Re: package validation (was Re: [PEAR-QA] Re: PHP_Beautifier-0.0.6.1
| From: | Greg Beaver | Date: | Fri, 04 Jun 2004 17:35:56 +0000 |
| Subject: | Re: package validation (was Re: [PEAR-QA] Re: PHP_Beautifier-0.0.6.1 | ||
| References: | 1 2 3 | Groups: | php.pear.qa |
| Request: | Send a blank email to pear-qa+get-1373@lists.php.net to get a copy of this message | ||
Tomas V.V.Cox wrote:
Stefan Neufeind wrote:You really need to check out the code in PEAR-1.4.0dev5.tgz, in particular PEAR_PackageFile and PEAR_ChannelFile classes. A package name validation regex is already in channel.xml and is fully supported by the installer. Far better than doing this in pre- and post- scripts is to leverage the existing power of polymorphism. Simply allow users to extend PEAR_PackageFile, and redefine the validation of package version. Most of this stuff should be server-side, and not in the pear packager, as things like this tend to change as the political decisions change. The only other alternative is to add a field to the channel.xml for describing version validation, which would be far more complex than I'm willing to deal with. GregOn 4 Jun 2004 at 14:32, Tomas V.V.Cox wrote:Well, is not trivial at all, remember that pear is not PEAR only. The goal of pear (the pear cmd) is to be an universal installer for software written in php, so we can't enforce our rules. Even with channels comming fast this point will become worse. The pear installer needs urgently, apart of a decent DTD validation, one of these two: a) Remove any PEAR coding standar rule that is hardcoded today. b) Provide something like a plug-in system which validates the package.xml fields according to the rules of each "channel" developers.Stefan Neufeind wrote: I didn't know that was even supported by the installer, maybe the solution is just to restrict version numbers to three digits in package.xml validation.Yes, to strictly restrict - no more and not less then three (afaik we had a two-digit-number recently as well). Volunteers that have maybe already worked on the package.xml-validation? Sounds like a quickfix to me ...