From 61f21e383429ad26a9388ead39268cbac2d81163 Mon Sep 17 00:00:00 2001
From: Mika Vilpas <mika.vilpas@gmail.com>
Date: Sun, 19 Jan 2025 19:58:55 +0200
Subject: [PATCH 1/2] docs: add some code comments

---
 lua/blink-ripgrep/highlighting.lua  | 3 +++
 lua/blink-ripgrep/visualization.lua | 2 ++
 2 files changed, 5 insertions(+)

diff --git a/lua/blink-ripgrep/highlighting.lua b/lua/blink-ripgrep/highlighting.lua
index f9603fc..4cd2adc 100644
--- a/lua/blink-ripgrep/highlighting.lua
+++ b/lua/blink-ripgrep/highlighting.lua
@@ -1,5 +1,8 @@
 local M = {}
 
+--- In the blink documentation window, when the context for the match is being
+--- shown, highlight the match so that the user can easily see where the match
+--- is.
 ---@param bufnr number
 ---@param match blink-ripgrep.RipgrepMatch
 ---@param highlight_ns_id number
diff --git a/lua/blink-ripgrep/visualization.lua b/lua/blink-ripgrep/visualization.lua
index cfaa416..a5b9888 100644
--- a/lua/blink-ripgrep/visualization.lua
+++ b/lua/blink-ripgrep/visualization.lua
@@ -8,6 +8,8 @@ vim.api.nvim_set_hl(
   { link = "Search", default = true }
 )
 
+-- Temporarily flash the search prefix so that the user can see what searches
+-- are being performed. Should be called in debug mode only.
 ---@param prefix string
 function visualization.flash_search_prefix(prefix)
   vim.api.nvim_buf_clear_namespace(0, ns_id, 0, -1)

From aaeb8b95cbf50cc4522b424d0f1589d77fb68857 Mon Sep 17 00:00:00 2001
From: Mika Vilpas <mika.vilpas@gmail.com>
Date: Sun, 19 Jan 2025 20:06:23 +0200
Subject: [PATCH 2/2] refactor: extract search prefix logic to a separate
 module

---
 lua/blink-ripgrep/init.lua             |  42 +------
 lua/blink-ripgrep/search_prefix.lua    |  45 ++++++++
 spec/blink-ripgrep/get_prefix_spec.lua | 146 ++++++++++++-------------
 3 files changed, 120 insertions(+), 113 deletions(-)
 create mode 100644 lua/blink-ripgrep/search_prefix.lua

diff --git a/lua/blink-ripgrep/init.lua b/lua/blink-ripgrep/init.lua
index 1c13f77..7ccbb2e 100644
--- a/lua/blink-ripgrep/init.lua
+++ b/lua/blink-ripgrep/init.lua
@@ -27,27 +27,6 @@ pcall(function()
 end)
 vim.api.nvim_set_hl(0, "BlinkRipgrepMatch", { link = "Search", default = true })
 
-local word_pattern
-do
-  -- match an ascii character as well as unicode continuation bytes.
-  -- Technically, unicode continuation bytes need to be applied in order to
-  -- construct valid utf-8 characters, but right now we trust that the user
-  -- only types valid utf-8 in their project.
-  local char = vim.lpeg.R("az", "AZ", "09", "\128\255")
-
-  local non_starting_word_character = vim.lpeg.P(1) - char
-  local word_character = char + vim.lpeg.P("_") + vim.lpeg.P("-")
-  local non_middle_word_character = vim.lpeg.P(1) - word_character
-
-  word_pattern = vim.lpeg.Ct(
-    (
-      non_starting_word_character ^ 0
-      * vim.lpeg.C(word_character ^ 1)
-      * non_middle_word_character ^ 0
-    ) ^ 0
-  )
-end
-
 ---@type blink-ripgrep.Options
 RgSource.config = {
   prefix_min_len = 3,
@@ -67,24 +46,6 @@ function RgSource.setup(options)
   RgSource.config = vim.tbl_deep_extend("force", RgSource.config, options or {})
 end
 
----@param text_before_cursor string "The text of the entire line before the cursor"
----@return string
-function RgSource.match_prefix(text_before_cursor)
-  local matches = vim.lpeg.match(word_pattern, text_before_cursor)
-  local last_match = matches and matches[#matches]
-  return last_match or ""
-end
-
----@param context blink.cmp.Context
----@return string
-local function default_get_prefix(context)
-  local line = context.line
-  local col = context.cursor[2]
-  local text = line:sub(1, col)
-  local prefix = RgSource.match_prefix(text)
-  return prefix
-end
-
 ---@param input_opts blink-ripgrep.Options
 function RgSource.new(input_opts)
   local self = setmetatable({}, RgSource)
@@ -92,7 +53,8 @@ function RgSource.new(input_opts)
   RgSource.config =
     vim.tbl_deep_extend("force", RgSource.config, input_opts or {})
 
-  self.get_prefix = RgSource.config.get_prefix or default_get_prefix
+  self.get_prefix = RgSource.config.get_prefix
+    or require("blink-ripgrep.search_prefix").default_get_prefix
 
   self.get_command = RgSource.config.get_command
 
diff --git a/lua/blink-ripgrep/search_prefix.lua b/lua/blink-ripgrep/search_prefix.lua
new file mode 100644
index 0000000..f96075d
--- /dev/null
+++ b/lua/blink-ripgrep/search_prefix.lua
@@ -0,0 +1,45 @@
+local M = {}
+
+--- The pattern that is used to match the prefix of a search query. The prefix
+--- is the text before the cursor that is used to search for matching words in
+--- the project.
+local word_pattern
+do
+  -- match an ascii character as well as unicode continuation bytes.
+  -- Technically, unicode continuation bytes need to be applied in order to
+  -- construct valid utf-8 characters, but right now we trust that the user
+  -- only types valid utf-8 in their project.
+  local char = vim.lpeg.R("az", "AZ", "09", "\128\255")
+
+  local non_starting_word_character = vim.lpeg.P(1) - char
+  local word_character = char + vim.lpeg.P("_") + vim.lpeg.P("-")
+  local non_middle_word_character = vim.lpeg.P(1) - word_character
+
+  word_pattern = vim.lpeg.Ct(
+    (
+      non_starting_word_character ^ 0
+      * vim.lpeg.C(word_character ^ 1)
+      * non_middle_word_character ^ 0
+    ) ^ 0
+  )
+end
+
+---@param text_before_cursor string "The text of the entire line before the cursor"
+---@return string
+function M.match_prefix(text_before_cursor)
+  local matches = vim.lpeg.match(word_pattern, text_before_cursor)
+  local last_match = matches and matches[#matches]
+  return last_match or ""
+end
+
+---@param context blink.cmp.Context
+---@return string
+function M.default_get_prefix(context)
+  local line = context.line
+  local col = context.cursor[2]
+  local text = line:sub(1, col)
+  local prefix = M.match_prefix(text)
+  return prefix
+end
+
+return M
diff --git a/spec/blink-ripgrep/get_prefix_spec.lua b/spec/blink-ripgrep/get_prefix_spec.lua
index 888168d..ae0094d 100644
--- a/spec/blink-ripgrep/get_prefix_spec.lua
+++ b/spec/blink-ripgrep/get_prefix_spec.lua
@@ -1,82 +1,82 @@
 local assert = require("luassert")
-local blink_ripgrep = require("blink-ripgrep")
+local search_prefix = require("blink-ripgrep.search_prefix")
 
 describe("match_prefix", function()
   it("for simple strings", function()
-    assert.are_same(blink_ripgrep.match_prefix("hello"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("abc123"), "abc123")
-    assert.are_same(blink_ripgrep.match_prefix("abc-123"), "abc-123")
-    assert.are_same(blink_ripgrep.match_prefix("abc_123"), "abc_123")
+    assert.are_same(search_prefix.match_prefix("hello"), "hello")
+    assert.are_same(search_prefix.match_prefix("abc123"), "abc123")
+    assert.are_same(search_prefix.match_prefix("abc-123"), "abc-123")
+    assert.are_same(search_prefix.match_prefix("abc_123"), "abc_123")
   end)
 
   it(
     "matches when there is one nonmatching piece of input in the beginning",
     function()
-      assert.are_same(blink_ripgrep.match_prefix(".hello"), "hello")
-      assert.are_same(blink_ripgrep.match_prefix(",hello"), "hello")
+      assert.are_same(search_prefix.match_prefix(".hello"), "hello")
+      assert.are_same(search_prefix.match_prefix(",hello"), "hello")
       assert.are_same(
-        blink_ripgrep.match_prefix(".,,!@!@$@%<<@$<hello"),
+        search_prefix.match_prefix(".,,!@!@$@%<<@$<hello"),
         "hello"
       )
-      assert.are_same(blink_ripgrep.match_prefix(" hello"), "hello")
-      assert.are_same(blink_ripgrep.match_prefix("random_text hello"), "hello")
-      assert.are_same(blink_ripgrep.match_prefix("-- hello"), "hello")
-      assert.are_same(blink_ripgrep.match_prefix("-- abc123"), "abc123")
-      assert.are_same(blink_ripgrep.match_prefix("-- abc-123"), "abc-123")
-      assert.are_same(blink_ripgrep.match_prefix("-- abc_123"), "abc_123")
+      assert.are_same(search_prefix.match_prefix(" hello"), "hello")
+      assert.are_same(search_prefix.match_prefix("random_text hello"), "hello")
+      assert.are_same(search_prefix.match_prefix("-- hello"), "hello")
+      assert.are_same(search_prefix.match_prefix("-- abc123"), "abc123")
+      assert.are_same(search_prefix.match_prefix("-- abc-123"), "abc-123")
+      assert.are_same(search_prefix.match_prefix("-- abc_123"), "abc_123")
     end
   )
 
   it("matches when there is nonmatching input at the end", function()
-    assert.are_same(blink_ripgrep.match_prefix(".hello)"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix(",hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix(".hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix(",hello)"), "hello")
     assert.are_same(
-      blink_ripgrep.match_prefix(".,,!@!@$@%<<@$<hello)"),
+      search_prefix.match_prefix(".,,!@!@$@%<<@$<hello)"),
       "hello"
     )
-    assert.are_same(blink_ripgrep.match_prefix(" hello)"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("random_text hello)"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("-- hello)"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("-- abc123)"), "abc123")
-    assert.are_same(blink_ripgrep.match_prefix("-- abc-123)"), "abc-123")
-    assert.are_same(blink_ripgrep.match_prefix("-- abc_123)"), "abc_123")
+    assert.are_same(search_prefix.match_prefix(" hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix("random_text hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix("-- hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix("-- abc123)"), "abc123")
+    assert.are_same(search_prefix.match_prefix("-- abc-123)"), "abc-123")
+    assert.are_same(search_prefix.match_prefix("-- abc_123)"), "abc_123")
   end)
 
   it("matches when word_characters are in the front", function()
-    assert.are_same(blink_ripgrep.match_prefix("--hello)"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("__hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix("--hello)"), "hello")
+    assert.are_same(search_prefix.match_prefix("__hello)"), "hello")
   end)
 
   it(
     "matches when there is nonmatching input at the beginning and at the end",
     function()
-      assert.are_same(blink_ripgrep.match_prefix('"hello"'), "hello")
+      assert.are_same(search_prefix.match_prefix('"hello"'), "hello")
     end
   )
 
   it(
     "matches when there are multiple nonmatching pieces of input in the beginning",
     function()
-      assert.are_same(blink_ripgrep.match_prefix(".hello.hello"), "hello")
-      assert.are_same(blink_ripgrep.match_prefix(",hello,hello"), "hello")
+      assert.are_same(search_prefix.match_prefix(".hello.hello"), "hello")
+      assert.are_same(search_prefix.match_prefix(",hello,hello"), "hello")
       assert.are_same(
-        blink_ripgrep.match_prefix(".,,!@!@$@%<<@$<hello.,,!@!@$@%<<@$<hello"),
+        search_prefix.match_prefix(".,,!@!@$@%<<@$<hello.,,!@!@$@%<<@$<hello"),
         "hello"
       )
-      assert.are_same(blink_ripgrep.match_prefix(" hello hello"), "hello")
+      assert.are_same(search_prefix.match_prefix(" hello hello"), "hello")
 
       assert.are_same(
-        blink_ripgrep.match_prefix("random_text hello hello"),
+        search_prefix.match_prefix("random_text hello hello"),
         "hello"
       )
-      assert.are_same(blink_ripgrep.match_prefix("-- hello hello"), "hello")
-      assert.are_same(blink_ripgrep.match_prefix("-- abc123 abc123"), "abc123")
+      assert.are_same(search_prefix.match_prefix("-- hello hello"), "hello")
+      assert.are_same(search_prefix.match_prefix("-- abc123 abc123"), "abc123")
       assert.are_same(
-        blink_ripgrep.match_prefix("-- abc-123 abc-123"),
+        search_prefix.match_prefix("-- abc-123 abc-123"),
         "abc-123"
       )
       assert.are_same(
-        blink_ripgrep.match_prefix("-- abc_123 abc_123"),
+        search_prefix.match_prefix("-- abc_123 abc_123"),
         "abc_123"
       )
     end
@@ -85,72 +85,72 @@ describe("match_prefix", function()
   it("for multipart strings", function()
     -- three parts
     assert.are_same(
-      blink_ripgrep.match_prefix("hello-world-today"),
+      search_prefix.match_prefix("hello-world-today"),
       "hello-world-today"
     )
 
     -- three parts with numbers
     assert.are_same(
-      blink_ripgrep.match_prefix("abc123-def456-ghi789"),
+      search_prefix.match_prefix("abc123-def456-ghi789"),
       "abc123-def456-ghi789"
     )
 
     -- multiple parts with mixed dashes and underscores
     assert.are_same(
-      blink_ripgrep.match_prefix("abc-123_def-456_ghi-789"),
+      search_prefix.match_prefix("abc-123_def-456_ghi-789"),
       "abc-123_def-456_ghi-789"
     )
   end)
 
   it("matches unicode characters", function()
     -- umlauts and other special characters
-    assert.are_same(blink_ripgrep.match_prefix("yöllä"), "yöllä") -- Finnish word with 'ö' and 'ä'
-    assert.are_same(blink_ripgrep.match_prefix("über"), "über") -- German word with 'ü'
-    assert.are_same(blink_ripgrep.match_prefix("übermensch"), "übermensch") -- German compound word with 'ü'
-    assert.are_same(blink_ripgrep.match_prefix("mañana"), "mañana") -- Spanish word with 'ñ'
-    assert.are_same(blink_ripgrep.match_prefix("Ångström"), "Ångström") -- Swedish word with 'Å' and 'ö'
-    assert.are_same(blink_ripgrep.match_prefix("Straße"), "Straße") -- German word with 'ß'
-    assert.are_same(blink_ripgrep.match_prefix("český"), "český") -- Czech word with 'č'
-    assert.are_same(blink_ripgrep.match_prefix("naïve"), "naïve") -- French word with 'ï'
-    assert.are_same(blink_ripgrep.match_prefix("façade"), "façade") -- French word with 'ç'
-    assert.are_same(blink_ripgrep.match_prefix("résumé"), "résumé") -- French word with 'é'
-    assert.are_same(blink_ripgrep.match_prefix("космос"), "космос") -- Russian word with Cyrillic characters
-    assert.are_same(blink_ripgrep.match_prefix("你好"), "你好") -- Chinese characters
-    assert.are_same(blink_ripgrep.match_prefix("日本語"), "日本語") -- Japanese characters
-    assert.are_same(blink_ripgrep.match_prefix("한국어"), "한국어") -- Korean characters
-    assert.are_same(blink_ripgrep.match_prefix("τοπική"), "τοπική") -- Greek word with 'π' and 'ή'
+    assert.are_same(search_prefix.match_prefix("yöllä"), "yöllä") -- Finnish word with 'ö' and 'ä'
+    assert.are_same(search_prefix.match_prefix("über"), "über") -- German word with 'ü'
+    assert.are_same(search_prefix.match_prefix("übermensch"), "übermensch") -- German compound word with 'ü'
+    assert.are_same(search_prefix.match_prefix("mañana"), "mañana") -- Spanish word with 'ñ'
+    assert.are_same(search_prefix.match_prefix("Ångström"), "Ångström") -- Swedish word with 'Å' and 'ö'
+    assert.are_same(search_prefix.match_prefix("Straße"), "Straße") -- German word with 'ß'
+    assert.are_same(search_prefix.match_prefix("český"), "český") -- Czech word with 'č'
+    assert.are_same(search_prefix.match_prefix("naïve"), "naïve") -- French word with 'ï'
+    assert.are_same(search_prefix.match_prefix("façade"), "façade") -- French word with 'ç'
+    assert.are_same(search_prefix.match_prefix("résumé"), "résumé") -- French word with 'é'
+    assert.are_same(search_prefix.match_prefix("космос"), "космос") -- Russian word with Cyrillic characters
+    assert.are_same(search_prefix.match_prefix("你好"), "你好") -- Chinese characters
+    assert.are_same(search_prefix.match_prefix("日本語"), "日本語") -- Japanese characters
+    assert.are_same(search_prefix.match_prefix("한국어"), "한국어") -- Korean characters
+    assert.are_same(search_prefix.match_prefix("τοπική"), "τοπική") -- Greek word with 'π' and 'ή'
   end)
 
   it("matches emoji", function()
     -- because why not 😄
-    assert.are_same(blink_ripgrep.match_prefix("👍👎"), "👍👎")
-    assert.are_same(blink_ripgrep.match_prefix("👍-👎"), "👍-👎")
+    assert.are_same(search_prefix.match_prefix("👍👎"), "👍👎")
+    assert.are_same(search_prefix.match_prefix("👍-👎"), "👍-👎")
   end)
 
   it("does not include punctuation characters", function()
-    assert.are_same(blink_ripgrep.match_prefix("!hello"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("?world"), "world")
-    assert.are_same(blink_ripgrep.match_prefix("#hashtag"), "hashtag")
-    assert.are_same(blink_ripgrep.match_prefix("$money"), "money")
-    assert.are_same(blink_ripgrep.match_prefix("%value"), "value")
-    assert.are_same(blink_ripgrep.match_prefix("&and"), "and")
-    assert.are_same(blink_ripgrep.match_prefix("*star"), "star")
-    assert.are_same(blink_ripgrep.match_prefix("@email"), "email")
-    assert.are_same(blink_ripgrep.match_prefix("~tilde"), "tilde")
-    assert.are_same(blink_ripgrep.match_prefix(";semicolon"), "semicolon")
-    assert.are_same(blink_ripgrep.match_prefix(":colon"), "colon")
+    assert.are_same(search_prefix.match_prefix("!hello"), "hello")
+    assert.are_same(search_prefix.match_prefix("?world"), "world")
+    assert.are_same(search_prefix.match_prefix("#hashtag"), "hashtag")
+    assert.are_same(search_prefix.match_prefix("$money"), "money")
+    assert.are_same(search_prefix.match_prefix("%value"), "value")
+    assert.are_same(search_prefix.match_prefix("&and"), "and")
+    assert.are_same(search_prefix.match_prefix("*star"), "star")
+    assert.are_same(search_prefix.match_prefix("@email"), "email")
+    assert.are_same(search_prefix.match_prefix("~tilde"), "tilde")
+    assert.are_same(search_prefix.match_prefix(";semicolon"), "semicolon")
+    assert.are_same(search_prefix.match_prefix(":colon"), "colon")
   end)
 
   it("does not include whitespace and control characters", function()
-    assert.are_same(blink_ripgrep.match_prefix(" hello"), "hello")
-    assert.are_same(blink_ripgrep.match_prefix("world "), "world")
-    assert.are_same(blink_ripgrep.match_prefix("\t\ttext"), "text")
-    assert.are_same(blink_ripgrep.match_prefix("\nnewline"), "newline")
+    assert.are_same(search_prefix.match_prefix(" hello"), "hello")
+    assert.are_same(search_prefix.match_prefix("world "), "world")
+    assert.are_same(search_prefix.match_prefix("\t\ttext"), "text")
+    assert.are_same(search_prefix.match_prefix("\nnewline"), "newline")
   end)
 
   it("includes symbols", function()
-    assert.are_same(blink_ripgrep.match_prefix("©copyright"), "©copyright")
-    assert.are_same(blink_ripgrep.match_prefix("®registered"), "®registered")
-    assert.are_same(blink_ripgrep.match_prefix("™trademark"), "™trademark")
+    assert.are_same(search_prefix.match_prefix("©copyright"), "©copyright")
+    assert.are_same(search_prefix.match_prefix("®registered"), "®registered")
+    assert.are_same(search_prefix.match_prefix("™trademark"), "™trademark")
   end)
 end)