RE: [PEAR-DEV] RE: [PEAR-CVS] cvs: pear /Cache OutputCompression.php

Hi, I have 3 things to say. First, I personally do not like using die() or exit() in any of my class. Second, Yasuo Ohgaki told me(and php-dev ml) that it might be related to how PHP handles output buffering and he was going to take a look at the PHP source(not sure exactly when though). Thus, when it's fixed, my changes will be safely reverted. Last, why leave unstable option as default? Who does want to use option that might cause their web site to fail serving web pages to end users? Please think more about people who relies on this class. Regards, Naoki Shima naoki@avantexchange.com murahachibu@php.net -----Original Message----- From: Christian Stocker [mailto:chregu@bitflux.ch] Sent: Thursday, April 25, 2002 10:41 PM To: pear-dev@lists.php.net Subject: Re: [PEAR-DEV] RE: [PEAR-CVS] cvs: pear /Cache OutputCompression.php Hi I suggest the attached diff. It adds a second optional parameter to printContent. If this is set to true, it dies with $content, otherwise it prints the content. This should not break any other scripts and still has the possibility to die :) If noone objects, i will commit it (then we still have a die in pear, but not by default..) chregu On Thu, 25 Apr 2002, Naoki Shima wrote: > The server where I deployed serves around 1000 connections at a time. > I couldn't find any other way to workaround this problem. > I believe that PEAR needs to be as stable as it can be and also works > in a blackbox way. > I have submitted this bug to bugs.php.net and hope it's gonna be fixed soon. > I think adding inline documentation might help not to confuse people. > If you have any other way to workaround without modifying the API, > please let me know. > > Best regards, > > Naoki Shima > -----Original Message----- > From: Christian Stocker [mailto:usenet@chregu.tv] > Sent: Thursday, April 25, 2002 9:03 PM > To: pear-dev@lists.php.net; pear-cvs@lists.php.net > Subject: RE: [PEAR-CVS] cvs: pear /Cache OutputCompression.php > > > In <LLEGIABELPNEABIBKDFEOEOCCJAA.naoki@avantexchange.com>, Naoki Shima > wrote: > > >> could you please explain this? > >> tc > > > > I have submitted this bug report to PEAR-DEV few days ago. The copy of > > the message sent follows: > > I'm really really not happy about having a die() in the "middle" of a > class.. We should either make a function printAndDieContent() or you do > the die() by yourself in your code. But having this die within a method, > will certainly confuse some people. > > What do others think about that? > > chregu > > > > > I've deployed PEAR Cache OutputComression on fairly busy site. It worked > > well when I viewed the page via Netscape 6.2. However, when I viewed the > > page using > > IE 6.0/5.0, it just displays white page as when data is not sent to the > > browser. > > When I was hacking the code in Cache/OutputCompression.php, I found out > > that it needed to call exit() just after printContent() does the > > print(). I have modified "print $content;" to "die($content);" in > > printContent(). > > I've confirmed that destructor is called even when die() is called > > during the execution by calling > > error_log() to send mail inside the destructor, so I guess the > > garbageCollection() is called properly. There're about 200 processes of > > apache on 5 of each load-balanced servers and it seems working fine for > > 3 days now. > > I could not find any other way to work around this problem. Should I > > commit this change to cvs? > > > > Naoki Shima > > naoki@avantexchange.com > > murahachibu@php.net > > -- > PEAR CVS Mailing List (http://pear.php.net/) > To unsubscribe, visit: http://www.php.net/unsub.php > > > > -- nam...christian stocker adr...bremgartnerstr. 66, ch-8003 zurich pho...+41 1 451 6021 www...http://phant.ch/chregu mob...+41 76 561 8860 ema...chregu@phant.ch wor...+41 1 240 5670 gpg...0x5CE1DECB