aboutsummaryrefslogtreecommitdiffstats
path: root/lib/syntax_tools/src/erl_comment_scan.erl
blob: 04caa39eb32b25add92c631de0a52c639697ac7b (plain) (blame)
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
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
%% =====================================================================
%% This library is free software; you can redistribute it and/or modify
%% it under the terms of the GNU Lesser General Public License as
%% published by the Free Software Foundation; either version 2 of the
%% License, or (at your option) any later version.
%%
%% This library is distributed in the hope that it will be useful, but
%% WITHOUT ANY WARRANTY; without even the implied warranty of
%% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
%% Lesser General Public License for more details.
%%
%% You should have received a copy of the GNU Lesser General Public
%% License along with this library; if not, write to the Free Software
%% Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301
%% USA
%%
%% =====================================================================
%% @copyright 1997-2006 Richard Carlsson
%% @author Richard Carlsson <[email protected]>
%% @end
%% =====================================================================

%% @doc Functions for reading comment lines from Erlang source code.

-module(erl_comment_scan).

-export([file/1, join_lines/1, scan_lines/1, string/1]).

-export_type([comment/0]).

%% =====================================================================

-type comment()     :: {integer(), integer(), integer(), [string()]}.
-type commentLine() :: {integer(), integer(), integer(), string()}.

%% =====================================================================
%% @spec file(FileName::file:filename()) -> [Comment]
%%
%%	    Comment = {Line, Column, Indentation, Text}
%%	    Line = integer()
%%          Column = integer()
%%          Indentation = integer()
%%          Text = [string()]
%%
%% @doc Extracts comments from an Erlang source code file. Returns a
%% list of entries representing <em>multi-line</em> comments, listed in
%% order of increasing line-numbers. For each entry, `Text'
%% is a list of strings representing the consecutive comment lines in
%% top-down order; the strings contain <em>all</em> characters following
%% (but not including) the first comment-introducing `%'
%% character on the line, up to (but not including) the line-terminating
%% newline.
%%
%% Furthermore, `Line' is the line number and
%% `Column' the left column of the comment (i.e., the column
%% of the comment-introducing `%' character).
%% `Indent' is the indentation (or padding), measured in
%% character positions between the last non-whitespace character before
%% the comment (or the left margin), and the left column of the comment.
%% `Line' and `Column' are always positive
%% integers, and `Indentation' is a nonnegative integer.
%%
%% Evaluation exits with reason `{read, Reason}' if a read
%% error occurred, where `Reason' is an atom corresponding to
%% a Posix error code; see the module {@link //kernel/file} for details.

-spec file(file:filename()) -> [comment()].

file(Name) ->
    Name1 = filename(Name),
    case catch {ok, file:read_file(Name1)} of
	{ok, V} ->
	    case V of
		{ok, B} ->
                    Encoding = epp:read_encoding_from_binary(B),
                    Enc = case Encoding of
                              none -> epp:default_encoding();
                              Enc0 -> Enc0
                          end,
                    case catch unicode:characters_to_list(B, Enc) of
                        String when is_list(String) ->
                            string(String);
                        R when Encoding =:= none ->
                            case
                              catch unicode:characters_to_list(B, latin1)
                            of
                                String when is_list(String) ->
                                    string(String);
                                _ ->
                                    error_read_file(Name1),
                                    exit(R)
                            end;
                        R ->
                            error_read_file(Name1),
                            exit(R)
                    end;
		{error, E} ->
		    error_read_file(Name1),
		    exit({read, E})
	    end;
	{'EXIT', E} ->
	    error_read_file(Name1),
	    exit(E);
	R ->
	    error_read_file(Name1),
	    throw(R)
    end.


%% =====================================================================
%% @spec string(string()) -> [Comment]
%%
%%	    Comment = {Line, Column, Indentation, Text}
%%	    Line = integer()
%%          Column = integer()
%%          Indentation = integer()
%%          Text = [string()]
%%
%% @doc Extracts comments from a string containing Erlang source code.
%% Except for reading directly from a string, the behaviour is the same
%% as for {@link file/1}.
%%
%% @see file/1

-spec string(string()) -> [comment()].

string(Text) ->
    lists:reverse(join_lines(scan_lines(Text))).


%% =====================================================================
%% @spec scan_lines(string()) -> [CommentLine]
%%
%%	    CommentLine = {Line, Column, Indent, Text}
%%	    Line = integer()
%%	    Column = integer()
%%	    Indent = integer()
%%	    Text = string()
%%
%% @doc Extracts individual comment lines from a source code string.
%% Returns a list of comment lines found in the text, listed in order of
%% <em>decreasing</em> line-numbers, i.e., the last comment line in the
%% input is first in the resulting list. `Text' is a single
%% string, containing all characters following (but not including) the
%% first comment-introducing `%' character on the line, up
%% to (but not including) the line-terminating newline. For details on
%% `Line', `Column' and `Indent', see {@link file/1}.

-spec scan_lines(string()) -> [commentLine()].

scan_lines(Text) ->
    scan_lines(Text, 1, 0, 0, []).

scan_lines([$\040 | Cs], L, Col, M, Ack) ->
    scan_lines(Cs, L, Col + 1, M, Ack);
scan_lines([$\t | Cs], L, Col, M, Ack) ->
    scan_lines(Cs, L, tab(Col), M, Ack);
scan_lines([$\n | Cs], L, _Col, _M, Ack) ->
    scan_lines(Cs, L + 1, 0, 0, Ack);
scan_lines([$\r, $\n | Cs], L, _Col, _M, Ack) ->
    scan_lines(Cs, L + 1, 0, 0, Ack);
scan_lines([$\r | Cs], L, _Col, _M, Ack) ->
    scan_lines(Cs, L + 1, 0, 0, Ack);
scan_lines([$% | Cs], L, Col, M, Ack) ->
    scan_comment(Cs, "", L, Col, M, Ack);
scan_lines([$$ | Cs], L, Col, _M, Ack) ->
    scan_char(Cs, L, Col + 1, Ack);
scan_lines([$" | Cs], L, Col, _M, Ack) ->
    scan_string(Cs, $", L, Col + 1, Ack);
scan_lines([$' | Cs], L, Col, _M, Ack) ->
    scan_string(Cs, $', L, Col + 1, Ack);
scan_lines([_C | Cs], L, Col, _M, Ack) ->
    N = Col + 1,
    scan_lines(Cs, L, N, N, Ack);
scan_lines([], _L, _Col, _M, Ack) ->
    Ack.

tab(Col) ->
    Col - (Col rem 8) + 8.

scan_comment([$\n | Cs], Cs1, L, Col, M, Ack) ->
    seen_comment(Cs, Cs1, L, Col, M, Ack);
scan_comment([$\r, $\n | Cs], Cs1, L, Col, M, Ack) ->
    seen_comment(Cs, Cs1, L, Col, M, Ack);
scan_comment([$\r | Cs], Cs1, L, Col, M, Ack) ->
    seen_comment(Cs, Cs1, L, Col, M, Ack);
scan_comment([C | Cs], Cs1, L, Col, M, Ack) ->
    scan_comment(Cs, [C | Cs1], L, Col, M, Ack);
scan_comment([], Cs1, L, Col, M, Ack) ->
    seen_comment([], Cs1, L, Col, M, Ack).

%% Add a comment line to the ackumulator and return to normal
%% scanning. Note that we compute column positions starting at 0
%% internally, but the column values in the comment descriptors
%% should start at 1.

seen_comment(Cs, Cs1, L, Col, M, Ack) ->
    %% Compute indentation and strip trailing spaces
    N = Col - M,
    Text = lists:reverse(string:strip(Cs1, left)),
    Ack1 = [{L, Col + 1, N, Text} | Ack],
    scan_lines(Cs, L + 1, 0, 0, Ack1).

scan_string([Quote | Cs], Quote, L, Col, Ack) ->
    N = Col + 1,
    scan_lines(Cs, L, N, N, Ack);
scan_string([$\t | Cs], Quote, L, Col, Ack) ->
    scan_string(Cs, Quote, L, tab(Col), Ack);
scan_string([$\n | Cs], Quote, L, _Col, Ack) ->
    %% Newlines should really not occur in strings/atoms, but we
    %% want to be well behaved even if the input is not.
    scan_string(Cs, Quote, L + 1, 0, Ack);
scan_string([$\r, $\n | Cs], Quote, L, _Col, Ack) ->
    scan_string(Cs, Quote, L + 1, 0, Ack);
scan_string([$\r | Cs], Quote, L, _Col, Ack) ->
    scan_string(Cs, Quote, L + 1, 0, Ack);
scan_string([$\\, _C | Cs], Quote, L, Col, Ack) ->
    scan_string(Cs, Quote, L, Col + 2, Ack);  % ignore character C
scan_string([_C | Cs], Quote, L, Col, Ack) ->
    scan_string(Cs, Quote, L, Col + 1, Ack);
scan_string([], _Quote, _L, _Col, Ack) ->
    %% Finish quietly.
    Ack.

scan_char([$\t | Cs], L, Col, Ack) ->
    N = tab(Col),
    scan_lines(Cs, L, N, N, Ack);    % this is not just any whitespace
scan_char([$\n | Cs], L, _Col, Ack) ->
    scan_lines(Cs, L + 1, 0, 0, Ack);    % handle this, just in case
scan_char([$\r, $\n | Cs], L, _Col, Ack) ->
    scan_lines(Cs, L + 1, 0, 0, Ack);
scan_char([$\r | Cs], L, _Col, Ack) ->
    scan_lines(Cs, L + 1, 0, 0, Ack);
scan_char([$\\, _C | Cs], L, Col, Ack) ->
    N = Col + 2,    % character C must be ignored
    scan_lines(Cs, L, N, N, Ack);
scan_char([_C | Cs], L, Col, Ack) ->
    N = Col + 1,    % character C must be ignored
    scan_lines(Cs, L, N, N, Ack);
scan_char([], _L, _Col, Ack) ->
    %% Finish quietly.
    Ack.


%% =====================================================================
%% @spec join_lines([CommentLine]) -> [Comment]
%%
%%	    CommentLine = {Line, Column, Indent, string()}
%%	    Line = integer()
%%	    Column = integer()
%%	    Indent = integer()
%%	    Comment = {Line, Column, Indent, Text}
%%	    Text = [string()]
%%
%% @doc Joins individual comment lines into multi-line comments. The
%% input is a list of entries representing individual comment lines,
%% <em>in order of decreasing line-numbers</em>; see
%% {@link scan_lines/1} for details. The result is a list of
%% entries representing <em>multi-line</em> comments, <em>still listed
%% in order of decreasing line-numbers</em>, but where for each entry,
%% `Text' is a list of consecutive comment lines in order of
%% <em>increasing</em> line-numbers (i.e., top-down).
%%
%% @see scan_lines/1

-spec join_lines([commentLine()]) -> [comment()].

join_lines([{L, Col, Ind, Txt} | Lines]) ->
    join_lines(Lines, [Txt], L, Col, Ind);
join_lines([]) ->
    [].

%% In the following, we assume that the current `Txt' is never empty.
%% Recall that the list is in reverse line-number order.

join_lines([{L1, Col1, Ind1, Txt1} | Lines], Txt, L, Col, Ind) ->
    if L1 =:= L - 1, Col1 =:= Col, Ind + 1 =:= Col ->
	    %% The last test above checks that the previous
	    %% comment was alone on its line; otherwise it won't
	    %% be joined with the current; this is not always what
	    %% one wants, but works well in general.
	    join_lines(Lines, [Txt1 | Txt], L1, Col1, Ind1);
       true ->
	    %% Finish the current comment and let the new line
	    %% start the next one.
	    [{L, Col, Ind, Txt}
	     | join_lines(Lines, [Txt1], L1, Col1, Ind1)]
    end;
join_lines([], Txt, L, Col, Ind) ->
    [{L, Col, Ind, Txt}].


%% =====================================================================
%% Utility functions for internal use

filename([C|T]) when is_integer(C), C > 0 ->
    [C | filename(T)];
filename([]) ->
    [];
filename(N) ->
    report_error("bad filename: `~P'.", [N, 25]),
    exit(error).

error_read_file(Name) ->
    report_error("error reading file `~ts'.", [Name]).

report_error(S, Vs) ->
    error_logger:error_msg(lists:concat([?MODULE, ": ", S, "\n"]), Vs).

%% =====================================================================