feat: add an option for a function to ignore files (#223)

* refactor: use vim.pesc instead of implementing

* feat: add optional func to ignore in registering

* test: add tests for ignore_register function

* docs: add note for `ignore_register` option

* docs: add note for use case of `ignore_patterns`

cspell.json

@@ -1 +1 @@
-{"words":["frecency","nvim","frecently","frecent","devicons","Neovim","shada","oldfiles","keymap","smartcase","frecencydelete","frecencyvalidate","frecencyopts","stdpath","substr","unindexed","fdfind","telescopebufferloaded","telescopepathseparator","telescopefrecencyscores","telescopequeryfilter","norl","noet","compl"],"flagWords":[],"language":"en","version":"0.2"}
+{"flagWords":[],"words":["frecency","nvim","frecently","frecent","devicons","Neovim","shada","oldfiles","keymap","smartcase","frecencydelete","frecencyvalidate","frecencyopts","stdpath","substr","unindexed","fdfind","telescopebufferloaded","telescopepathseparator","telescopefrecencyscores","telescopequeryfilter","norl","noet","compl","bufnr","nobuflisted","buflisted"],"language":"en","version":"0.2"}

doc/telescope-frecency.txt

@@ -443,9 +443,31 @@ Default:
   for Windows: `{ [[*.git\*]], [[*\tmp\*]], "term://*" }`
 Type: `string`
 
-Patterns in this table control which files are indexed (and subsequently which
+Glob patterns in this table control which files are indexed (and subsequently
-you'll see in the finder results).
+which you'll see in the finder results).
 
+			      *telescope-frecency-configuration-ignore_register*
+ignore_register ~
+
+Default: `nil`
+Type: `fun(bufnr: integer): boolean`
+
+In opening buffers, it registers the filename and timestamps into DB to
+calculate frecency scores later. `ignore_register` can take a buffer number
+and you can control whether the file should be registered or not.
+
+For example, you can ignore |'nobuflisted'| buffers to register with this below.
+>lua
+    telescope.setup {
+      extensions = {
+	frecency = {
+	  ignore_register = function(bufnr)
+	    return not vim.bo[bufnr].buflisted
+	  end,
+	}
+      },
+    }
+<
 				      *telescope-frecency-configuration-matcher*
 matcher ~
 
@@ -592,6 +614,11 @@ workspace files. It tries in order until it works successfully.
 If you like another commands, set them to this option, like
 `workspace_scan_cmd = { "find", ".", "-type", "f" }`.
 
+NOTE: When the plugin uses “4. Native Lua code”, it uses
+|telescope-frecency-configuration-ignore_patterns| to ignore candidates from
+listed entries. With other ways, such as `rg` and `fd`, it does not use the option
+because of performance consideration.
+
 				   *telescope-frecency-configuration-workspaces*
 workspaces ~
 

lua/frecency/config.lua

@@ -13,6 +13,7 @@ local os_util = require "frecency.os_util"
 ---@field filter_delimiter? string default: ":"
 ---@field hide_current_buffer? boolean default: false
 ---@field ignore_patterns? string[] default: { "*.git/*", "*/tmp/*", "term://*" }
+---@field ignore_register? fun(bufnr: integer): boolean
 ---@field matcher? "default"|"fuzzy" default: "default"
 ---@field scoring_function? fun(recency: integer, fzy_score: number): number default: see lua/frecency/config.lua
 ---@field max_timestamps? integer default: 10
@@ -40,6 +41,7 @@ local Config = {}
 ---@field filter_delimiter string default: ":"
 ---@field hide_current_buffer boolean default: false
 ---@field ignore_patterns string[] default: { "*.git/*", "*/tmp/*", "term://*" }
+---@field ignore_register? fun(bufnr: integer): boolean default: nil
 ---@field matcher "default"|"fuzzy" default: "default"
 ---@field scoring_function fun(recency: integer, fzy_score: number): number default: see lua/frecency/config.lua
 ---@field max_timestamps integer default: 10
@@ -65,6 +67,7 @@ Config.new = function()
     filter_delimiter = true,
     hide_current_buffer = true,
     ignore_patterns = true,
+    ignore_register = true,
     matcher = true,
     max_timestamps = true,
     path_display = true,

lua/frecency/fs.lua

@@ -17,14 +17,13 @@ local FS = {}
 ---@param fs_config? FrecencyFSConfig
 ---@return FrecencyFS
 FS.new = function(fs_config)
-  local self = setmetatable(
+  local self= setmetatable(
     { config = vim.tbl_extend("force", { scan_depth = 100 }, fs_config or {}), os_homedir = assert(uv.os_homedir()) },
     { __index = FS }
   )
   ---@param pattern string
   self.ignore_regexes = vim.tbl_map(function(pattern)
-    local escaped = pattern:gsub("[%-%.%+%[%]%(%)%$%^%%%?%*]", "%%%1")
+    local regex = vim.pesc(pattern):gsub("%%%*", ".*"):gsub("%%%?", ".")
-    local regex = escaped:gsub("%%%*", ".*"):gsub("%%%?", ".")
     return "^" .. regex .. "$"
   end, config.ignore_patterns)
   return self

lua/frecency/init.lua

@@ -113,6 +113,9 @@ end
 ---@param bufnr integer
 ---@param epoch? integer
 function Frecency:register(bufnr, epoch)
+  if config.ignore_register and config.ignore_register(bufnr) then
+    return
+  end
   local path = vim.api.nvim_buf_get_name(bufnr)
   if self.buf_registered[bufnr] or not self.fs:is_valid_path(path) then
     return

lua/frecency/tests/frecency_spec.lua

@@ -232,6 +232,29 @@ describe("frecency", function()
         end)
       end)
     end)
+
+    describe("when ignore_register is set", function()
+      with_files({ "hoge1.txt", "hoge2.txt" }, {
+        ignore_register = function(bufnr)
+          local _, bufname = pcall(vim.api.nvim_buf_get_name, bufnr)
+          local should_ignore = not not (bufname and bufname:find "hoge2%.txt$")
+          log.debug { bufnr = bufnr, bufname = bufname, should_ignore = should_ignore }
+          return should_ignore
+        end,
+      }, function(frecency, finder, dir)
+        local register = make_register(frecency, dir)
+        local epoch1 = make_epoch "2023-07-29T00:00:00+09:00"
+        register("hoge1.txt", epoch1)
+        local epoch2 = make_epoch "2023-07-29T01:00:00+09:00"
+        register("hoge2.txt", epoch2)
+        it("ignores the file the func returns true", function()
+          local results = finder:get_results(nil, make_epoch "2023-07-29T02:00:00+09:00")
+          assert.are.same({
+            { count = 1, path = filepath(dir, "hoge1.txt"), score = 10, timestamps = { epoch1 } },
+          }, results)
+        end)
+      end)
+    end)
   end)
 
   describe("benchmark", function()