Files
org-vim/src/README.md
Girish Palya 7e0df5eee9 patch 9.1.1627: fuzzy matching can be improved
Problem:  fuzzy-matching can be improved
Solution: Implement a better fuzzy matching algorithm
          (Girish Palya)

Replace fuzzy matching algorithm with improved fzy-based implementation

The
[current](https://www.forrestthewoods.com/blog/reverse_engineering_sublime_texts_fuzzy_match/)
fuzzy matching algorithm has several accuracy issues:

* It struggles with CamelCase
* It fails to prioritize matches at the beginning of strings, often
  ranking middle matches higher.

After evaluating alternatives (see my comments
[here](https://github.com/vim/vim/issues/17531#issuecomment-3112046897)
and
[here](https://github.com/vim/vim/issues/17531#issuecomment-3121593900)),
I chose to adopt the [fzy](https://github.com/jhawthorn/fzy) algorithm,
which:

* Resolves the aforementioned issues.
* Performs better.

Implementation details

This version is based on the original fzy
[algorithm](https://github.com/jhawthorn/fzy/blob/master/src/match.c),
with one key enhancement: **multibyte character support**.

* The original implementation supports only ASCII.
* This patch replaces ascii lookup tables with function calls, making it
  compatible with multibyte character sets.
* Core logic (`match_row()` and `match_positions()`) remains faithful to
  the original, but now operates on codepoints rather than single-byte
  characters.

Performance

Tested against a dataset of **90,000 Linux kernel filenames**. Results
(in milliseconds) show a **\~2x performance improvement** over the
current fuzzy matching algorithm.

```
Search String            Current Algo    FZY Algo
-------------------------------------------------
init                          131.759    66.916
main                          83.688     40.861
sig                           98.348     39.699
index                         109.222    30.738
ab                            72.222     44.357
cd                            83.036     54.739
a                             58.94      62.242
b                             43.612     43.442
c                             64.39      67.442
k                             40.585     36.371
z                             34.708     22.781
w                             38.033     30.109
cpa                           82.596     38.116
arz                           84.251     23.964
zzzz                          35.823     22.75
dimag                         110.686    29.646
xa                            43.188     29.199
nha                           73.953     31.001
nedax                         94.775     29.568
dbue                          79.846     25.902
fp                            46.826     31.641
tr                            90.951     55.883
kw                            38.875     23.194
rp                            101.575    55.775
kkkkkkkkkkkkkkkkkkkkkkkkkkkkk 48.519     30.921
```

```vim
vim9script

var haystack = readfile('/Users/gp/linux.files')

var needles = ['init', 'main', 'sig', 'index', 'ab', 'cd', 'a', 'b',
'c', 'k',
    'z', 'w', 'cpa', 'arz', 'zzzz', 'dimag', 'xa', 'nha', 'nedax',
'dbue',
    'fp', 'tr', 'kw', 'rp', 'kkkkkkkkkkkkkkkkkkkkkkkkkkkkk']
for needle in needles
    var start = reltime()
    var tmp = matchfuzzy(haystack, needle)
    echom $'{needle}' (start->reltime()->reltimefloat() * 1000)
endfor
```

Additional changes

* Removed the "camelcase" option from both matchfuzzy() and
  matchfuzzypos(), as it's now obsolete with the improved algorithm.

related: neovim/neovim#34101
fixes #17531
closes: #17900

Signed-off-by: Girish Palya <girishji@gmail.com>
Signed-off-by: Christian Brabandt <cb@256bit.org>
2025-08-12 22:22:52 +02:00

8.2 KiB

Vim Logo

Vim source code

Here are a few hints for finding your way around the source code. This doesn't make it less complex than it is, but it gets you started.

You might also want to read :help development.

Jumping around

First of all, use :make tags to generate a tags file, so that you can jump around in the source code.

To jump to a function or variable definition, move the cursor on the name and use the CTRL-] command. Use CTRL-T or CTRL-O to jump back.

To jump to a file, move the cursor on its name and use the gf command.

Most code can be found in a file with an obvious name (incomplete list):

File name Description
alloc.c memory management
arglist.c handling argument list
autocmd.c autocommands
blob.c blob data type
buffer.c manipulating buffers (loaded files)
bufwrite.c writing a buffer to file
change.c handling changes to text
cindent.c C and Lisp indentation
clientserver.c client server functionality
clipboard.c handling the clipboard
cmdexpand.c command-line completion
cmdhist.c command-line history
debugger.c Vim script debugger
diff.c diff mode (vimdiff)
drawline.c drawing a window line
drawscreen.c drawing the windows
eval.c expression evaluation
evalbuffer.c buffer related built-in functions
evalfunc.c built-in functions
evalvars.c vim variables
evalwindow.c window related built-in functions
fileio.c reading and writing files
filepath.c dealing with file names and paths
findfile.c search for files in 'path'
fold.c folding
fuzzy.c fuzzy matching
getchar.c getting characters and key mapping
gc.c garbage collection
help.c vim help related functions
highlight.c syntax highlighting
indent.c text indentation
insexpand.c Insert mode completion
locale.c locale/language handling
map.c mapping and abbreviations
mark.c marks
match.c highlight matching
float.c floating point functions
mbyte.c multi-byte character handling
memfile.c storing lines for buffers in a swapfile
memline.c storing lines for buffers in memory
menu.c menus
message.c (error) messages
mouse.c handling the mouse
ops.c handling operators ("d", "y", "p")
option.c options
optionstr.c handling string options
popupmenu.c popup menu
popupwin.c popup window
profiler.c Vim script profiler
quickfix.c quickfix commands (":make", ":cn")
regexp.c pattern matching
register.c handling registers
scriptfile.c runtime directory handling and sourcing scripts
screen.c lower level screen functions
search.c pattern searching
session.c sessions and views
sign.c signs
spell.c spell checking core
spellfile.c spell file handling
spellsuggest.c spell correction suggestions
strings.c string manipulation functions
syntax.c syntax and other highlighting
tag.c tags
term.c terminal handling, termcap codes
testing.c testing: assert and test functions
textformat.c text formatting
textobject.c text objects
textprop.c text properties
time.c time and timer functions
typval.c Vim script type/value functions
undo.c undo and redo
usercmd.c user defined commands
userfunc.c user defined functions
viminfo.c viminfo handling
window.c handling split windows

Debugging

If you have a reasonable recent version of gdb, you can use the :Termdebug command to debug Vim. See :help :Termdebug.

When something is time critical or stepping through code is a hassle, use the channel logging to create a time-stamped log file. Add lines to the code like this:

ch_log(NULL, "Value is now %02x", value);

After compiling and starting Vim, do:

:call ch_logfile('debuglog', 'w')

And edit debuglog to see what happens. The channel functions already have ch_log() calls, thus you always see that in the log.

Important Variables

The current mode is stored in State. The values it can have are NORMAL, INSERT, CMDLINE, and a few others.

The current window is curwin. The current buffer is curbuf. These point to structures with the cursor position in the window, option values, the file name, etc. These are defined in structs.h.

All the global variables are declared in globals.h.

The main loop

This is conveniently called main_loop(). It updates a few things and then calls normal_cmd() to process a command. This returns when the command is finished.

The basic idea is that Vim waits for the user to type a character and processes it until another character is needed. Thus there are several places where Vim waits for a character to be typed. The vgetc() function is used for this. It also handles mapping.

Updating the screen is mostly postponed until a command or a sequence of commands has finished. The work is done by update_screen(), which calls win_update() for every window, which calls win_line() for every line. See the start of screen.c for more explanations.

Command-line mode

When typing a :, normal_cmd() will call getcmdline() to obtain a line with an Ex command. getcmdline() contains a loop that will handle each typed character. It returns when hitting CR or Esc or some other character that ends the command line mode.

Ex commands

Ex commands are handled by the function do_cmdline(). It does the generic parsing of the : command line and calls do_one_cmd() for each separate command. It also takes care of while loops.

do_one_cmd() parses the range and generic arguments and puts them in the exarg_t and passes it to the function that handles the command.

The : commands are listed in ex_cmds.h. The third entry of each item is the name of the function that handles the command. The last entry are the flags that are used for the command.

Normal mode commands

The Normal mode commands are handled by the normal_cmd() function. It also handles the optional count and an extra character for some commands. These are passed in a cmdarg_t to the function that handles the command.

There is a table nv_cmds in normal.c which lists the first character of every command. The second entry of each item is the name of the function that handles the command.

Insert mode commands

When doing an i or a command, normal_cmd() will call the edit() function. It contains a loop that waits for the next character and handles it. It returns when leaving Insert mode.

Options

There is a list with all option names in option.c, called options[].

The GUI

Most of the GUI code is implemented like it was a clever terminal. Typing a character, moving a scrollbar, clicking the mouse, etc. are all translated into events which are written in the input buffer. These are read by the main code, just like reading from a terminal. The code for this is scattered through gui.c. For example, gui_send_mouse_event() for a mouse click and gui_menu_cb() for a menu action. Key hits are handled by the system-specific GUI code, which calls add_to_input_buf() to send the key code.

Updating the GUI window is done by writing codes in the output buffer, just like writing to a terminal. When the buffer gets full or is flushed, gui_write() will parse the codes and draw the appropriate items. Finally the system-specific GUI code will be called to do the work.

Debugging the GUI

Remember to prevent that gvim forks and the debugger thinks Vim has exited, add the -f argument. In gdb: run -f -g.

When stepping through display updating code, the focus event is triggered when going from the debugger to Vim and back. To avoid this, recompile with some code in gui_focus_change() disabled.

Contributing

If you would like to help making Vim better, see the CONTRIBUTING.md file.

This is README.md for version 9.1 of the Vim source code.