edoc_lib.erl

来自「OTP是开放电信平台的简称」· ERL 代码 · 共 222 行

%% =====================================================================
%% 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., 59 Temple Place, Suite 330, Boston, MA 02111-1307
%% USA
%%
%% $Id$
%%
%% @private
%% @copyright 2001-2003 Richard Carlsson
%% @author Richard Carlsson <[e-mail]>
%% @see edoc
%% @end
%% =====================================================================

%% @doc Utility functions for EDoc.

-module(edoc_lib).

-export([count/2, lines/1, split_at/2, split_at_stop/1,
	 split_at_space/1, filename/1, transpose/1, segment/2,
	 get_first_sentence/1, is_space/1, strip_space/1, parse_expr/2,
	 parse_contact/2, escape_uri/1, join_uri/2, is_relative_uri/1,
	 is_name/1, to_label/1, find_doc_dirs/0, find_sources/2,
	 find_sources/3, find_file/3, try_subdir/2, unique/1,
	 write_file/3, write_file/4, write_info_file/4,
	 read_info_file/1, get_doc_env/1, get_doc_env/4, copy_file/2,
	 uri_get/1, run_doclet/2, run_layout/2,
	 simplify_path/1, timestr/1, datestr/1]).

-import(edoc_report, [report/2, warning/2]).

-include("edoc.hrl").
-include("xmerl.hrl").

-define(FILE_BASE, "/").


%% ---------------------------------------------------------------------
%% List and string utilities

timestr({H,M,Sec}) ->
    lists:flatten(io_lib:fwrite("~2.2.0w:~2.2.0w:~2.2.0w",[H,M,Sec])).

datestr({Y,M,D}) ->
    Ms = ["Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep",
	  "Oct", "Nov", "Dec"],
    lists:flatten(io_lib:fwrite("~s ~w ~w",[lists:nth(M, Ms),D,Y])).

count(X, Xs) ->
    count(X, Xs, 0).

count(X, [X | Xs], N) ->
    count(X, Xs, N + 1);
count(X, [_ | Xs], N) ->
    count(X, Xs, N);
count(_X, [], N) ->
    N.

lines(Cs) ->
    lines(Cs, [], []).

lines([$\n | Cs], As, Ls) ->
    lines(Cs, [], [lists:reverse(As) | Ls]);
lines([C | Cs], As, Ls) ->
    lines(Cs, [C | As], Ls);
lines([], As, Ls) ->
    lists:reverse([lists:reverse(As) | Ls]).

split_at(Cs, K) ->
    split_at(Cs, K, []).

split_at([K | Cs], K, As) ->
    {lists:reverse(As), Cs};
split_at([C | Cs], K, As) ->
    split_at(Cs, K, [C | As]);
split_at([], _K, As) ->
    {lists:reverse(As), []}.

split_at_stop(Cs) ->
    split_at_stop(Cs, []).

split_at_stop([$., $\s | Cs], As) ->
    {lists:reverse(As), Cs};
split_at_stop([$., $\t | Cs], As) ->
    {lists:reverse(As), Cs};
split_at_stop([$., $\n | Cs], As) ->
    {lists:reverse(As), Cs};
split_at_stop([$.], As) ->
    {lists:reverse(As), []};
split_at_stop([C | Cs], As) ->
    split_at_stop(Cs, [C | As]);
split_at_stop([], As) ->
    {lists:reverse(As), []}.

split_at_space(Cs) ->
    split_at_space(Cs, []).

split_at_space([$\s | Cs], As) ->
    {lists:reverse(As), Cs};
split_at_space([$\t | Cs], As) ->
    {lists:reverse(As), Cs};
split_at_space([$\n | Cs], As) ->
    {lists:reverse(As), Cs};
split_at_space([C | Cs], As) ->
    split_at_space(Cs, [C | As]);
split_at_space([], As) ->
    {lists:reverse(As), []}.

is_space([$\s | Cs]) -> is_space(Cs);
is_space([$\t | Cs]) -> is_space(Cs);
is_space([$\n | Cs]) -> is_space(Cs);
is_space([_C | _Cs]) -> false;
is_space([]) -> true.

strip_space([$\s | Cs]) -> strip_space(Cs);
strip_space([$\t | Cs]) -> strip_space(Cs);
strip_space([$\n | Cs]) -> strip_space(Cs);
strip_space(Cs) -> Cs.

segment(Es, N) ->
    segment(Es, [], [], 0, N).

segment([E | Es], As, Cs, N, M) when N < M ->
    segment(Es, [E | As], Cs, N + 1, M);
segment([_ | _] = Es, As, Cs, _N, M) ->
    segment(Es, [], [lists:reverse(As) | Cs], 0, M);
segment([], [], Cs, _N, _M) ->
    lists:reverse(Cs);
segment([], As, Cs, _N, _M) ->
    lists:reverse([lists:reverse(As) | Cs]).

transpose([]) -> [];
transpose([[] | Xss]) -> transpose(Xss);
transpose([[X | Xs] | Xss]) ->
    [[X | [H || [H | _T] <- Xss]]
     | transpose([Xs | [T || [_H | T] <- Xss]])].

%% Note that the parser will not produce two adjacent text segments;
%% thus, if a text segment ends with a period character, it marks the
%% end of the summary sentence only if it is also the last segment in
%% the list, or is followed by a 'p' or 'br' ("whitespace") element.

get_first_sentence([#xmlElement{name = p, content = Es} | _]) ->
    %% Descend into initial paragraph.
    get_first_sentence_1(Es);
get_first_sentence(Es) ->
    get_first_sentence_1(Es).

get_first_sentence_1([E = #xmlText{value = Txt} | Es]) ->
    Last = case Es of
	       [#xmlElement{name = p} | _] -> true;
	       [#xmlElement{name = br} | _] -> true;
	       [] -> true;
	       _ -> false
	   end,
    case end_of_sentence(Txt, Last) of
	{value, Txt1} ->
	    [E#xmlText{value = Txt1}];
	none ->
	    [E | get_first_sentence_1(Es)]
    end;
get_first_sentence_1([E | Es]) ->
    % Skip non-text segments - don't descend further
    [E | get_first_sentence_1(Es)];
get_first_sentence_1([]) ->
    [].

end_of_sentence(Cs, Last) ->
    end_of_sentence(Cs, Last, []).

%% We detect '.' and '!' as end-of-sentence markers.

end_of_sentence([C=$., $\s | _], _, As) ->
    end_of_sentence_1(C, true, As);
end_of_sentence([C=$., $\t | _], _, As) ->
    end_of_sentence_1(C, true, As);
end_of_sentence([C=$., $\n | _], _, As) ->
    end_of_sentence_1(C, true, As);
end_of_sentence([C=$.], Last, As) ->
    end_of_sentence_1(C, Last, As);
end_of_sentence([C=$!, $\s | _], _, As) ->
    end_of_sentence_1(C, true, As);
end_of_sentence([C=$!, $\t | _], _, As) ->
    end_of_sentence_1(C, true, As);
end_of_sentence([C=$!, $\n | _], _, As) ->
    end_of_sentence_1(C, true, As);
end_of_sentence([C=$!], Last, As) ->
    end_of_sentence_1(C, Last, As);
end_of_sentence([C | Cs], Last, As) ->
    end_of_sentence(Cs, Last, [C | As]);
end_of_sentence([], Last, As) ->
    end_of_sentence_1($., Last, strip_space(As)).  % add a '.'

end_of_sentence_1(C, true, As) ->
    {value, lists:reverse([C | As])};
end_of_sentence_1(_, false, _) ->
    none.

%% For handling ISO 8859-1 (Latin-1) we use the following information:
%%
%% 000 - 037	NUL - US	control
%% 040 - 057	SPC - /		punctuation
%% 060 - 071	0 - 9		digit
%% 072 - 100	: - @		punctuation
%% 101 - 132	A - Z		uppercase
%% 133 - 140	[ - `		punctuation
%% 141 - 172	a - z		lowercase
%% 173 - 176	{ - ~		punctuation
%% 177		DEL		control
%% 200 - 237			control
%% 240 - 277	NBSP -