I would like to clarify what is the proper way to filter user input with php. For example I have a web form that a user enters information into. When submitted the data from the form will be entered into a database.
My understanding is you don't want to sanitize the data going into the database, except for escaping it such as mysql_escape_string, you want to sanitize it when displaying it on the front end with something like htmlentities or htmlspecialchars. However if you want you can validate/filter the user input when they submit the form to make sure the data is in the proper format such as if a field is for an email address you want to validate that it has the proper email format. Is that correct?
My next question is what do you do with the data when you re-display it in a web form? Lets say the user is allowed to edit the information in that form after they filled it out and the information was added to the database. They then go back in and see the data in the fields they originally entered, do you have to sanitize the data for it to show correctly in the form fields? For example there is a field called My Title, the person enters My title is "Manager". You see the quotations around manager, when you display it as is into the form field it breaks because of the quotations:
<input type="text" name="title" value="My title is "Manager"">
So don't you have to do something like htmlentities to turn the quotations into its html entities? Otherwise the value of the field would look like My title is
Hope this makes sense.
Nothing says you can't sanitize data before database insertion. After all, if your script/site/company has a certain policy regarding what's acceptable in a form field, it's best to strip out anything that's not allowed before saving it. That way you only sanitize once, before data insertion/update, rather than EVERY TIME you retrieve the data.
If you allow HTML entities for (say) accented characters, but not HTML tags, then you have to both check for invalid entities (&foobar;?) and HTML tags as well. Since you don't allow them, don't bother storing them. If you require a valid email address, then check if it's at RFC 5322 compliant and only store it once the user's entered proper data. (Whether that email address actually exists is another matter).
Now, let's get one thing straight. There's a difference between sanitization and escaping. Sanitization means literally to clean up - you're removing anything you don't want from the data. You can either silently drop it, or present an error to the user and tell them to fix it. On the other hand, escaping is just a means of encoding data so it's displayed properly.
With your My title is "Manager" string, you don't need to sanitize it, as there's nothing really wrong or offensive about it. What you do need to do is escape it, with at least htmlspecialchars(), so that the embedded double quotes don't "break" your form. If you embed it verbatim, most browsers will see it as having value="My title is" and some bogus attribute/garbage Manager"". So, you run it through htmlspecialchars and end up My title is "Manager", which embeds into the value="" perfectly with no trouble. No sanitization, just proper encoding.
Now, when that form is submitted, then you do have to sanitize/validate again, as the data's been in the hands of a potentially malicious user, and the data could have been changed to My title is <script>document.location='http://attacksite.com';</script>pwn me.
Basically, the workflow should be:
present form to user
get data submitted.
sanitize data
if form is not correctly filled out, displays errors and go to 1)
escape data for sql query
insert into database
then later
retrieve data from database
escape/encode as appropriate for however it will be displayed
display data. if data's going into a form, do 1-6 as before.
Related
I have a simple app programmed in PHP using CodeIgniter 4 framework and, as a web application, it has some HTML forms for user input.
I am doing two things:
In my Views, all variables from the database that come from user input are sanitized using CodeIgniter 4's esc() function.
In my Controllers, when reading HTTP POST data, I am using PHP filters:
$data = trim($this->request->getPost('field', FILTER_SANITIZE_SPECIAL_CHARS));
I am not sure if sanitizing both when reading data from POST and when printing/displaying to HTML is a good practice or if it should only be sanitized once.
In addition, FILTER_SANITIZE_SPECIAL_CHARS is not working as I need. I want my HTML form text input to prevent users from attacking with HTML but I want to keep some 'line breaks' my database has from the previous application.
FILTER_SANITIZE_SPECIAL_CHARS will NOT delete HTML tags, it will just store them in the database, not as HTML, but it is also changing my 'line breaks'. Is there a filter that doesn't remove HTML tags (only stores them with proper condification) but that respects \n 'line breaks'?
You don't need to sanitize User input data as explained in the question below:
How can I sanitize user input with PHP?
It's a common misconception that user input can be filtered. PHP even
has a (now deprecated) "feature", called
magic-quotes,
that builds on this idea. It's nonsense. Forget about filtering (or
cleaning, or whatever people call it).
In addition, you don't need to use FILTER_SANITIZE_SPECIAL_CHARS, htmlspecialchars(...), htmlentities(...), or esc(...) either for most use cases:
-Comment from OP (user1314836)
I definitely think that I don't need to sanitize user-input data
because I am not writing SQL directly but rather using CodeIgniter 4's
functions to create SQL safe queries. On the other hand, I do
definitely need to esc() that same information when showing to avoid
showing html where just text is expected.
The reason why you don't need the esc() method for most use cases is:
Most User form input in an application doesn't expect a User to submit/post HTML, CSS, or JavaScript that you plan on displaying/running later on.
If the expected User input is just plain text (username, age, birth date, etc), images, or files, use form validation instead to disallow unexpected data.
I.e: Available Rules and Creating Custom Rules
By using the Query Builder for your database queries and rejecting unexpected User input data using validation rules (alpha, alpha_numeric_punct, numeric, exact_length, min_length[8], valid_date, regex_match[/regex/], uploaded, etc), you can avoid most potential security holes i.e: SQL injections and XSS attacks.
Answer from steven7mwesigwa gets my vote, but here is how you should be thinking about it.
Rules Summary
You should always hold in memory the actual data that you want to process.
You should always convert the data on output into a format that the output can process.
Inputs:
You should strip from all untrusted inputs (user forms, databases that you didn't write to, XML feeds that you don't control etc)
any data that you are unable to process (e.g. if you are not able to handle multi-byte strings as you are not using the right functions, or your DB won't support it, or you can't handle UTF8/16 etc, strip those extra characters you can't handle).
any data that will never form part of the process or output (e.g. if you can only have an integer/bool than convert to int/bool; if you are only showing data on an HTML page, then you may as well trim spaces; if you want a date, strip anything that can't be formatted as a date [or reject*]).
This means that many "traditional" cleaning functions are not needed (e.g. Magic Quotes, strip_tags and so on): but you need to know you can handle the code. You should only strip_tags or escape or so on if you know it is pointless having that data in that field.
Note: For user input I prefer to hold the data as the user entered and reject the form allowing them to try again. e.g. If I'm expected a number and I get "hello" then I'll reload the form with "hello" and tell the user to try again. steven7mwesigwa has links to the validation functions in CI that make that happen.
Outputs:
Choose the correct conversion for the output: and don't get them muddled up.
htmlspecialchars (or family) for outputting to HTML or XML; although this is usually handled by any templating engine you use.
Escaping for DB input; although this should be left to the DB engine you use (e.g. parameterised queries, query builder etc).
urlencode for outputting a URL
as required for saving images, json, API responses etc
Why?
If you do out output conversion on input, then you can easily double-convert an input, or lose track of if you need to make it safe before output, or lose data the user wanted to enter. Mistakes happen but following clean rules will prevent it.
This also mean there is no need to reject special characters (those forms that reject quote marks are horrible user experience, for example, and anyone putting restrictions on what characters can go in a password field are only weakening security)
In your particular case:
Drop the FILTER_SANITIZE_SPECIAL_CHARS on input, hold the data as the user gave it to you
Output using template engine as you have it: this will display < > tags as the user entered then, but won't break your output.
You will essentially sanitize each and every output (that you appear to want to avoid), but that's safer than accidentally missing a sanitize on output and a better user experience than losing stuff they typed.
From my understanding,
FILTER_SANITIZE_SPECIAL_CHARS is used to sanitize the user input before you act on it or store it.
Whereas esc is used to escape HTML etc in the string so they don't interfere with normal html, css etc. It is used for viewing the data.
So, you need both, one for input and the other for output.
Following from codeigniter.com. Note, it uses the Laminas Escaper library.
esc($data[, $context = 'html'[, $encoding]])
Parameters
$data (string|array) – The information to be escaped.
$context (string) – The escaping context. Default is ‘html’.
$encoding (string) – The character encoding of the string.
Returns
The escaped data.
Return type
mixed
Escapes data for inclusion in web pages, to help prevent XSS attacks. This uses the Laminas Escaper library to handle the actual filtering of the data.
If $data is a string, then it simply escapes and returns it. If $data is an array, then it loops over it, escaping each ‘value’ of the key/value pairs.
Valid context values: html, js, css, url, attr, raw
From docs.laminas.dev
What laminas-Escaper is not
laminas-escaper is meant to be used only for escaping data for output, and as such should not be misused for filtering input data. For such tasks, use laminas-filter, HTMLPurifier or PHP's Filter functionality should be used.
Some of the functions they do are similar. Such as both may/will convert < to <. However, your stored data may not have come just from user input and it may have < in it. It is perfectly safe to store it this way
but it needs to be escaped for output otherwise the browser could get confused, thinking its html.
I think for this situation using esc is sufficient. FILTER_SANITIZE_SPECIAL_CHARS is a PHP sanitize filter that encode '"<>& and optionally strip or encode other special characters according to the flag. To do that you need to set the flag. It is third parameter in getPost() method. Here is an example
$this->request->getPost('field', FILTER_SANITIZE_SPECIAL_CHARS, FILTER_FLAG_ENCODE_HIGH)
This flag can be change according to your requirements. You can use any PHP filter with a flag. Please refer php documentation for more info.
I have a website which runs on PHP and a MySQL database. I was wondering how to best treat user input in regard to HTML encoding (I am well aware that I should store as received and decode in output: that's what I do) and this cycle in particular:
user registers filling in a form with a username field, the content of the field is validated and sent and stored in the DB as is (no HTML encoding) as it will be required to output HTML, XML, JSON, plaintext and other formats;
on any page requiring the username to be shown, it will be fetched from the database, HTML-encoded and displayed in the page;
on a particular page the username is placed in the "value" field of an html text input: obviously this means that the username must be HTML encoded (otherwise XSS and all those fantastic things...). However this also means that if the original username was "però" the text field will be <input value="però"> and when the user submits it the server will receive però instead of però.
Now my question is: should the server decode all the received inputs so that però gets decoded to the original però?
My doubt is that this would mean that if an user inputs è as his username it will be registered as è and not as he actually intended...
I know this is not such a big problem (don't know of many users which would want to use HTML special characters encoding literals in their usernames...), but it puzzled me and I could not find a completely satisfying solution.
Unless I've misunderstood what you're asking, you seem to have the wrong impression about the effect of outputting HTML encoded strings into text inputs. Here's a basic example of what will happen. Let's say you have a user who wants to be named PB&J. Sure, it's weird, but not everyone can pick a nice non-weird username like "Bonvi" or "Don't Panic".
So you save that in your database as is.
Later, when you're using it in another form, you escape it for output.
<input type="text" name="username" value="<?= htmlspecialchars($username) ?>">
In your page source, you'll see
<input type="text" name="username" value="PB&J">
with the ampersand converted to an HTML entity. (Which is what you want, in case they really wanted to be named bob"><script>alert("però!")</script><p class="ha or something worse.)
But the value displayed in the text box will be PB&J, and when the user submits the form, the value in $_POST['username'] will be PB&J, not PB&J. It will not be changed to the encoded value.
(I used htmlspecialchars in this example, but the same would apply with your example using però with htmlentities.)
I'm trying to explain it basically, so I apologize if I did misunderstand you - I don't intend to sound condescending.
So I have been working on an API for a little side project and all the user input won't be directly used for SQL or displayed to the users so do I need to sanitize it? If I do what should I do exactly?
I am currently checking to make sure if they integers, strings, arrays, etc. but other than that is there anything else I need to do?
The question is always for what purpose? If you just take values you get from a user and you don't do anything with them (you just store and display them), then there's nothing to sanitize. If you let those values "actively do" something, you may want to sanitize them to avoid them doing something you don't like.
For instance, you accept HTML input from a user and want HTML formatted content, but you want to avoid XSS problems; in this case you will want to selectively remove HTML elements, i.e. you want to sanitize the input.
// some HTML is allowed, but not everything
echo remove_unwanted_html_elements($_POST['content']);
If OTOH you do not allow HTML input to be interpreted anyway, i.e. whatever the user posts is just displayed literally back to him without any of being interpreted as HTML, then you do not need to sanitize anything. You may just need to escape the content according to its target format.
// don't care what the user enters, just display it right back as is
echo htmlspecialchars($_POST['content']);
Sanitization is only relevant if you evaluate the value in some not entirely predictable way. Sanitization means to take a value and change it into something else, typically removing something from it. This must be very targeted and purposeful since it can be a very error prone operation; you don't just sanitize data somehow just because. The other alternative is simple validation, i.e. checking that a value conforms to expected norms and otherwise rejecting it outright.
Even taking a supposed number entered by the user and casting it to an int is a very simple form of sanitization; it's effective since it means you are guaranteed to get a harmless number, but that number may or may not have anything to do with the value the user submitted. Validation may be the better option here.
I am trying to figure out what is the best way to manage the data a user inputs concerning non desirable tags he might insert:
strip_tags() - the tags are removed and they are not inserted in the database
the tags are inserted in the database, but when reading that field and displaying it to the user we would use htmlspecialchars()
What's the better, and is there any disadvantage in any of these?
Regards
This depends on what your priority is:
if it's important to display special characters from user input (like on StackOverflow, for example), then you'll need to store this information in the database and sanitize it on display - in this case, you'll want to at least use htmlspecialchars() to display the output (if not something more sophisticated)
if you just want plain text comments, use strip_tags() before you stick it in the database - this way you'll reduce the amount of data that you need to store, and reduce processing time when displaying the data on the screen
the tags are inserted in the database, but when reading that field and displaying it to the user we would use htmlspecialchars()
This. You usually want people to be able to type less-than signs and ampersands and have them displayed as such on the page. htmlspecialchars on every text-to-HTML output step (whether that text came directly from user input, or from the database, or from somewhere else entirely) is the right way to achieve this. Messing about with the input is a not-at-all-appropriate tactic for dealing with an output-encoding issue.
Of course, you will need a different escape — or parameterisation — for putting text in an SQL string.
The measures taken to secure user input depends entirely on in what context the data is being used. For instance:
If you're inserting it into a SQL database, you should use parameterized statements. PHP's mysql_real_escape_string() works decently, as well.
If you're going to display it on an HTML page, then you need to strip or escape HTML tags.
In general, any time you're mixing user input with another form of mark-up or another language, that language's elements need to be escaped or stripped from the input before put into that context.
The last point above segues into the next point: Many feel that the original input should always be maintained. This makes a lot of sense when, later, you decide to use the data in a different way and, for instance, HTML tags aren't a big deal in the new context. Also, if your site is in some way compromised, you have a record of the exact input given.
Specifically related to HTML tags in user input intended for display on an HTML page: If there is any conceivable reason for a user to input HTML tags, then simply escape them. If not, strip them before display.
PLATFORM:
PHP & mySQL
For my experimentation purposes, I have tried out few of the XSS injections myself on my own website. Consider this situation where I have my form textarea input. As this is a textarea, I am able to enter text and all sorts of (English) characters. Here are my observations:
A). If I apply only strip_tags and mysql_real_escape_string and do not use htmlentities on my input just before inserting the data into the database, the query is breaking and I am hit with an error that shows my table structure, due to the abnormal termination.
B). If I am applying strip_tags, mysql_real_escape_string and htmlentities on my input just before inserting the data into the database, the query is NOT breaking and I am able to successfully able to insert data from the textarea into my database.
So I do understand that htmentities must be used at all costs but unsure when exactly it should be used. With the above in mind, I would like to know:
When exactly htmlentities should be used? Should it be used just before inserting the data into DB or somehow get the data into DB and then apply htmlentities when I am trying to show the data from the DB?
If I follow the method described in point B) above (which I believe is the most obvious and efficient solution in my case), do I still need to apply htmlentities when I am trying to show the data from the DB? If so, why? If not, why not? I ask this because it's really confusing for me after I have gone through the post at: http://shiflett.org/blog/2005/dec/google-xss-example
Then there is this one more PHP function called: html_entity_decode. Can I use that to show my data from DB (after following my procedure as indicated in point B) as htmlentities was applied on my input? Which one should I prefer from: html_entity_decode and htmlentities and when?
PREVIEW PAGE:
I thought it might help to add some more specific details of a specific situation here. Consider that there is a 'Preview' page. Now when I submit the input from a textarea, the Preview page receives the input and shows it html and at the same time, a hidden input collects this input. When the submit button on the Preview button is hit, then the data from the hidden input is POST'ed to a new page and that page inserts the data contained in the hidden input, into the DB. If I do not apply htmlentities when the form is initially submitted (but apply only strip_tags and mysql_real_escape_string) and there's a malicious input in the textarea, the hidden input is broken and the last few characters of the hidden input visibly seen as " /> on the page, which is undesirable. So keeping this in mind, I need to do something to preserve the integrity of the hidden input properly on the Preview page and yet collect the data in the hidden input so that it does not break it. How do I go about this? Apologize for the delay in posting this info.
Thank you in advance.
Here's the general rule of thumb.
Escape variables at the last possible moment.
You want your variables to be clean representations of the data. That is, if you are trying to store the last name of someone named "O'Brien", then you definitely don't want these:
O'Brien
O\'Brien
.. because, well, that's not his name: there's no ampersands or slashes in it. When you take that variable and output it in a particular context (eg: insert into an SQL query, or print to a HTML page), that is when you modify it.
$name = "O'Brien";
$sql = "SELECT * FROM people "
. "WHERE lastname = '" . mysql_real_escape_string($name) . "'";
$html = "<div>Last Name: " . htmlentities($name, ENT_QUOTES) . "</div>";
You never want to have htmlentities-encoded strings stored in your database. What happens when you want to generate a CSV or PDF, or anything which isn't HTML?
Keep the data clean, and only escape for the specific context of the moment.
Only before you are printing value(no matter from DB or from $_GET/$_POST) into HTML. htmlentities have nothing to do with database.
B is overkill. You should mysql_real_escape_string before inserting to DB, and htmlentities before printing to HTML. You don't need to strip tags, after htmlentities tags will be displayed on screen as < b r / > e.t.c
Theoretically you may do htmlentities before inserting to DB, but this might make further data processing harder, if you would need original text.
3. See above
In essence, you should use mysql_real_escape_string prior to database insertion (to prevent SQL injection) and then htmlentities, etc. at the point of output.
You'll also want to apply sanity checking to all user input to ensure (for example) that numerical values are really numeric, etc. Functions such as is_int, is_float, etc. are useful at this point. (See the variable handling functions section of the PHP manual for more information on these functions and other similar ones.)
I've been through this before and learned two important things:
If you're getting values from $_POST/$_GET/$_REQUEST and plan to add to DB, use mysql_real_escape_string function to sanitize the values. Do not encode them with htmlentities.
Why not just encode them with htmlentities and put them in database? Well, here's the thing - the goal is to make data as meaningful and clean as possible and when you encode the data with htmlentities like Jeff's Dog becomes Jeff"s Dog ... that will cause the context of data to lose its meaning. And if you decide to implement REST servcies and you fetch that string from DB and put it in JSON - it'll come up like Jeff"s Dog which isn't pretty. You'd have to add another function to decode as well.
Suppose you want to search for "Jeff's Dog" using SQL "select * from table where field='Jeff\'s Dog'", you won't find it since "Jeff's Dog" does not match "Jeff"s Dog." Bad, eh?
To output alphanumeric strings (from CHAR type) to a webpage, use htmlentities - ALWAYS!