[PEPr] Comment on RFC::EvalForbiddance

Philippe Jausions (http://pear.php.net/user/jausions) has commented on the proposal for RFC::EvalForbiddance. Comment: It should also be added that a workaround such as writing code to a file, then including it, should be avoided just as well. Another rational against the use of eval: performance. The code is evaluated each time making it impossible to be code-byte cached by optimizers. What would really be useful is actual peer-review. But everybody is time-crunched with their own packages, it could be difficult to keep on eye on everybody's else package. Maybe some automated scan and post of "suspicious" code to the PEAR-dev list on a regular basis would help for that. (yes I know there is the CVS list.) At least if the existing packages are scanned, we'll know where we stand. Do we know how many packages are actually currently using eval()? There may not be a reason for a complete ban, if the eval() is not widely improperly used. Proposal information: http://pear.php.net/pepr/pepr-proposal-show.php?id=288 -- Sent by PEPr, the automatic proposal system at http://pear.php.net