Small talk on php regular extraction of image address

I am obsessed with regular rules and keep trying new tricks. First of all, thanks to TNA for the incomplete output of RSS, and then again to SH for his compulsive learning. Without TNA, I wouldn’t have looked at regular expressions, and I didn’t know there were such awesome expressions in the world. If SH wasn’t just saying that he didn’t understand, I wouldn’t have had the guts to figure it out and improve it. To achieve the same goal, regular expressions may not be unique, and there is nothing that cannot be done, it’s just that you didn’t expect it. You can put it this way, regularity is playing with setting rules. I love this kind of thing. Nothing excites me more and makes me feel awesome than setting rules to filter things.

Share some tips on using regular expressions to extract image addresses in the PHP environment:

The html code of the image URL specification is nothing more than

Copy code The code is as follows:

囧1 and 囧2 are not required. To pass XHTML certification, 囧4, 囧5, and 囧6 are essential. 囧3 is the core content and of course it is indispensable.

As far as regularity is concerned, the shortest match I wrote is

Copy code The code is as follows:

(?<=img.+?src=").*?(? =")

However, this does not work in php, it will appear:

Warning: preg_match_all() [function.preg-match-all]: Compilation failed: lookbehind assertion is not fixed length at offset *** in ***

I have been struggling for a long time, but it doesn’t work. What’s the reason? After trying many times, I finally found that the problem is in the zero-width assertion (?<=img.+?src="). In PHP, zero-width assertions do not support unlimited times like "*" and "+". Something, so an error was reported, just change ".+?" to a fixed length. However, it is basically impossible to fix the length between "img" and "src=". Usually, the img and src of the image address are only. It will be separated by a very simple space, but it does not rule out that in some cases there are alt, titlte and other things before src and after img.

Copy the code The code is as follows:

(?<=img.src=").*? (?=")

Copy code The code is as follows:

(?< ;=imgssrc=").*?(?=")

It may be possible, but there is no guarantee that it will be 100% ok.

You may ask, simple

Copy code The code is as follows:

(?<=src=").*?(?=")

No? Normally, yes, but friends who have searched the page should know that in addition to image addresses starting with src, javascript addresses also start with src! Moreover, there are too many mysterious and unpredictable factors hidden in it, so this seemingly short and perfect writing method will not work.

You may also ask, clever and short is not enough, I will list the suffix of the picture, it should be ok, such as

Copy code The code is as follows:

(?<=src=").*?.(jpg|jpeg| gif|png|bmp|JPG|JPEG|GIF|PNG|BMP)

Indeed, this way of writing is really honest, but have you ever seen pictures without suffixes? wwe.com has many examples of this

RAW http://us.wwe.com/content/media/images/Headers/15559182

SmackDown http://us.wwe.com/content/media/images/Headers/15854138
NXT http://us.wwe.com/content/media/images/Headers/15929136
Superstars http://us.wwe.com/content/media/images/Headers/15815850

The URLs above are all pictures, but they don’t have traditional suffixes. It’s no use being honest and you still can’t get them.

What to do? It can still be like this

Copy code The code is as follows:

Different from the above expression, the content of array[0] in this result is not what we want. The image address we want is in array[2]. Why? Because we used 2 (.*?), each "()" thing will automatically exist in a group, and array[0] represents the summary of the results, array[1] contains everything in img and src , array[2] is the turn for the image address we want. This matching method can match not only images with traditional suffixes, but also some image files without suffixes, while not killing other src= files by mistake. Personally, I feel pretty good about it, haha. Of course, if you have any better suggestions, please leave a message immediately, and people around the world will thank you!

What kind of picture do you want? Is it a fixed format or something else? It has to be analyzed in detail.

My suggestion is:

If the format of the image address you want is img space src=, please use: (?<=img.src=").*?(?="), the array is unique, you know.

Otherwise, please use

Let’s talk about php regular extraction of image address

The day before yesterday I wrote a small talk about extracting image addresses using php regular rules, but in fact, extracting the image address in src= is not enough, because there is no guarantee that the address must be an absolute address or a complete address. What if it is relative? If the address is something like:

albums/Candids/thumb_P1050338.jpg
/content/media/touts/5271608/5271654/15320982

What should we do?

Sometimes you need to add http://example1.com/ in front of these addresses, and some even need to add http://example1.com/example2/.../ Therefore, you need to write a rule that meets all the requirements, It’s simply a fantasy. We can only act according to the circumstances and prescribe the right medicine. Sometimes, the knife needs to be cut from the front, sometimes it needs to be cut from the back.

Today, I was surprised to learn something. It turns out that http://example.com/ and http://example.com////// are the same!

http://img3.douban.com/pics/nav/lg_main_a6.png

and

http://img3.douban.com////pics////nav///lg_main_a6.png

Eventually you will arrive

So, if you want to forcibly add a prefix to the two relative addresses mentioned at the beginning to restore them to absolute addresses, it doesn’t matter whether there is a "/" in front of it, just add a "/", "There is a mistake, Didn't let it go." Well, the display will still be normal if there is one more "/", but if there is one less "/", hey, you can't even think about success. I didn't realize this kind of thing at the beginning, so I copied a large section of code and made two copies of the same thing, one with "./." added and one without. I'm from Mars, I'm wasting my time.

Release 2 addresses for public testing of how to obtain images from the webpage:

For any web page, except those that require login: http://xyark.serw5.com/img.php
For Coppermine Photo Gallery system: http://xyark.serw5.com/g.php (if you If you think that the js page that pops up the original image is also needed, I will have to embarrass you)

The general page is an attempt to capture any picture, and the system page is to show what is meant by detailed analysis of specific situations. Those who have tried it will know that the general page does not work for some websites that use the Coppermine Photo Gallery system. What is the reason? It’s that prefix that’s responsible! But the system page can avoid this problem very well.

If you find any bugs during testing, please leave a message to let us know. Please test in a low-key manner, thank you for your cooperation.

Note: The above topics are purely for discussing rules and technology, and cannot be used for illegal purposes. I am not responsible for any cups or tableware that may be damaged due to improper use.

When reprinting, please indicate the original source and author information of the article and this statement in the form of a hyperlink
http://www.blogbus.com/xrspook-logs/85330456.html

Small talk on php regular extraction of image address_PHP tutorial