aboutsummaryrefslogtreecommitdiff
path: root/modules/textadept/mime_types.lua
blob: 2675b9fcd33acfd6e24427371988464cde94d121 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
-- Copyright 2007-2012 Mitchell mitchell.att.foicica.com. See LICENSE.

local M = {}

--[[ This comment is for LuaDoc.
---
-- Handles file type detection.
-- @field _G.events.LANGUAGE_MODULE_LOADED (string)
--   Called after loading a language-specific module.
--   This is useful for overriding a language-specific module's key bindings
--   or other properties since the module is not loaded when Textadept starts.
--   Arguments:
--
--   * `lang`: The language lexer name.
module('_M.textadept.mime_types')]]

-- Events.
local events, events_connect = events, events.connect
events.LANGUAGE_MODULE_LOADED = 'language_module_loaded'

---
-- Table of file extensions with their associated lexers.
-- If the file type is not recognized by shebang words or first-line patterns,
-- each file extension is matched against the file's extension.
-- @class table
-- @name extensions
M.extensions = {}

---
-- Table of shebang words and their associated lexers.
-- If the file has a shebang line, a line that starts with "#!" and is the first
-- line in the file, each shebang word is matched against that line.
-- @class table
-- @name shebangs
M.shebangs = {}

---
-- Table of first-line patterns and their associated lexers.
-- If a file type is not recognized by shebang words, each pattern is matched
-- against the first line in the file.
-- @class table
-- @name patterns
M.patterns = {}

-- Load mime-types from *modules/textadept/mime_types.conf*.
local mime_types
local f = io.open(_HOME..'/modules/textadept/mime_types.conf', 'rb')
if f then
  mime_types = f:read('*all')
  f:close()
end
f = io.open(_USERHOME..'/mime_types.conf', 'rb')
if f then
  mime_types = mime_types..'\n'..f:read('*all')
  f:close()
end
for line in mime_types:gmatch('[^\r\n]+') do
  if not line:find('^%s*%%') then
    if line:find('^%s*[^#/]') then -- extension definition
      local ext, lexer_name = line:match('^%s*(.+)%s+(%S+)$')
      if ext and lexer_name then M.extensions[ext] = lexer_name end
    else -- shebang or pattern
      local ch, text, lexer_name = line:match('^%s*([#/])(.+)%s+(%S+)$')
      if ch and text and lexer_name then
        (ch == '#' and M.shebangs or M.patterns)[text] = lexer_name
      end
    end
  end
end

---
-- List of detected lexers.
-- Lexers are read from *lexers/* and *~/.textadept/lexers/*.
-- @class table
-- @name lexers
M.lexers = {}

-- Generate lexer list
local lexers_found = {}
for lexer in lfs.dir(_HOME..'/lexers') do
  if lexer:find('%.lua$') and lexer ~= 'lexer.lua' then
    lexers_found[lexer:match('^(.+)%.lua$')] = true
  end
end
if lfs.attributes(_USERHOME..'/lexers') then
  for lexer in lfs.dir(_USERHOME..'/lexers') do
    if lexer:find('%.lua$') and lexer ~= 'lexer.lua' then
      lexers_found[lexer:match('^(.+)%.lua$')] = true
    end
  end
end
for lexer in pairs(lexers_found) do M.lexers[#M.lexers + 1] = lexer end
table.sort(M.lexers)

---
-- Prompts the user to select a lexer from a filtered list for the current
-- buffer.
-- @see buffer.set_lexer
-- @name select_lexer
function M.select_lexer()
  local lexer = gui.filteredlist(_L['Select Lexer'], _L['Name'], M.lexers)
  if lexer then buffer:set_lexer(lexer) end
end

-- LuaDoc is in core/.buffer.luadoc.
local function get_style_name(buffer, style_num)
  buffer:check_global()
  if style_num < 0 or style_num > 255 then error('0 <= style_num < 256') end
  return buffer:private_lexer_call(style_num)
end

-- Contains the whitespace styles for lexers.
-- These whitespace styles are used to determine the lexer at the current caret
-- position since the styles have the name "[lang]_whitespace".
-- @class table
-- @name ws_styles
local ws_styles = {}
local SETDIRECTPOINTER = _SCINTILLA.properties.doc_pointer[2]
local SETLEXERLANGUAGE = _SCINTILLA.properties.lexer_language[2]
-- LuaDoc is in core/.buffer.luadoc.
local function set_lexer(buffer, lang)
  buffer:check_global()
  buffer._lexer = lang
  buffer:private_lexer_call(SETDIRECTPOINTER, buffer.direct_pointer)
  buffer:private_lexer_call(SETLEXERLANGUAGE, lang)
  if package.searchpath(lang, package.path) then
    _M[lang] = require(lang)
    local post_init = lang..'.post_init'
    if package.searchpath(post_init, package.path) then require(post_init) end
    if _M[lang].set_buffer_properties then _M[lang].set_buffer_properties() end
    events.emit(events.LANGUAGE_MODULE_LOADED, lang)
  end
  buffer:colourise(0, -1)
  -- Create the ws_styles[lexer] lookup table for `get_lexer()`.
  if ws_styles[lang] then return end
  local ws = {}
  for i = 0, 255 do
    ws[i] = buffer:private_lexer_call(i):find('whitespace') ~= nil
  end
  ws_styles[lang] = ws
end

local GETLEXERLANGUAGE = _SCINTILLA.properties.lexer_language[1]
-- LuaDoc is in core/.buffer.luadoc.
local function get_lexer(buffer, current)
  buffer:check_global()
  local lexer = buffer:private_lexer_call(GETLEXERLANGUAGE)
  if not current then return lexer end
  local i, ws, style_at = buffer.current_pos, ws_styles[lexer], buffer.style_at
  if ws then while i > 0 and not ws[style_at[i]] do i = i - 1 end end
  return get_style_name(buffer, style_at[i]):match('^(.+)_whitespace$') or lexer
end

events_connect(events.BUFFER_NEW, function()
  buffer.set_lexer, buffer.get_lexer = set_lexer, get_lexer
  buffer.get_style_name = get_style_name
end, 1)
-- Scintilla's first buffer does not have these.
if not RESETTING then
  buffer.set_lexer, buffer.get_lexer = set_lexer, get_lexer
  buffer.get_style_name = get_style_name
end

-- Performs actions suitable for a new buffer.
-- Sets the buffer's lexer language and loads the language module.
local function handle_new()
  local lexer
  local line = buffer:get_line(0)
  if line:find('^#!') then
    for word in line:gsub('[/\\]', ' '):gmatch('%S+') do
      lexer = M.shebangs[word]
      if lexer then break end
    end
  end
  if not lexer then
    for patt, lex in pairs(M.patterns) do
      if line:find(patt) then lexer = lex break end
    end
  end
  if not lexer and buffer.filename then
    lexer = M.extensions[buffer.filename:match('[^/\\.]+$')]
  end
  buffer:set_lexer(lexer or 'text')
end
events_connect(events.FILE_OPENED, handle_new)
events_connect(events.FILE_SAVED_AS, handle_new)

-- Sets the buffer's lexer based on filename, shebang words, or
-- first line pattern.
local function restore_lexer()
  buffer:private_lexer_call(SETDIRECTPOINTER, buffer.direct_pointer)
  buffer:private_lexer_call(SETLEXERLANGUAGE, buffer._lexer or 'text')
end
events_connect(events.BUFFER_AFTER_SWITCH, restore_lexer)
events_connect(events.VIEW_NEW, restore_lexer, 1)

events_connect(events.RESET_AFTER,
               function() buffer:set_lexer(buffer._lexer or 'text') end)

return M