| 1 |
Sérgio Almeida wrote: |
| 2 |
> user action bin { |
| 3 |
> description "Change Python's Version" |
| 4 |
> type sym |
| 5 |
> sym python { |
| 6 |
> bin python |
| 7 |
> target /usr/bin/python |
| 8 |
> prefix /usr/bin/ |
| 9 |
> regexp python([0-9]+\.[0-9]+$) |
| 10 |
> sym python-config { |
| 11 |
> bin python-config |
| 12 |
> destination /usr/bin/python-config |
| 13 |
> prefix /usr/bin/ |
| 14 |
> regexp python([0-9]+\.[0-9]+)-config($) |
| 15 |
> } python-config |
| 16 |
> } python |
| 17 |
> } bin |
| 18 |
> |
| 19 |
> Soon urged the need for more complex lexical analysis and started |
| 20 |
> implementing lex rules and yacc skeleton. |
| 21 |
> |
| 22 |
> With this step a question bounced into my head. |
| 23 |
> |
| 24 |
> Am I reinventing the wheel? |
| 25 |
> Why implement lex/yacc to translate a block of code into a python's |
| 26 |
> block of code? |
| 27 |
> Why not use plain python in modules? |
| 28 |
> |
| 29 |
> After discussing with mentor, we decided to adopt python as the base |
| 30 |
> language for uselect modules. |
| 31 |
|
| 32 |
It seems to me that the original langauge is "static"/"descriptive" |
| 33 |
while Python is not. Why not move to XML or JSON (former seems more |
| 34 |
common with Gentoo) instead of Python? Think about how much easier it |
| 35 |
is to pull information from metadata.xml than from .ebuild files - it's |
| 36 |
the same difference in your case. |
| 37 |
|
| 38 |
You know much better where you want to go with this than I do, but |
| 39 |
please triple-check this move, as you cannot go back. |
| 40 |
|
| 41 |
Thanks for listening, |
| 42 |
|
| 43 |
|
| 44 |
|
| 45 |
Sebastian |
Archives