On 12 April 2011 13:29, Eli Orr <eli.orr@xxxxxxxxxxxx> wrote: > Hi Richard, > > Thanks. > I've already got a solution to simply use Notes++ and save the PHP script > with Save As encoding set to ANSI (It was UTF-8 indeed that creates the BOM...). > > Thanks again > > Eli > > -----Original Message----- > From: Richard Quadling [mailto:rquadling@xxxxxxxxx] > Sent: Tuesday, April 12, 2011 2:59 PM > To: Eli Orr > Cc: php-general@xxxxxxxxxxxxx > Subject: Re: Eliminatimg PHP UTF-8 BOM in a returned stream to a Mobile App > > On 12 April 2011 12:50, Richard Quadling <rquadling@xxxxxxxxx> wrote: >> On 12 April 2011 11:59, Eli Orr <eli.orr@xxxxxxxxxxxx> wrote: >>> >>> Hi Richard, >>> >>> Thanks. >>> Indeed, that is the case - I've included a code that has UTF-8 string >>> contants -so I guess the PHP parser set the UTF-8 mode to ON so that the returned string to the client has the UTF-8 BOM. >>> >>> It is not a big issue as the mobile app guys aware of this and make the proper 3 bytes offset. >>> Anyhow I was looking for a service to control that behaviour. >>> >>> ÂThanks >>> >>> Eli >>> >>> -----Original Message----- >>> From: Richard Quadling [mailto:rquadling@xxxxxxxxx] >>> Sent: Tuesday, April 12, 2011 12:45 PM >>> To: Eli Orr >>> Cc: php-general@xxxxxxxxxxxxx >>> Subject: Re: Eliminatimg PHP UTF-8 BOM in a returned stream to >>> a Mobile App >>> >>> 2011/4/12 Eli Orr <eli.orr@xxxxxxxxxxxx>: >>>> Dear PHP Gurus, >>>> >>>> I would like to Eliminate the 3 UTF-8 BOM enforced on my returned BLOB: >>>> >>>> The PHP server adds Âutf-8 BOM (UTF-8 Byte Order Mark Â- in the >>>> beginning of >>>> UTF-8 Âfiles) which >>>> consists of three bytes: EF BB BF. >>>> >>>> The Mobile App served by the server Does not need that. How can I >>>> eliminate it?? >>>> >>>> Thanks. >>>> >>>> UTF-8 Byte Order Mark â BOM: >>>> http://unicode.org/faq/utf_bom.html#BOM >>>> >>>> Best Regards, >>>> >>>> Eli ÂOrr >>>> CTO & Founder >>>> Mimmage.com >>>> My virtual vCard >>>> LogoDial Ltd. >>>> M:+972-54-7379604 >>>> O:+972-74-703-2034 >>>> F: +972-77-3379604 >>>> >>>> Plaut 10, Rehovot, Israel >>>> Email: Â Eli.Orr@xxxxxxxxxxxx >>>> Skype: Âeliorr.com >>>> >>>> >>>> ----- >>>> >>>> >>>> >>>> -- >>>> PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: >>>> http://www.php.net/unsub.php >>>> >>>> >>> >>> Can you show us the PHP script that DOES output the BOM? >>> >>> Normally, PHP doesn't do this automatically (AFAIK). The main reason being is that it is often the case that the BOM appears in the source code file before the <?php opening tag, so would block headers (session cookie for example). >>> >>> If a BOM is being issued by PHP, it is being done programmatically, or is being missed due to the initial source code file having the BOM set. >>> >>> See http://docs.php.net/manual/en/function.session-start.php#102431, >>> http://docs.php.net/manual/en/function.header.php#95864, etc. >>> >>> Now. Having said all of that, you may find you are using some sort of output buffering and that is setting the BOM after the headers are sent. >>> >>> But, as it stands, PHP will not be generating the BOM for you. >>> >>> -- >>> Richard Quadling >>> Twitter : EE : Zend >>> @RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY >>> >>> -- >>> PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: >>> http://www.php.net/unsub.php >>> >>> >>> >> >> No. The parser does not _ADD_ the BOM. >> >> The bom already exists in your source code. Nothing to do with PHP. >> >> The file you included that has the UTF-8 constants has the BOM. >> >> You need to edit that file and remove the BOM. The actions you need to >> take will depend upon your editor. >> >> -- >> Richard Quadling >> Twitter : EE : Zend >> @RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY >> > > To be a bit more specific... > > The BOM is the 3 bytes you correctly identified earlier. > > Most editors won't show these when you edit the files. > > But, for the sake of argument, let's just pretend they are visible and look like ... > > #@& > > In the php script that contains some UTF-8 constants, the file would look like ... > > #@&<php > echo 'ï'; // The Fullwidth Won sign. > ?> > > As PHP will only actually parse the content between <?php and ?>, the #@& string (the BOM) is simply sent straight through to the web server > -> the client with no interruption. > > Now, if your code was ... > > #@&<php > session_start(); > ?> > > you would see the headers already sent error message, as the BOM tells the webserver that data is now being received and to send any headers it already has. > > So when the session_start() wants to send the session cookie (which is done as a HTTP Header), PHP already knows some content has gone (the > BOM) and reports the error. > > To iterate, PHP is NOT generating the BOM. You already did that in your code. Well, the editor did it for you. > > Ideally, you want to turn off the BOM in your editor. > > -- > Richard Quadling > Twitter : EE : Zend > @RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY > > -- > PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php > > > Hmmm. You really want to save the code as UTF-8 without BOM. What is the editor? The references for Notes++ I can find is for a post-it notes app. -- Richard Quadling Twitter : EE : Zend @RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
