Re: Mystery Binary Characters in My Files

At 06.04.2002 08:36, you wrote:

I read in the contents of an old mysql database, and made XML files out of the data contained in it. The total job came out to about 6800 files, or "Documents". I'm finding that, sporadically, when reading these XML files, and passing the xmldata to sablotron, I'm getting sablotron errors. These errors stem from characters I'm finding throughout these documents. This is the meat of my problem. I take a look at a file which is giving errors by using 'less'. When looking at the file, I'll see chars like this: <A1> or <91> or <92> and so on, and so on. The chars are hilighted in bold reverse text, indicating that they are 'binary'??. I'm not sure whether to call them binary or hex... perhaps someone can tell me how to appropriately address these chars... Anyhow... these chars sometimes correspond to valid characters. Such as <A9>... this is a "copyright" char, or &amp;copy. I've been manually replacing these characters as errors are generated, but it's getting a little tiring. Is there anyway I can force PHP to either strip out these 'binary' characters, or whatever they are, when I read the file? Is there any way to keep php from saving these chars to NEW documents when they are created? Does anyone even know what I'm talking about?? hehe. Seems that you work on a win*box, as a tip (don´t know if it works, as I use LAMP)

try manual -> string functions -> get_html_translation_table — Returns the translation table used by htmlspecialchars() and htmlentities() htmlentities — Convert all applicable characters to HTML entities htmlspecialchars — Convert special characters to HTML entities Maybe here you´ll find something that make you happy. Otherwise, as I learned the last two weeks on the list, ereg_replace or eregi could make your life easier. HTH Oliver