The Proxomitron

Filter Language   originally at http://www.sankey.ws

Scott R. Lemmon
and
John Sankey

Introduction
Tech Talk
Filter Language
  The Text Matching Language
  Tips and Tricks
  How do I ...? The Proxomitron FAQ
  Limitations
Filter Examples

The Proxomitron
Keeping an eye on the Web.
For You.

The Text Matching Language

The text matching language is the key to understanding how The Proxomitron's filters work. It allows you to match complex combinations of HTML tags and store parts of the matched text into variables which can later be used in the replacement text.

The rules have been specifically designed to make working with HTML easier. For instance, since case is seldom important in HTML, all matching is case insensitive - saving you the trouble of testing for both upper and lower case.

The Proxomitron's meta characters

Here is a quick list of The Proxomitron's meta commands and what they do:

Matching meta characters
*         Match a string of any characters.
?         match any single character.
[abc]     Match any single character listed in the brackets.
[^a-z]    Match any single character not listed in the brackets.
[#n-n]    Prox3 Numeric range match.
[#x:y]    Prox4 Numeric range match.  Supports negative numbers.
" "       Always matches but also consumes any whitespace.
\s        Match string of whitespace only.
\w        Match any number of non-space characters except ">".
\t        Matches a single tab character.
\r        Matches a single carriage return character.
\n        Matches a single newline character.
\0-9      put match into a variable - works like "*" unless following stuff in parentheses: "( ... )\1"
\#        put match into the replacement stack. Each time it is used, the matched item is added to the stack.
&         AND function.
|         OR function.
^         NOT function.
(...)     Group a sub-expression.  Negate with "(^ ... )"
+         Repeat previous match until there are no more left.
++        Same as '+' except matches to up to the point where what follows is true.
+{5}      The + or ++ pattern run will match only 5 repetitions.
+{2,7}    The + or ++ pattern run will match 2 to 7 repetitions.
+{3,*}    The + or ++ pattern run will match 3 or more repetitions.
\         Escape any meta character's special meaning.
=         Magic equal - absorbs leading/trailing spaces.
"         Magic quote - matches double or single quote.
'         Smart ending quote - use to deal with nested quotes.
$LST      include a list in any matching expression
$SET      set a positional variable to a specific value.
$CON      test current connection number
$IHDR     test input headers
$OHDR     test output headers
$URL      test the URL inside the matching portion of a filter.
$CTYP     limit a filter to certain types of pages
$FILTER   force filtering or not
$USEPROXY override remote proxy selection
$SETPROXY force a particular proxy.
$UESC     JavaScript unescape()

Replacement Text Escapes
\0-9      Insert a variable into the replacement text.
\#        Inserts one item from the replacement stack each time it is used - first in first out
\@        Insert all the items in the replacement stack.
\\        Insert a single backslash.
\a        Insert any anchor text from the current URL (anything following a "#").
\d        Insert The Proxomitron base directory in "file://" URL format.
\h        Insert the host portion of the current URL.
\k        Kills the current connection.
\p        Insert the path portion of the current URL.
\q        Insert any query string from the current URL (anything following a "?").
\u        Insert the full URL of the current web page.
<start>   Insert this at the beginning of a page
<end>     Insert this at the end of a page
$RDIR     redirect URL (transparent)
$JUMP     redirect URL (non-transparent)

What they do, in detail:
*matches any string of characters, including no characters at all. For example, "foo*bar" would match "foobar", "fooma babar", or even "foo goat bat bison bar". Basically the asterisk means, "search forward matching anything until you find what follows the asterisk"
?matches any single character no matter what it is. "?oat" would match "boat" or "goat" but not "oat". ?* matches 1 or more characters, but not nothing at all.
^negation (NOT). Used to reverse the action of any of the following tests. Note that a negated expression consumes no characters if it doesn't match any.
[...] matches any single character listed within the [ and ] e.g. [abc] matches "a", "b" or "c". Ranges are checked by using a dash: [A-Z] will match any letter "A" through "Z" while [0-9] will match any single digit. If the first character is a "^" (NOT) it will match any character not within the brackets - [^0-9] will match any character that's not a digit. Note the difference between an escaped character (\t \r \n) and a Proxomitron metacharacter (\w \s etc.). Metacharacters have no meaning within [ ], only escaped characters do. Except for [trn], the escape of any character is the same character e.g. \s is the same as s within [ ]. If ^ is not the first character, and - is the last character, they do not need to be escaped (although it won't do any harm if you do escape them). \ and ] must always be escaped, of course, if they are to be matched.
[#n-n]numeric match. This is used to check for numeric value ranges in HTML tags. For example, to check for a number between 100 and 150 use "[#100-150]". If the second number is a '*' it acts as if it's infinitely large, "[#40-*]" would match any number greater than 40. To check for a number less than 40 simply use "[#0-40]". To check for an exact number the second number can be left out (as in "[#100]"). The numeric match will match regardless of leading zeros or quotes surrounding a number - tag="0100". Note that currently, this rule only works with positive numbers, but you can use the dash to test for negative values: "-[#2-7]" for instance, could be used to match -2 through -7.
  (space)matches a series of spaces (in HTML, tabs and line breaks are spaces), including none at all. Use it where there may or may not be a space between items. For example "tag value" matches "tag value" or "tag  value" or "tagvalue".
\slike the space, but there must be at least one for it to match. For example "tag\s" would match "tag " or "tag  " but not "tag"
\wmatches any number of non-space characters ('word'). It stops when it hits a space or a ">" (which marks the end of a HTML tag). Useful for matching tag values and URLs.
\tmatches a tab character and nothing else. Be careful with this and the following two - they depend upon the system that created the HTML so will be site dependent in their results.
\rmatches a carriage return character and nothing else
\nmatches a newline character and nothing else
\0-9backslash followed by one digit 0-9: save the match into a variable. It matches just like the "*" character, but stores whatever is matched into one of ten variables. These variables can then be used in the replacement text to include parts of the original HTML.
( )\0-9more complex matches can be captured by placing the \0-9 directly after a set of parenthesis with no spaces between, as in "(foo*bar)\1" Then, anything matched within those parenthesis will be placed into the variable. Never forget: "(foo*bar)\1" and "(foo*bar) \1" are completely different things.
\#put the matched text into the replacement stack. Each time it is used, the matched item is added to the end of the stack. Allows recursive substitutions. The maximum size is 20 items.
|the "OR" function. For example "foo|bar" would match either "foo" or "bar".
&the "AND" function. For example "*foo&*bar" would match "foo bar" or "bar foo" but not "foo foo". Note the use of the asterisk - something like this is always needed with the AND function since a word could never be both "foo" and "bar" at the same place and time. AND is useful for situations where tag values may come in any order...
<img src="picture" height=60 width=200>
and
<img width=200 height=60 src="picture">
are both matched by...
<img (*src="picture" & *height=60 & *width=200)*>
(...)parenthesis permit matching sub-expressions within phrases. For example "foo(bar|bear|goat)" would match "foobar", "foobear" or "foogoat". Parenthesis can be nested, as in "foo(bar|(black|brown|puce) bear|goat)" which would match "foobar" "fooblackbear" "foobrownbear" etc.. Also, as with "[...]", if the first character following a "(" is "^" the expression will match only if the expression within does not match. For example, "(^foo|bar)" would match anything that's not "foo" or "bar".
+a run of repeating characters. For instance, "a+" would match "" (i.e. no a's at all), "a" or "aaaaaa". You can use it after other meta characters or square bracket terms for more complex runs. For example, [abc]+ would match a run of any characters "a","b",or "c" like "ababccba". ([a-z]&[^n])+ would match a run of letters "a" through "z" but not "n". (foo)+ would match "foo", "foofoo", "foofoofoo", etc. Note that "a+ab" will never match anything because a+ will "use up" all the "a's" leaving nothing for "ab".
++same as '+' except it matches only up to the point where what follows the ++ is true. So, "a++ab" will match "aaab" because a++ will stop at "ab". This is especially useful for negations: [^/]++microsoft.com will look for microsoft.com but stop at the first /. So, it will match a URL like "qwerty.microsoft.com..." but not "www.somewhere.com/microsoft.com/info.html"
+{}+{5} means that the + or ++ pattern run will match precisely 5 repetitions and nothing else. Use +{2,7} to match 2 to 7 repetitions; +{3,*} to match 3 or more repetitions.
\the backslash is used to "escape" any character that has special meaning and treat it as a normal character. For example, to match a parenthesis in the HTML text use "\(", to match a backslash itself use "\\".
=the equal character matches not only the "=" itself, but also any whitespace before or after - making tests for tag values easier. For example foo="bar" also matches foo= "bar" or foo = "bar"
"a double quote will match either double or single quotes (since either may be used in HTML). for example " * " would match "oh happy mongoose" or 'oh happy mengeese'. If you want to test for double quotes explicitly, use the backslash - \"
'The single quote is smarter than your average quote: It attempts to match the appropriate ending quote for any quote previously matched by the double quote even if there are other quotes in between! Why? In HTML it's common to use a mixture of single and double quotes when you need "quotes within quotes", as in
href=" javascript:window.open( ' bison.html ' ); "
or
href=' javascript:window.open( " bison.html " ); '
Both these are matched by href=( " * ' ) - use the double quote to match the initial quote and the single quote to match the ending quote.
There are some restrictions here: First both the starting and ending quotes must be in the same sub-expression - that means in the same set of nested parenthesis. For example, " some text ', ( " some text ' ), and " ( some text | other text) ' work, but neither " ( some text ' ) nor ( " | ) some text ( ' | ) work. Another restriction is that start and end quotes can't be nested in the same sub-expression: " something " something else ' end of something ' won't work. However, you can nest them using a different sub-expression, like " something ( " something else ' ) end of something '. It's also worth noting that if no previous double quote was matched, the single quote just matches a normal single quote. Still it's safer to use \' to explicitly check for a single quote if you need to.

Matching functions: these do more complicated things than metacharacters.
$LST(filename)include a file in any matching expression. The contents of the file are tested line by line against the text to be matched until a match is found. If a match is found, the expression returns TRUE, otherwise FALSE.
$SET(#)set a positional variable to a specific value. By placing $SET commands within a matching expression, you can set various values if the matching expression reaches that point. This can be used for an if/then/else effect. So
match: name=(one $SET(0=Naoko) | $SET(0=Default))
replace: "\0 Matched"

will produce
"Naoko Matched" if name=one
else "Default matched"
Two details on how The Proxomitron handles variables are needed here. "$SET(\#=\1)" does not make \# equal to the value of \1, instead SET makes \# equal to the string literal "\1". The expansion of \1 isn't done until the match is complete and the filter moves on to the replacement section. Each variable just contains a pointer and length to a bit of text that's already stored somewhere (either in the input buffer, or in the case of SET in the filter itself). So, $SET(\1=\1\2) doesn't work to append to a variable.
$CON:(x,y[,z])Will be true only if the current connection number is 'x' of 'y' (optionally for every 'z' connections). Use to rotate values based on connection. For example, if you want to pretend to be two different people rather than just one, the following for example will alternate between two values in \0 ...
($CON(1,2) $SET(0=personna 1)|$CON(2,2) $SET(0=personna 2))
You might wish to do this with a cookie-faker or browser-identification filter.
$IHDR(name:match)
$OHDR(name:match)
test the value of an HTTP header. The command will be true if the named header's value matches the 'matching' section. $OHDR tests outputed headers while $IHDR tests incoming headers. To only match a "Referer" header that contains 'microsoft.com', use
$OHDR(Referer:*.microsoft.com)
Using these you can have web filter only match if specific header values are also true, or to capture and use header values into a variables to use in a filter's replacement. You can use also them in HTTP header filters to check combinations of headers for a match.
$URL(matching)test the URL inside the matching portion of a filter. You can use the filter's URL match for this, but by using this command you can check for different URL based on the text matched. It's also useful to capture portions of a URL into variables. The following would capture the URL's path...
$URL(www.somehost.com/\1)
As elsewhere, the URL matching starts directly with the hostname so there's no need to match the "http://" portion.
$CTYP(code)Content Type check command. used to limit a filter to only affect certain types of pages (like JavaScript files only). The "code" must be one of the following known types...
htm - Web pages
css - Cascading style sheets
js - JavaScript
vbs - VB Script
oth - Anything else
For more complex content-type checks you can use "$OHDR(Content-Type:some matching value)"
$RDIR(url)The RDIR (redirect) command is used to transparently redirect URLs to a different location. It's also possible to redirect to a local file by using the "http://file//filename" URL command syntax. The new URL must be of a type The Proxomitron understands (http, or with SSLeay, https).
$JUMP(url)Similar to the RDIR command, the JUMP command is used to redirect a URL to a different location. However, JUMP just tells your browser to go to the new location. With JUMP your browser is aware of the redirection and the URL you see will be updated to reflect the new location. It works best for redirecting an entire page, while RDIR is better at invisibly redirecting individual page elements like images or java apps. Use RDIR when you want the redirection to happen "behind the scenes" and use JUMP when you want to simply go to a different site from the one requested. Use both RDIR and JUMP commands in the replacement section of header filters only. It's important to note that for outgoing headers the redirect will happen before the original site is ever contacted, but when used with incoming headers, the initial site must be contacted first. These commands have no effect in web filters since by this point the original page has already begun loading into your browser. In such cases you can often use JavaScript to change to a new location as in...
$FILTER(bool)force a particular request to be filtered or not filtered regardless of it's type. Normally only specific types are filtered (like text/html, text/css, image/gif, etc). $FILTER can be used in the match or replace of any header filter and takes a "true" or "false" value. If true, the request will be run through the web filters regardless of it's type. This only makes sense for content that's text based. You can also use it to avoid "freezing" certain GIF images by using it in a header filter along with a URL match. Take for example...
Out = "True"
Key="URL: Don't freeze this gif"
URL="www.somewhere.com/animate_me.gif"
Replace="$FILTER(False)"
$USEPROXY(bool)override the "Use remote proxy" check box for a single filter. It is used to ensure that a proxy is or isn't used with a given site or for a particular action. To have effect this command must be called in either the match or replace of an outgoing header filter. This is because the proxy setting must be established prior to connecting to the site.
$SETPROXY(proxy:port)force a request to use a particular proxy. It overrides both the "Use remote proxy" checkbox and the current proxy chosen in the proxy selector. The proxy must be one entered into the External Proxy Selector list - SETPROXY simply looks up and sets a proxy from that list. Like the previous command, this command must be called in either the match or replace of an outgoing header filter. (It's usually only necessary to type the first part of the proxy name - the first proxy matched in the list will be used. The partial match must be exact though - no wildcards.)
$UESC(text)similar to the JavaScript unescape(). It will convert most URL escaped characters back to their ASCII form. It's useful for unescaping URLs that may be embedded in other URLs (an increasingly common trick used by many sites to track the links you click). Often characters like ":" and "/" will be escaped by their hex equivilents ("%3A" and "%2F") making the real URL hard to use. $UESC can be used in the replacement text of a filter, and can be given any valid replacement text as input (such as \1 variables). It will convert most escaped characters back to their correct form, but spaces and any non-displayable ASCII characters will remain escaped.

Special Replacement Text codes

These let you put special codes into replacement text:
\0 - \9insert value stored into the corresponding variable from the matching expression
\#insert value stored in the stack - first in, first out
\@insert everything in the stack
\\insert a single backslash
\aincludes any anchor text from the current URL (anything following a "#")
\dincludes The Proxomitron's base directory in a "file://" URL format
\hincludes the host portion of the current URL (before the first /)
\kkills the current connection - useful in HTTP headers to ban specific URLs and in web page filters to skip loading the rest of a page.
\pincludes the path portion of the current URL (after the first /).
\qincludes any query string from the current URL (anything following a "?").
\uincludes the full URL of the current web page.

Note that \a \h \p \q \u only work when you are online - connected to a web page. The algorithm for \h is still under development to try to handle relative URL's and the recent tendency to drop the www. prefix for HTML pages from most but not all domain name servers.

Special URL codes

These commands are inserted before the hostname in the URL. Commands are separated from the URL with either ".." or "//" e.g. http://bypass..www.special.site/ bypasses The Proxomitron when accessing http://www.special.site/. All of these affect only the URL specified in your filter, not others. You can combine these commands e.g. http://src..bypass..www.host.com/specificpage
bypass bypass The Proxomitron totally
binbypass The Proxomitron just for header input
boutbypass The Proxomitron just for header output
bwebbypass The Proxomitron just for web filters
file//filter files on your own computer in the same way as remote files. Useful for checking.
httpsload a 'secure' https: web page without having the local page encrypted. Can be use to access secure pages from browsers that don't directly support https or to avoid the normal https warning messages browsers spit out. The actual remote connection is still encrypted, but The Proxomitron sends the decrypted and filtered page to your browser. (You need the SSL DLL's for this to work.)
load//loads a Proxomitron config file. It can optionally be followed by a '?' and a URL to go to once the config has been loaded e.g. http://load//microsoft.cfg?http://www.microsoft.com/ will load your special config file whenever you visit Microsoft.
srcdisplays the real source of the web page (not just unaffected by The Proxomitron, but unaffected by JavaScript stunts too!)
dbug as src, but also includes The Proxomitron debug stuff. Similar to turning on the "HTML Debug Info" option in the log window, but only for one URL.

Tips and Tricks

The matching rules are the most complex part of The Proxomitron. Understanding them can be confusing at first - especially if you've never used a pattern matching language before. However, don't despair, even very simple rules can accomplish quite a bit. Take it a step at a time and soon it'll become second nature. To get you started, here are a few tips covering some basic HTML matching tasks.

This section assumes you know a little about HTML - if not there are many excellent tutorials available on the net. If you don't intend to write your own rules, you can ignore this information entirely.

Formatting your matching rules

Complex matching rules can often be hard to read. However, to make them more legible, both the matching expressions and replacement text can be split over multiple lines. These line breaks are ignored when deciding what your matching rule means. A techie warning though: spaces are not ignored. This is the reverse of HTML, where a line break is treated exactly the same as a space. So, if you want to tell The Proxomitron "ab" (as opposed to "a b"), and indent lines (as Scott does in his .cfg files), to be safe you should use the format
  "a"
  "b"
But, if you write
"a
b"
(no leading or trailing spaces), you don't need the extra double quotes.

To include a line break in the replacement text or match one in the matching clause use "\n". Spaces always match true (i.e. whether there is a space in the input page or not), so you can normally use them freely to separate elements of a matching expression. Just remember that they consume all spaces they find, and that "<a pplet" is not the same as "<applet"!

Some general info

When designing a new rule it's more common for it to not match at all rather than to match too much. Always start simple - then add refinements as needed. That way, when a rule suddenly doesn't match when you expected it to, you'll have an idea which part is causing the trouble.

Use the log window to see when a filter matches, and use your browser's "view source" option to see the results of a match. These are two very helpful tools when designing rules. Even more helpful is the Match Testing Window which allows you to see exactly how a filter will change a bit of HTML text.

Spaces are very important in many cases. For example, if you want to test an anchor tag <a *</a>, the space after the 'a' is essential, otherwise it would match 'applet'! Always use \s (e.g. <a\s*</a>) after tag names so that no matter what other tag names turn up in the future, The Proxomitron will know that the name must be followed by at least one space to match.

Cutting and pasting HTML

Often a good way to get started on a new filter is to cut and pase the HTML you're interested in directly into the matching clause. Remember: in order to permit formatting rules over multiple lines for clarity, the matching clause ignores line breaks. So, a line that looks like
<br>
<p>
seems just like "<br><p>" to The Proxomitron - no space. This can cause trouble since, in HTML, the newline character behaves as a space. The solution is to place a space at the begining or end of each matching clause line; this will match all "whitespace" including any newlines.

Disabling a tag and tag elements

Since browsers are supposed to ignore any tags and element they don't understand, an effective way to disable a tag or one of its elements is to rename it. This comes in especially useful when the same element may be used by several different tags. Take "onload" for example, this element auto-runs a JavaScript. Although normally in the "<body ... >" tag, it may occur elsewhere as well. To stop it you could use...
Matching: onload=
Replace: LoadOff=

which would make a tag like
<body background="bak.gif" onload="window.open(myadd);" >
become
<body background="bak.gif" LoadOff="window.open(myadd);" >

Notice how simple this rule is! It's a bit risky, since there's a chance the phrase "onload=" could occur outside a tag in the actual text of a web page. In practice however this seldom happens (including the equal sign helps guard against this a bit).

Changing start and end tags

Here's a simple trick for changing both a start and end tag with the same rule. This trick is used by the "Blink to Bold" rule among others. In this rule we want to convert "<blink>" to "<b>" and "</blink>" to "</b>" - Let's take a look how it's done:
Matching: <\1blink>
Replace: <\1b>

By using the "\1" meta character, the rule will match both the start tag "<blink>" and also the end tag: "</blink>". Additionally, the "\1" captures the end tag's slash for use in the replacement text. A safer, but more complex, version of the rule might be...
Matching: < ( / | )\1 blink>
Replace: <\1b>

Can you tell why? If not read Testing for something or nothing.

Capturing a tag's contents

Often you'll want to change only one element of a tag while leaving the rest as they are. This is where the number variable "\#" matching is very useful. Take the following example of a rule to kill web page backgrounds.
Matching: <body \1 background=\w \2 >
Replace: <body \1 \2>

When they don't directly follow parenthesis ( ... )'s the \# variables act just like an asterisk "*". Here, the "\1" captures anything before the background element, while the "\2" captures everything afterwards. In the replacement text, the background element is simply left out, but you could also include your own background here.

Adding an new element to a tag

Here's quick trick to add an element to a tag. Although the "proper" method would be to replace an element if it already exists and add it only if it doesn't, this can sometimes be difficult. It's often simpler to just add the element regardless. We just need to make sure the browser will use our tag instead of any pre-existing one. For example, to add a border to all "<img ... >" tags, you could use
Matching: <img \1 >
Replace: <img border=1 \1 border=1>

Why add border twice? Well, when Netscape finds a duplicate element it uses the first one and ignores the rest, but Internet Explorer uses the last one! By placing the element at the beginning and end of the tag, it works for both. Note that being browser independent isn't as important here as it is for designing web pages. You're likely to know what browser you intend to use, so it's ok to just arrange thing in the way your browser expects unless you plan to publish your code.

Capturing specific tag attribute values

The values of a tag's attributes can often be tricky to match. Take "<a href=... >" for instance. "href" indicates a URL, but the URL value could be surrounded by single quotes, double quotes, or even no quotes at all. (There are some pretty wild approximations of proper HTML out there!) This is where the word match "\w" rule comes in handy. It will match everything, including any quotes it may find, until it hits a space or the end of the tag. For example, if you wanted to capture the URL into the \1 variable you could use
<a * href=(\w)\1 * >

Remember that when a "\#" immediately follows parenthesis it captures whatever text those parenthesis matched. A more interesting example is
<a * href=(\w(banner|advert|cgi)\w)\1 * >
which would only match URLs containing the words "banner", "advert", or "cgi". We now have the beginnings of a "banner blaster" type rule.

Testing for something or nothing

Often you'll find you'll want an expression to match whether a particular value is there or not. To do that use the following rule
"( something | )"
This will first test for the word "something" but if it isn't found the expression is still true. Why? Notice there's an OR symbol (vertical bar) with nothing between it and the closing parenthesis. This creates an empty expression and an empty expression is always true and consumes no characters. Read this as saying - match "something" OR nothing.

Note that if the expression had been written "(|something)", the word "something" would never be matched! Since ORs are processed from left to right, the empty expression would always match first before the word "something" got a chance.

Another example is
( " | ) * ( " | )
which tests for something that may or may not be surrounded by quotes.

Here's a more elaborate example which grabs the "border" value from an "<img ... >" tag if it exists and places it into the variable \1
<img ( * (border=\w)\1 | ) * >
Be careful of the placement of asterisks here, for example
"<img*(border=\w|)\1*>"
might not do what you expect. Upon scanning the first character after "<img ", if it turned out not to be "border" the sub-expression would still match! Then when "border" occurred later in the string, it would be matched by the second asterisk instead, since the initial test had already passed by.

Using "AND" to capture multiple tag attributes

By using the ampersand "&" you can capture certain tag attributes regardless of the order they're found in. For example. Say you wanted to re-write an "<img ... >" tag to contain an image of your own, but wanted to keep the original "width" and "height" values. You might use...
Matching: <img ( (*(height=\w)\1*| ) & (*(width=\w)\2*| ) ) >
Replace: <img src="file://d|/my_pictures/shonen.gif" \1 \2 >

The height is captured into variable \1 and the width into \2. Also by using the "something or nothing" syntax described above, the expression will still match even if the width or height value is missing from the tag. In which case the corresponding \# variables will simply be blank.

Using "smart" quotes

Most of the time "\w" works well for capturing a tag's attributes. However there are times when you need something more. Something like an "<img ... >" tag's "alt" element often contains spaces as in... alt="this is some text" or alt='also some text'. To capture this sort of thing use the double quotes
Matching: alt=( " * " )\1

An even more complex situation often arises with JavaScript. Here it is common to have "quotes within quotes" as in the following
onmouseover="javascript:window.open( ' mywindow.html ' ); "
which could also be written
onmouseover= ' javascript:window.open( " mywindow.html " ); '

This is where the single quote comes in, it looks for the correct closing quote for the previous starting quote. Take the following rule: it can be used to deal with just about any value it comes up against:
Matching: onmouseover=( " * ' | \w )\1
This will get single quotes, double quotes, nested quotes, or even no quotes at all!

Using file URLs to include your own stuff

A "file URL" is a URL that points to a file on your hard drive rather than to some location on the Internet. Browsers use file URLs to view web pages stored off-line, but they are also a very handy way to insert your own pictures, web pages, even JavaScripts into pages you view.

The Proxomitron makes it very easy to insert a file URL into a matching rule's replacement text. First position your cursor where you wish to insert the URL then right-click and select "Insert file URL" from the context menu. A file requester will open up allowing you to choose the file to insert.

Here's an example of a "background replacer" rule that uses a file URL
Matching: <body ( \1 background=\w | ) \2 >
Replace: <body \1 background="file://c:/pictures/background.gif" \2 >

Note the matching expression has a space between the ")" and the "\2". Don't forget this space! This would result in the \2 containing what was matched in the "(...)" phrase instead of whatever follows the background attribute.

Inserting JavaScript or other items in every web page you view.

Here's an trick for really taking control of a web page. JavaScript can be a very powerful tool - in the right hands... Now those hands can be your own! The matching clause can have two special values <start> and <end> - they simply insert the replacement text either at the beginning or end of a web page. They're easy to use and very efficient since no searching has to be done.
Matching: <start>
Replace: <script> onerror=null; </script>

This works great for JavaScripts (and overriding JavaScript functions - see below). There's no need to worry about allowing for multiple matches, and it doesn't matter how badly written the page is (there are pages that omit even the supposedly-required <html> and <title> tags; <head> and <body> are optional even in W3C standards.)

Small scripts like the one above can be included directly in the replacement text, but for larger scripts you may prefer to use a file URL such as
<html>\n<script src= "file://c:/scripts/myscript.js" >

Overriding JavaScript Functions with your own

In Netscape and Internet Explorer 4.0+ there's a very effective trick for warping JavaScripts. Any JavaScript function - even built-in ones - can be redefined to do whatever we want. Say we wanted to get rid of those "alert( ... )" and "confirm( ... )" boxes. We could do it by simply inserting the following script at the start of a web page (use the <start> technique above)
<script>
function alert( ){ return(1); }
function confirm( ){ return(1); }
</start>

Now whenever any other script on the page attempts to call an alert or confirm box, our functions get called instead. By returning a "1" we even make the unsuspecting script think we answered "yes" to the confirm box's prompt!

This is really a very powerful concept - although the functions in this example don't really do anything, more complex replacement functions could filter only certain alert boxes or even do something else entirely. There's really no limit.

Since it's so efficient, The Proxomitron's default rules make good use of this trick. One drawback, while it works with most all versions of Netscape, and IE 4.0 or higher, it won't work with Microsoft's JScript in IE 3.x The Proxomitron provides an alternate rule set for IE 3.0 users which use the normal search and replace techniques to accomplish these things instead.

Using recursive matching

Recursive matching is when a rule matches its own previous results. Normally this is something to avoid, especially if it leads to infinite recursion - where the rule matches endlessly against itself. However, used carefully, this can be a powerful technique.

Take this scenario - say you want to eliminate any pop-up windows the occurs between the "<script ..." and "</script>" tags. Since JavaScript uses the "open(...)" command to pop open a window, the resulting rule might look like this...
Matching: <script \1 open \( * \) \2 </script>
Replace: <script \1 \2 </script>

(In actuality, you'd also want to use scope here, but we'll discuss that later. Also note the use of the "\" to escape the special meaning of the parenthesis). This might work if there was only one open command in the JavaScript, but if there was more than that only the first would be eliminated.

The solution? Well, two things. First, check "Allow for multiple matches" to let the rule match against its own results. Secondly, we need to change the replacement text to read
Replace: \n<script \1 \2 </script>
Why? Well, in order to help guard against accidental infinite recursion, The Proxomitron's matching engine always advances one character forward after all rules have been checked. This means the next time through our rule would see only "script ..." instead of "<script ...". To get around this we simply insert a "newline" character in front of the replacement text. Although in the final output this will create a blank line before the "<script ..." tag, the browser will ignore it. However it lets the rule see "<script ..." next time around.

A leading space could have also been used, in fact so could anything that doesn't affect the web browser's function. The idea is simply to push the entire result at least one character forward. Once all the "open(...)" commands have been removed, the rule will no longer match, so there's no danger of infinite recursion.

A word about Scope Bounds

For simplicity, the examples sofar have made no use of the web filter's scope bounds settings. Bounds can be used to control how far ahead The Proxomitron's matching engine scans a web page when searching for a match (see the web page filter editor for a more detailed explanation). With bounds you normally give the rule fixed starting and ending points to search between. Reusing an example from above, take the following rule...

Matching: <script \1 open \( * \) \2 </script>
Replace: <script \1 \2 </script>

Written using bounds, this would be

Bounds: <script\s*</script>
Limit: 4096
Matching: \1 open \( * \) \2
Replace: \1 \2

Notice we just move the fixed beginning and ending text into the scope's bounds check. The byte limit is the maximum characters to search before giving up. Nothing to it really. Also we've used the \1 and \2 in the matching expression to capture the start and end of what the bounds matched - including <script and </script>. So we no longer need put them in the replacement text.

There are more tips on The Proxomitron Filter Explanation Page. If you have any ideas of your own please let me know. Also remember you can look at the rules that come with The Proxomitron! Use them as a starting point when writing new rules - often you'll find one that already comes close to what you want to do.

How do I ...? The Proxomitron FAQ

I've run The Proxomitron, but my browser doesn't seem to use it ...

  1. You need to set your browser's proxy option to use The Proxomitron - see installation.
  2. The first time you use The Proxomitron you may need to reload the page or clear your browser's cache before the changes will appear.
  3. Your system may lack the file that defines localhost: it's in the windows system directory, called hosts (no extension) and contains the line
    127.0.0.1 localhost
  4. If you use a 'free' service browser, it may be designed to prevent you from using any proxy. (That's because the service has to get their money from somewhere - if they don't get it from you they have to get it from ad pushers and data miners.) In some cases, you may be able to make the initial connection to the Internet using the 'free' browser, then continue using an unmodified browser from then on which will accept The Proxomitron.
  5. If you are using a firewall, you will need to tell it The Proxomitron is OK.

My browser doesn't seem to work anymore - what gives?

Once you set your browser to use The Proxomitron, both programs must be running for things to work. See installation. If you don't want to use The Proxomitron, set your browser back to not using a HTTP proxy.

I'm getting JavaScript errors on web pages I view - why?

Are you using Internet Explorer version 3.x? It doesn't support some of the JavaScript tricks The Proxomitron uses to filter things like pop-up windows and alert boxes. Use the special Internet Explorer 3.0 rule set included with The Proxomitron distribution.

A filter can also on occasion cause an error by changing things in a way the original script never expected (like closing a window it thought should be open for instance). The Proxomitron's rules are designed to work transparently when they can, but it's not possible to account for every situation. One solution is to enable the "Suppress all JavaScript errors" filter - the browser will then ignore any errors if they occur. Since it's not unusual for scripts to fail even without The Proxomitron's meddling, it's a good filter to always use.

When using certain filters, some web pages don't seem to work anymore - why?

When filtering a web page there's always a chance something necessary to the page's function might be eliminated - this is especially true of JavaScript based filters. Although it's bad HTML design, some pages might actually need pop-up windows or dynamic HTML in order to work. The quick solution is just to bypass the web filters and hit reload on your browser. Or, ignore dumb sites like that.

For the more adventurous, open the log window. This will show which rules are used on a particular page, and can be useful in tracking down exactly which rule is causing the problem. Once you find it you can exclude the offending page by adding it to the rule's URL match, like (^www.dont.filter.me.com)

Help! Hotmail doesn't work any more!

Like most 'free' services, Hotmail gets the money to run their operation by selling data they collect about you to advertisers, spammers and the like. So, they block service to users they can't make money off. The simplest way to get Hotmail to work is to add "[^/]++hotmail" to The Proxomitron's bypass list.

How do I see the original unfiltered web page?

Click The Proxomitron's bypass button then force a browser reload: In Netscape hold down SHIFT while you click reload. In IE 4.x and Opera, press CTRL+F5. (With IE 3.x, use F5 and hope for the best.)

How come some rules don't appear to work on my browser?

What browser are you using? The Proxomitron can use any browser, but some items (especially JavaScript) vary from browser to browser. The default rule set was designed around Netscape and I.E. 4.x - there's also a special I.E. 3.x rule set. If you're using a different browser however, some rules might need tweaked to work properly - the good part is anyone can change them! If you do find an incompatibility please let me know.

How come some sites slow to a crawl with The Proxomitron active?

When a remote site doesn't respond, browsers are set to try again after a timeout period. Most browsers default to a sensible retry value like 2 or 3, but some 'browser accellerator' packages increase this to a very large number so that all the pages referred to by the page you are viewing can be preloaded into your browser cache. Not surprisingly, some sites block these excess demands to limit the load on them. When connected direct, a browser sees these blocks as connect errors and makes no new connections. However it won't see them through a proxy because the browser-to-proxy connection is good, so the browser keeps telling The Proxomitron to retry the connection. The solution is to find where in your system (usually the registry) the connections value has been set too high and put it back to something sensible - definitely less than 10.

How many filters can I use at the same time?

As many as you want! It's really only limited by your computers memory and speed. However, on slower systems such as 486's, using too many simultaneous filters could slow things down. If this is a problem:

Limitations

The Proxomitron can filter most web pages. However, there are some things it can't filter. These include

Filter Examples