htmlex

A powerful hypertext markup language preprocessor (HTML)

Copyright (C) 2001, 2002, 2003 by David Capello

Introduction

htmlex is a small but powerful HTML preprocesor, which has numerous advantages compared to conventional HTML files. On top of that, htmlex is fully compatible with the standard HTML syntax. This means that you can use any HTML file that you have previously made, rename it to .htex, and start using htmlex.

This software is distributed under the MIT license.

Options

htmlex interprets the following options from the command line:

-c   compiles all subsequent files (gets an output file name
     from '-o' or generates a name with .html extension, see -E)
-o   adds output files
-a   adds arguments for the input files
-I   adds search include paths (you can use old '-i' parameter)
-E   changes the HTML extension generated by -c option (.html by default)
-k   kills comments (old htmlex behavior)
-d   calculates dependencies of the input files (output to STDOUT)
-v   activates the verbose mode (to see what htmlex does)
-V   activates very verbose mode (to debug .htex files)
-h   displays help screen and exit
--   breaks -coaI arguments

Execution

References

• STDIN (standard input): from the keyboard or from file.
• STDOUT (standard output): to the screen or to file.

Common behavior

./htmlex

Process STDIN and leave the results in STDOUT.

./htmlex file arguments...

Process file and leave the results in STDOUT.

./htmlex -c files... -a arguments...

Process files and leave the results in files.html.

./htmlex -c files... -o destinations...
./htmlex -o destinations... -c files...

Process the files and leave the results in the destinations.

Some equivalent examples

./htmlex file.htex arg1 arg2 > file.html
./htmlex -c file.htex -a arg1 arg2
./htmlex -c file.htex -o file.html -a arg1 arg2
./htmlex -a arg1 arg2 < file.htex > file.html
./htmlex -a arg1 arg2 -o file.html < file.htex

Common mistakes

File to compile

./htmlex -I examples file.htex > file.html

Here the program waits for input from STDIN, because file.htex counts as another path for -I. Possible solutions:

./htmlex -I examples -- file.htex > file.html
./htmlex -I examples -c file.htex
./htmlex file.htex -I examples > file.html

Arguments for the input file

./htmlex -c index.htex en English

Here, htmlex will try to compile the files index.htex, en and English, instead of passing the arguments to index.htex. Possible solutions:

./htmlex -c index.htex -a en English
./htmlex index.htex en English > index.html

Changes from previous versions

• Now you can use -I instead of -i (like gcc).

• The argument order isn't necessary anymore, so you can do:

    ./htmlex -c src.htex -o dst.html -I dir
  

Features

Reference

This notation will be used along of the definitions:

• expression: The expression is mandatory.
• [ expression ]: The expression is optional.
• { expression }: The expression is optional and repeatable.
• source -> result: Indicates that results will be obtained when the program will process the source text.

Tags

Lists of tags which you can use in the .htex files (by alphabetical sorting):

args

<!args>

Returns the number of arguments which was passed to the input file.

Example:

You input <!args> argument(s).

argN

<!arg1> ... <!argN>

Returns the argument's value. The arguments are passed to the files via the command line's -a option, or from the <!include file arg1 ... argN> tag.

Example:

2nd argument = <!arg2>.

basename

<!basename file>

Returns the base name of the file. In other words, removes its extension.

Examples:

<!basename src/foo.c>    -> src/foo
<!basename src-1.0/bar>  -> src-1.0/bar
<!basename hacks>        -> hacks

chop

<!chop word>

Removes the last character of word.

Examples:

<!chop hello>       -> hell
<!chop directorio/> -> directorio

clean

<!clean word>

Cleans word removing blank space (tabulators and line breaks) around it.

Example:

*<!clean " hola ">* -> *hola*

dep

<!dep file>

Adds 'file' as a dependency. This tag takes effect only in the dependency generation process.

Example:

<!exec ./script.sh>
<!dep script.sh>

dir

<!dir file>

Returns the directory part of file. In other words, deletes the file name.

Examples:

<!dir src/foo.c> -> src/
<!dir a/b/foo.h> -> a/b/
<!dir hacks>     -> ./

elif

<!elif expression>

See if.

else

<!else>

See if.

end

<!end>

See function.

exec

<!exec command { arguments }>

Executes the indicated command passing the arguments to it. Whatever the command prints to the standard output (STDOUT) will be inserted at this position in the file.

Examples:

<!exec date>
<!exec gcc --version>

exec-proc

<!exec-proc file { arguments }>

Equal to <!exec ...> except that it will process the results as other htmlex file.

Examples:

<!exec-proc cat header.htex>
<!exec-proc cat "a footer.htex">
<!exec-proc cat <!find macros.htex>>

fi

<!fi>

See if.

file-size

<!file-size file>

Returns the size of file in a format that's more human-readable (bytes, KB, MB or GB).

Example:

image <!file-size image.jpg>

find

<!find file>

Tries to find file in the search path (the active directory and all directories added with the -I option), and will return the name of the file with the matching path. In the case that the file isn't found, nothing is returned.

Examples:

<!if <!find index.en.html>>
  english
<!fi>

function

<!function name { submacros }>

Creates a new functional macro like in C/C++. When the program see this tag, it follows reading the file in "raw" mode until finds <!end> tag. Then, the macro could be called like any other tag: <!name {args}>

Example:

  <!function my_macro a b c>a is b c<!end>
  <!my_macro This my macro>.

Output:

  This is my macro.

if

<!if expression>

One of the more powerful features of htmlex relative to conventional HTML is the conditional-blocks.

Through four tags you can do some of the conditional operations of "normal" programming languages. Although nowadays, htmlex has a poor handling of operations, improvements to this construct will made.

The general syntax is the following:

<!if [ expression ]> block
{ <!elif [ expression ]> block }
[ <!else> block ]
<!fi>

The block can be any text (or just nothing), can have other tags, and consequentialy, can have more nested <!if> tags.

The expression to be evaluated can have any of the formats below:

FORMAT               TRUE IF...
-------              ----------
number               number is not zero
! number             number is zero
word                 word has one character as minimum
! word               word is empty
number1 < number2    number1 is less than number2
number1 > number2    number1 is greater than number2
number1 <= number2   number1 is less or equal than number2
number1 >= number2   number1 is greater or equal than number2
number1 == number2   number1 is equal than number2
number1 != number2   number1 is different than number2
word1 == word2       word1 is identical than word2
word1 != word2       word1 differs in just one character than word2

If the expression format cannot be resolved to the table above, then false is returned.

numbers are distinguished from words when they have some of these formats:

[1-9][0-9]*      decimals
0x[0-9a-fA-F]*   hexadecimals
0[0-7]           octals

Examples:

<!if <!arg1> == --help>
  Do you need help?
<!else>
  Try with --help.
<!fi>
<!if <!tolower HELLO> == <!tolower Hello>>
  works
<!fi>
<!if>For here it won't never pass<!fi>

include

<!include file { arguments }>

Includes the file to be processed like an normal .htex file. This file will also be used for calculating dependencies. This is equivalent to C's #include.

Examples:

<!include header.htex>
<!include footer.htex en UK>

macro

<!macro name [ value ]>

Creates or modifies a macro with the specified name and value. If value is not specified, the macro will be eliminated from memory. The main use of macros is to replace a certain keyword (its name) for some predetermined text (its value).

Example:

<!macro hi bye>
hi -> bye   <!-- "hi" replaced by "bye" -->
<!macro hi>
hi -> hi    <!-- "hi" macro is deleted -->

macro-reset

<!macro-reset>

Removes any macro which was created before this tag.

Example:

<!macro a b>a<!macro-reset>a -> ba

notdir

<!notdir file>

Returns the name of the file without the directory.

Examples:

<!notdir src/foo.c> -> foo.c
<!notdir a/b/foo.h> -> foo.h
<!notdir hacks>     -> hacks

shift

<!shift word>

Removes the first character of word.

Examples:

<!shift your> -> our
<!shift .txt> -> txt

suffix

<!suffix file>

Returns the extension of file.

Examples:

<!suffix "src/foo.c">   -> .c
<!suffix "src-1.0/bar"> ->
<!suffix "hacks">       ->

tolower

<!tolower word [ beginning [ final ] ]>

See toupper.

toupper

<!toupper word [ beginning [ final ] ]>

Converts partially or completelly word to lower or upper case depending on if you specify or not the beginning and/or the final character index. Indecies start at 0.

If only beginning is specified, then the range starting from beginning to the last character is converted. If you also indicate final, you are specifying an upper bound to the conversion range.

Note: negative values for final indicates that values are to be taken from the end of the word.

Examples:

<!tolower HELLO>      -> hello
<!tolower HELLO 0>    -> hELLO
<!toupper hello 1 2>  -> hELlo
<!toupper hello 0 -2> -> HELLo

version

<!version>

Shows the program version which is processing the file.

Example:

This page was created with htmlex <!version>.

Important Notes

• <!macro X> deletes the X macro, so if you use <!if X> that expression will give you true (because the string "X" will be used, and not the value of X).

• <!macro X ""> assigns a null value to the X macro, this is necessary for a false result in <!if X> (because in this case, the X is replaced by its value).

Author

David Capello [email protected]

Any suggestion, thank-you letters, or (preferably) donations ;-), will be most welcomed.

You can visit http://davidcapello.com/donate/ for more information.