Re: Revert Tokenizer behavior for 5.4

On Tue, Sep 13, 2011 at 22:08, Ferenc Kovacs <tyra3l@gmail.com> wrote: > On Tue, Sep 13, 2011 at 9:56 PM, Christopher Jones > <christopher.jones@oracle.com> wrote: >> >> >> On 09/13/2011 12:25 PM, Nikita Popov wrote: >>> >>> I changed my previous patch to an SVN patch, so it is easier to apply >>> and added another patch (the one called "tokenizer_patch_full.txt"), >>> which additionally fetches the rest into a T_INLINE_HTML. (The "_full" >>> patch thus contains both changes. I didn't know how I could separate >>> them.) >> >> I haven't seen much "buy-in" about your email.  You might want to wait >> until after Beta and revisit the topic later.  If you add some concise >> explanation it would be easier for people to weigh the pros and cons, >> see what breaks and what gets fixed etc. > > I agree that the lack of response is kind of disturbing: > no response in the bugreport, no response on the mailing list, almost > zero comment on irc. :/ > I can accept that if it won't make into 5.3, or even 5.4.0, but at > least it could be merged to trunk, if the patch is fine. > I think that Nicolas and Nikita did more than enough on their part, > could somebody please review the patch? the change to ext/tokenizer/tests/token_get_all_variation16.phpt seems really weird? And I'm unsure how this should actually work. A file containing: <?php __halt_compiler() ?> stuff Will return "stuff" after this patch.. that doesn't seem expected. -Hannes