguments and class attributes. |
| | 38 | - Eric points out ChimeraX plugin developers may find these hints confusing. |
| | 39 | - Greg points out speed optimizers for Python are trying to use type hints. Guido has come out of retirement trying to increase CPython speed 2 fold. |
| | 40 | - Zach says type hints produce runtime error when wrong argument type passed that helps him debug code faster. |
| | 41 | - Zach suggests using a Sphinx extension so the type hints do not appear in function signatures in the documentation. |
| | 42 | - Decided that each developer can choose whether to use type hints. |
| | 43 | - Greg thinks might be possible that ChimeraX command registration could introspect function argument type hints instead of declaring the argument types in the registration call. |
| | 44 | |
| | 45 | * Blast threading |
| | 46 | - Zach asks how to avoid blocking after BLAST rest returns results while code is fetching files containing results. Can QThread be used? |
| | 47 | - Could use a Task. |
| | 48 | - Current blast gui in ChimeraX 1.2.5 blocks for 40 seconds on 4v6x chain A (900 hits) while loading results. |
| | 49 | - Suggest Zach complete new Qt blast gui before worrying about optimization. |
| | 50 | |
| | 51 | * mmCIF saving error |
| | 52 | - Greg says user tried to save NMR ensemble where chain names were changed, gave traceback. |
| | 53 | - Will put in error checking. |
| | 54 | - mmCIF save tries to extract info from chimerax data structures that can be inconsistent and needs checking for compatibility with mmcif requirements. |
| | 55 | |
| | 56 | * Roadmap for 1.3 release |
| | 57 | - Everyone update the 1.3 roadmap moving stuff that won't make it into 1.3 to the 1.4 milestone. |
| | 58 | |