Re: Weird parsing error in Pike 8.1
Chris Angelico <[email protected]>
| Newsgroups | gmane.comp.lang.pike.user |
|---|---|
| Message-ID | <CAPTjJmruV8D+kw5jYgMcryHS++d5bOCDUW8yLiXjeA1OFoW2yQ@mail.gmail.com> |
On Thu, Nov 20, 2014 at 8:02 PM, Stephen R. van den Berg <[email protected]> wrote: > Chris Angelico wrote: >>Well, I don't know what the actual cause of the problem is, but I >>found a stick that I can poke the bug with. Have a look at commit >>22dbe6 - there was a typo in an apparently-unrelated piece of code, >>the final _processloop function. I've no idea why that didn't result >>in its own compilation error. But with that fixed, reverting c7de86 >>doesn't reinstate the behaviour you referred to. Is that any sort of >>clue? It doesn't say anything to my mind. > > That's not a typo actually. > As of yesterday, there is a conxion and a conxiin class. > The conxiin is the one that does the read buffering, the conxion > is the one that does the write buffering (and has a conxiin as a member). > > So, please revert this patch. Ooooops! I thought I did a quick "search across files" to check for any other references to conxiin, and didn't come across any... but I must have done that in the wrong directory, or something. Sorry about that. Reverted. It did still poke the bug, though, so maybe there's something to be learned. What, I don't know, though. ChrisA