Factor is a concatenative, stack-based programming language with high-level features including dynamic types, extensible syntax, macros, and garbage collection. On a practical side, Factor has a full-featured library, supports many different platforms, and has been extensively documented.

The implementation is fully compiled for performance, while still supporting interactive development. Factor applications are portable between all common platforms. Factor can deploy stand-alone applications on all platforms. Full source code for the Factor project is available under a BSD license.

There are several differences between this fork and the upstream Factor implementation:

• main work is done in the mine branch;
• the commits starting with !mine!, !todo! or !wip! are not meant for upstream, not finished or are specific to my development environment;
• the "unmaintained" folder is undeleted, because I like to browse it sometimes, and even bring some things to life (this reverts commit 9aacb296);
• there is an alternative implementation of the fix for issue 1711, which suppresses the bad-vocab-name exception;
• the ui.gadgets.tables:line-gadget tuple contains the new slot fixed-column-widths, which allows one to pre-set the column widths of a table and avoid their recalculations;
• the custom ui.tools.inspector uses the fixed-column-widths to improve its performance when displaying large tables, while upstream developers decided to implement caching of the size calculations in this component instead of creating a mechanism available to all tables generically;
• another customization of the ui.tools.inspector is related to the inspection of strings: my version of the Inspector always displays two representations of each character - its printable version, if any, and its character code that could be used in the Factor source. The upstream developers decided they don't need to show character codes for the printable characters, and for the unprintable lower part of the ASCII table they only show the escape code, but not the hexadecimal code. I think that my version of the Inspector is better, because it allows one to see the code differences for the identically looking characters, and it produces a more consistent view of the lower ASCII table. See for example the string "\0\x01\tТорGеаr\n", which has some Cyrillic characters embedded in it, viewed in both Inspector versions:

My Inspector            Upstream Inspector
0   \0  \x00            0   \0
1       \x01            1   \x01
2   \t  \x09            2   \t
3   Т   \u{422}         3   Т
4   о   \u{43e}         4   о
5   р   \u{440}         5   р
6   G   \u{47}          6   G
7   е   \u{435}         7   е
8   а   \u{430}         8   а
9   r   \u{72}          9   r
10  \n  \x0a            10  \n

• the parameters for the limit-stream word of the io.streams.limited are swapped compared to the original Factor implementation. I think my version makes the usage simpler, requiring less stack shuffling for typical usage, but the upstream developers decided they don't want to break compatibility by introducing this change;
• a patch for strings.tables suggested by @mrjbq7 in issue 1840 is applied, so now the table cells containing numeric data are right-aligned;
• other changes may be developed and contributed to upstream when ready, but the ones listed above represent either things not meant for general distribution or things already rejected.

If you have a build environment set up, then you can build Factor from git. These scripts will attempt to compile the Factor binary and bootstrap from a boot image stored on factorcode.org.

To check out Factor:

• git clone git://github.com/factor/factor.git
• cd factor

To build the latest complete Factor system from git, either use the build script:

• Unix: ./build.sh update
• Windows: build.cmd

or download the correct boot image for your system from http://downloads.factorcode.org/images/master/, put it in the factor directory and run:

• Unix: make and then ./factor -i=boot.unix-x86.64.image
• Windows: nmake /f Nmakefile x86-64 and then factor.com -i=boot.windows-x86.64.image

Now you should have a complete Factor system ready to run.

More information on building factor and system requirements.

You can download a Factor binary from the grid on https://factorcode.org. The nightly builds are usually a better experience than the point releases.

• Windows: Double-click factor.exe, or run .\factor.com in a command prompt
• Mac OS X: Double-click Factor.app or run open Factor.app in a Terminal
• Unix: Run ./factor in a shell

A tutorial is available that can be accessed from the Factor environment:

"first-program" help

Some other simple things you can try in the listener:

"Hello, world" print

{ 4 8 15 16 23 42 } [ 2 * ] map .

1000 [1,b] sum .

4 <iota> [
    "Happy Birthday " write
    2 = "dear NAME" "to You" ? print
] each

For more tips, see Learning Factor.

The Factor environment includes extensive reference documentation and a short "cookbook" to help you get started. The best way to read the documentation is in the UI; press F1 in the UI listener to open the help browser tool. You can also browse the documentation online.

Factor supports a number of command line switches:

Usage: factor [Factor arguments] [script] [script arguments]

Common arguments:
    -help            print this message and exit
    -i=<image>       load Factor image file <image> (default factor.image)
    -run=<vocab>     run the MAIN: entry point of <vocab>
        -run=listener    run terminal listener
        -run=ui.tools    run Factor development UI
    -e=<code>        evaluate <code>
    -no-user-init    suppress loading of .factor-rc
    -roots=<paths>   a list of path-delimited extra vocab roots

Enter
    "command-line" help
from within Factor for more information.

You can also write scripts that can be run from the terminal, by putting #!/path/to/factor at the top of your scripts and making them executable.

The Factor source tree is organized as follows:

• vm/ - Factor VM source code (not present in binary packages)
• core/ - Factor core library
• basis/ - Factor basis library, compiler, tools
• extra/ - more libraries and applications
• misc/ - editor modes, icons, etc
• unmaintained/ - unmaintained contributions, please help!

During Factor's lifetime, sourcecode has lived in many repositories. Unfortunately, the first import in Git did not keep history. History has been partially recreated from what could be salvaged. Due to the nature of Git, it's only possible to add history without disturbing upstream work, by using replace objects. These need to be manually fetched, or need to be explicitly added to your git remote configuration.

Use: git fetch origin 'refs/replace/*:refs/replace/*'

or add the following line to your configuration file

[remote "origin"]
    url = ...
    fetch = +refs/heads/*:refs/remotes/origin/*
    ...
    fetch = +refs/replace/*:refs/replace/*

Then subsequent fetches will automatically update any replace objects.

Factor developers meet in the #concatenative channel on irc.freenode.net. Drop by if you want to discuss anything related to Factor or language design in general.

• Factor homepage
• Concatenative languages wiki
• Mailing list
• Search for "factorcode" on Gitter

Have fun!