glaemscribe 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/LICENSE.txt +19 -0
- data/bin/glaemscribe +307 -0
- data/glaemresources/charsets/cirth_ds.cst +205 -0
- data/glaemresources/charsets/sarati_eldamar.cst +256 -0
- data/glaemresources/charsets/tengwar_ds.cst +318 -0
- data/glaemresources/charsets/unicode_gothic.cst +64 -0
- data/glaemresources/charsets/unicode_runes.cst +120 -0
- data/glaemresources/modes/adunaic.glaem +251 -0
- data/glaemresources/modes/blackspeech-annatar.glaem +318 -0
- data/glaemresources/modes/blackspeech.glaem +260 -0
- data/glaemresources/modes/gothic.glaem +78 -0
- data/glaemresources/modes/khuzdul.glaem +141 -0
- data/glaemresources/modes/mercian.glaem +419 -0
- data/glaemresources/modes/oldnorse-medieval.glaem +127 -0
- data/glaemresources/modes/quenya-sarati.glaem +320 -0
- data/glaemresources/modes/quenya.glaem +307 -0
- data/glaemresources/modes/sindarin-beleriand.glaem +285 -0
- data/glaemresources/modes/sindarin-classical.glaem +276 -0
- data/glaemresources/modes/sindarin-daeron.glaem +182 -0
- data/glaemresources/modes/telerin.glaem +302 -0
- data/glaemresources/modes/valarin-sarati.glaem +210 -0
- data/glaemresources/modes/westron.glaem +340 -0
- data/glaemresources/modes/westsaxon.glaem +342 -0
- data/lib/api/charset.rb +84 -0
- data/lib/api/charset_parser.rb +55 -0
- data/lib/api/constants.rb +29 -0
- data/lib/api/debug.rb +36 -0
- data/lib/api/eval.rb +268 -0
- data/lib/api/fragment.rb +113 -0
- data/lib/api/glaeml.rb +200 -0
- data/lib/api/if_tree.rb +96 -0
- data/lib/api/mode.rb +112 -0
- data/lib/api/mode_parser.rb +314 -0
- data/lib/api/option.rb +64 -0
- data/lib/api/post_processor/reverse.rb +36 -0
- data/lib/api/pre_processor/downcase.rb +35 -0
- data/lib/api/pre_processor/elvish_numbers.rb +47 -0
- data/lib/api/pre_processor/rxsubstitute.rb +40 -0
- data/lib/api/pre_processor/substitute.rb +38 -0
- data/lib/api/pre_processor/up_down_tehta_split.rb +138 -0
- data/lib/api/resource_manager.rb +130 -0
- data/lib/api/rule.rb +99 -0
- data/lib/api/rule_group.rb +159 -0
- data/lib/api/sheaf.rb +70 -0
- data/lib/api/sheaf_chain.rb +86 -0
- data/lib/api/sheaf_chain_iterator.rb +108 -0
- data/lib/api/sub_rule.rb +40 -0
- data/lib/api/transcription_pre_post_processor.rb +118 -0
- data/lib/api/transcription_processor.rb +137 -0
- data/lib/api/transcription_tree_node.rb +91 -0
- data/lib/glaemscribe.rb +70 -0
- metadata +112 -0
data/lib/api/glaeml.rb
ADDED
@@ -0,0 +1,200 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
#
|
3
|
+
# Glǽmscribe (also written Glaemscribe) is a software dedicated to
|
4
|
+
# the transcription of texts between writing systems, and more
|
5
|
+
# specifically dedicated to the transcription of J.R.R. Tolkien's
|
6
|
+
# invented languages to some of his devised writing systems.
|
7
|
+
#
|
8
|
+
# Copyright (C) 2015 Benjamin Babut (Talagan).
|
9
|
+
#
|
10
|
+
# This program is free software: you can redistribute it and/or modify
|
11
|
+
# it under the terms of the GNU Affero General Public License as published by
|
12
|
+
# the Free Software Foundation, either version 3 of the License, or
|
13
|
+
# any later version.
|
14
|
+
#
|
15
|
+
# This program is distributed in the hope that it will be useful,
|
16
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
17
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
18
|
+
# GNU Affero General Public License for more details.
|
19
|
+
#
|
20
|
+
# You should have received a copy of the GNU Affero General Public License
|
21
|
+
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
22
|
+
|
23
|
+
module Glaemscribe
|
24
|
+
module API
|
25
|
+
module Glaeml
|
26
|
+
|
27
|
+
class Error
|
28
|
+
attr_accessor :text
|
29
|
+
attr_accessor :line
|
30
|
+
|
31
|
+
def initialize(line,text)
|
32
|
+
@line = line
|
33
|
+
@text = text
|
34
|
+
end
|
35
|
+
end
|
36
|
+
|
37
|
+
class Document
|
38
|
+
attr_accessor :root_node
|
39
|
+
attr_accessor :errors
|
40
|
+
|
41
|
+
def initialize
|
42
|
+
@errors = []
|
43
|
+
end
|
44
|
+
end
|
45
|
+
|
46
|
+
class Node
|
47
|
+
|
48
|
+
class Type
|
49
|
+
Text = 0
|
50
|
+
ElementInline = 1
|
51
|
+
ElementBlock = 2
|
52
|
+
end
|
53
|
+
|
54
|
+
def text?
|
55
|
+
@type == Type::Text
|
56
|
+
end
|
57
|
+
|
58
|
+
def element?
|
59
|
+
@type == Type::ElementInline || @type == Type::ElementBlock
|
60
|
+
end
|
61
|
+
|
62
|
+
def initialize(line, type, name)
|
63
|
+
@line = line
|
64
|
+
@type = type
|
65
|
+
@name = name
|
66
|
+
@args = []
|
67
|
+
@children = []
|
68
|
+
end
|
69
|
+
|
70
|
+
def pathfind_crawl(apath, found)
|
71
|
+
|
72
|
+
children.each{ |c|
|
73
|
+
if(c.name == apath[0])
|
74
|
+
if apath.count == 1
|
75
|
+
found << c
|
76
|
+
else
|
77
|
+
bpath = apath.dup
|
78
|
+
bpath.shift
|
79
|
+
c.pathfind_crawl(bpath, found)
|
80
|
+
end
|
81
|
+
end
|
82
|
+
}
|
83
|
+
end
|
84
|
+
|
85
|
+
def gpath(path)
|
86
|
+
apath = path.split(".")
|
87
|
+
found = []
|
88
|
+
pathfind_crawl(apath, found)
|
89
|
+
found
|
90
|
+
end
|
91
|
+
|
92
|
+
attr_accessor :line
|
93
|
+
|
94
|
+
attr_reader :type
|
95
|
+
attr_reader :name
|
96
|
+
|
97
|
+
attr_accessor :args
|
98
|
+
attr_reader :children
|
99
|
+
|
100
|
+
attr_accessor :parent_node
|
101
|
+
end
|
102
|
+
|
103
|
+
|
104
|
+
|
105
|
+
class Parser
|
106
|
+
|
107
|
+
def add_text_node(lnum, text)
|
108
|
+
n = Node.new(lnum, Node::Type::Text, "root")
|
109
|
+
n.args << text
|
110
|
+
n.parent_node = @current_parent_node
|
111
|
+
@current_parent_node.children << n
|
112
|
+
end
|
113
|
+
|
114
|
+
def parse(raw_data)
|
115
|
+
raw_data.gsub!(/\\\*\*(.*?)\*\*\\/m) { |found|
|
116
|
+
"\n" * found.count("\n")
|
117
|
+
}
|
118
|
+
|
119
|
+
lnum = 0
|
120
|
+
|
121
|
+
@document = Document.new
|
122
|
+
@document.root_node = Node.new(lnum, Node::Type::ElementBlock, nil)
|
123
|
+
|
124
|
+
@current_parent_node = @document.root_node
|
125
|
+
|
126
|
+
raw_data.lines.each{ |l|
|
127
|
+
lnum += 1
|
128
|
+
|
129
|
+
l = l.strip
|
130
|
+
add_text_node(lnum, l) and next if l.empty?
|
131
|
+
|
132
|
+
if(l[0..0] == "\\")
|
133
|
+
if(l.length == 1)
|
134
|
+
@document.errors << Error.new(lnum, "Incomplete node.")
|
135
|
+
else
|
136
|
+
if(l[1..1] == "\\") # First backslash is escaped
|
137
|
+
|
138
|
+
add_text_node(lnum, l[1..-1])
|
139
|
+
|
140
|
+
elsif l =~ /^(\\beg\s+)/
|
141
|
+
|
142
|
+
rest = $'.strip
|
143
|
+
args = []
|
144
|
+
name = "???"
|
145
|
+
if(! (rest =~ /^([a-z_]+)/) )
|
146
|
+
@document.errors << Error.new(lnum, "Bad element name #{rest}.")
|
147
|
+
else
|
148
|
+
name = $1
|
149
|
+
args = $'.shellsplit
|
150
|
+
end
|
151
|
+
|
152
|
+
n = Node.new(lnum, Node::Type::ElementBlock, name)
|
153
|
+
n.args += args
|
154
|
+
n.parent_node = @current_parent_node
|
155
|
+
|
156
|
+
@current_parent_node.children << n
|
157
|
+
@current_parent_node = n
|
158
|
+
|
159
|
+
elsif l =~ /^(\\end(\s|$))/
|
160
|
+
|
161
|
+
if !@current_parent_node.parent_node
|
162
|
+
@document.errors << Error.new(lnum, "Element 'end' unexpected.")
|
163
|
+
elsif $'.strip != ""
|
164
|
+
@document.errors << Error.new(lnum, "Element 'end' should not have arguments.")
|
165
|
+
else
|
166
|
+
@current_parent_node = @current_parent_node.parent_node
|
167
|
+
end
|
168
|
+
|
169
|
+
else
|
170
|
+
|
171
|
+
# Read the name of the node
|
172
|
+
l = l[1..-1]
|
173
|
+
l =~ /^([a-z_]+)/
|
174
|
+
name = $1
|
175
|
+
args = $'.shellsplit
|
176
|
+
|
177
|
+
n = Node.new(lnum, Node::Type::ElementInline, name)
|
178
|
+
n.args += args
|
179
|
+
n.parent_node = @current_parent_node
|
180
|
+
|
181
|
+
@current_parent_node.children << n
|
182
|
+
|
183
|
+
end
|
184
|
+
end
|
185
|
+
else
|
186
|
+
add_text_node(lnum, l)
|
187
|
+
end
|
188
|
+
}
|
189
|
+
|
190
|
+
if @current_parent_node != @document.root_node
|
191
|
+
@document.errors << Error.new(lnum, "Missing 'end' element.")
|
192
|
+
end
|
193
|
+
|
194
|
+
return @document
|
195
|
+
|
196
|
+
end
|
197
|
+
end
|
198
|
+
end
|
199
|
+
end
|
200
|
+
end
|
data/lib/api/if_tree.rb
ADDED
@@ -0,0 +1,96 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
#
|
3
|
+
# Glǽmscribe (also written Glaemscribe) is a software dedicated to
|
4
|
+
# the transcription of texts between writing systems, and more
|
5
|
+
# specifically dedicated to the transcription of J.R.R. Tolkien's
|
6
|
+
# invented languages to some of his devised writing systems.
|
7
|
+
#
|
8
|
+
# Copyright (C) 2015 Benjamin Babut (Talagan).
|
9
|
+
#
|
10
|
+
# This program is free software: you can redistribute it and/or modify
|
11
|
+
# it under the terms of the GNU Affero General Public License as published by
|
12
|
+
# the Free Software Foundation, either version 3 of the License, or
|
13
|
+
# any later version.
|
14
|
+
#
|
15
|
+
# This program is distributed in the hope that it will be useful,
|
16
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
17
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
18
|
+
# GNU Affero General Public License for more details.
|
19
|
+
#
|
20
|
+
# You should have received a copy of the GNU Affero General Public License
|
21
|
+
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
22
|
+
|
23
|
+
module Glaemscribe
|
24
|
+
module API
|
25
|
+
module IfTree
|
26
|
+
|
27
|
+
class IfCond
|
28
|
+
attr_accessor :line, :expression, :parent_if_term, :child_code_block
|
29
|
+
def initialize(line, parent_if_term, expression)
|
30
|
+
@parent_if_term = parent_if_term
|
31
|
+
@expression = expression
|
32
|
+
end
|
33
|
+
end
|
34
|
+
|
35
|
+
class Term
|
36
|
+
attr_accessor :parent_code_block
|
37
|
+
def initialize(parent_code_block)
|
38
|
+
@parent_code_block = parent_code_block
|
39
|
+
end
|
40
|
+
def is_code_lines?
|
41
|
+
false
|
42
|
+
end
|
43
|
+
def is_pre_post_processor_operators?
|
44
|
+
false
|
45
|
+
end
|
46
|
+
end
|
47
|
+
|
48
|
+
class IfTerm < Term
|
49
|
+
attr_accessor :if_conds
|
50
|
+
def initialize(parent_code_block)
|
51
|
+
super(parent_code_block)
|
52
|
+
@if_conds = []
|
53
|
+
end
|
54
|
+
end
|
55
|
+
|
56
|
+
class CodeLine
|
57
|
+
attr_accessor :expression, :line
|
58
|
+
def initialize(expression, line)
|
59
|
+
@expression = expression
|
60
|
+
@line = line
|
61
|
+
end
|
62
|
+
end
|
63
|
+
|
64
|
+
class PrePostProcessorOperatorsTerm < Term
|
65
|
+
attr_accessor :operators
|
66
|
+
def initialize(parent_code_block)
|
67
|
+
super(parent_code_block)
|
68
|
+
@operators = []
|
69
|
+
end
|
70
|
+
def is_pre_post_processor_operators?
|
71
|
+
true
|
72
|
+
end
|
73
|
+
end
|
74
|
+
|
75
|
+
class CodeLinesTerm < Term
|
76
|
+
attr_accessor :code_lines
|
77
|
+
def initialize(parent_code_block)
|
78
|
+
super(parent_code_block)
|
79
|
+
@code_lines = []
|
80
|
+
end
|
81
|
+
def is_code_lines?
|
82
|
+
true
|
83
|
+
end
|
84
|
+
end
|
85
|
+
|
86
|
+
class CodeBlock
|
87
|
+
attr_accessor :terms, :parent_if_cond
|
88
|
+
def initialize(parent_if_cond = nil)
|
89
|
+
@parent_if_cond = parent_if_cond
|
90
|
+
@terms = []
|
91
|
+
end
|
92
|
+
end
|
93
|
+
|
94
|
+
end
|
95
|
+
end
|
96
|
+
end
|
data/lib/api/mode.rb
ADDED
@@ -0,0 +1,112 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
#
|
3
|
+
# Glǽmscribe (also written Glaemscribe) is a software dedicated to
|
4
|
+
# the transcription of texts between writing systems, and more
|
5
|
+
# specifically dedicated to the transcription of J.R.R. Tolkien's
|
6
|
+
# invented languages to some of his devised writing systems.
|
7
|
+
#
|
8
|
+
# Copyright (C) 2015 Benjamin Babut (Talagan).
|
9
|
+
#
|
10
|
+
# This program is free software: you can redistribute it and/or modify
|
11
|
+
# it under the terms of the GNU Affero General Public License as published by
|
12
|
+
# the Free Software Foundation, either version 3 of the License, or
|
13
|
+
# any later version.
|
14
|
+
#
|
15
|
+
# This program is distributed in the hope that it will be useful,
|
16
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
17
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
18
|
+
# GNU Affero General Public License for more details.
|
19
|
+
#
|
20
|
+
# You should have received a copy of the GNU Affero General Public License
|
21
|
+
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
22
|
+
|
23
|
+
module Glaemscribe
|
24
|
+
module API
|
25
|
+
class Mode
|
26
|
+
|
27
|
+
attr_accessor :errors
|
28
|
+
attr_accessor :warnings
|
29
|
+
|
30
|
+
attr_accessor :name
|
31
|
+
|
32
|
+
attr_accessor :language, :writing, :human_name, :authors, :version
|
33
|
+
attr_accessor :pre_processor, :processor, :post_processor
|
34
|
+
|
35
|
+
attr_accessor :options
|
36
|
+
|
37
|
+
attr_accessor :supported_charsets
|
38
|
+
attr_accessor :default_charset
|
39
|
+
|
40
|
+
def initialize(name)
|
41
|
+
@name = name
|
42
|
+
@errors = []
|
43
|
+
@warnings = []
|
44
|
+
@supported_charsets = {}
|
45
|
+
@options = {}
|
46
|
+
|
47
|
+
@pre_processor = TranscriptionPreProcessor.new(self)
|
48
|
+
@processor = TranscriptionProcessor.new(self)
|
49
|
+
@post_processor = TranscriptionPostProcessor.new(self)
|
50
|
+
end
|
51
|
+
|
52
|
+
def to_s
|
53
|
+
" <Mode #{name}: Language '#{language}', Writing '#{writing}', Human Name '#{human_name}', Authors '#{authors}', Version '#{version}'> "
|
54
|
+
end
|
55
|
+
|
56
|
+
def inspect
|
57
|
+
to_s
|
58
|
+
end
|
59
|
+
|
60
|
+
def finalize(options={})
|
61
|
+
|
62
|
+
# Hash: option_name => value_name
|
63
|
+
trans_options = {}
|
64
|
+
|
65
|
+
# Build default options
|
66
|
+
@options.each { |oname, o|
|
67
|
+
trans_options[oname] = o.default_value_name
|
68
|
+
}
|
69
|
+
|
70
|
+
# Push user options
|
71
|
+
options.each { |oname, valname|
|
72
|
+
# Check if option exists
|
73
|
+
opt = @options[oname]
|
74
|
+
next if(!opt)
|
75
|
+
|
76
|
+
val = opt.value_for_value_name(valname)
|
77
|
+
next if val == nil
|
78
|
+
|
79
|
+
trans_options[oname] = valname
|
80
|
+
}
|
81
|
+
|
82
|
+
trans_options_converted = {}
|
83
|
+
|
84
|
+
# Do a conversion to values space
|
85
|
+
trans_options.each{ |oname,valname|
|
86
|
+
trans_options_converted[oname] = @options[oname].value_for_value_name(valname)
|
87
|
+
}
|
88
|
+
|
89
|
+
@pre_processor.finalize(trans_options_converted)
|
90
|
+
@post_processor.finalize(trans_options_converted)
|
91
|
+
@processor.finalize(trans_options_converted)
|
92
|
+
|
93
|
+
self
|
94
|
+
end
|
95
|
+
|
96
|
+
def transcribe(content, charset = nil)
|
97
|
+
charset = default_charset if !charset
|
98
|
+
return false, "*** No charset usable for transcription. Failed!" if !charset
|
99
|
+
|
100
|
+
ret = content.lines.map{ |l|
|
101
|
+
l = l.strip # Clean the lines
|
102
|
+
l = @pre_processor.apply(l)
|
103
|
+
l = @processor.apply(l, charset)
|
104
|
+
l = @post_processor.apply(l)
|
105
|
+
}.join("\n")
|
106
|
+
|
107
|
+
return true, ret
|
108
|
+
end
|
109
|
+
|
110
|
+
end
|
111
|
+
end
|
112
|
+
end
|
@@ -0,0 +1,314 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
#
|
3
|
+
# Glǽmscribe (also written Glaemscribe) is a software dedicated to
|
4
|
+
# the transcription of texts between writing systems, and more
|
5
|
+
# specifically dedicated to the transcription of J.R.R. Tolkien's
|
6
|
+
# invented languages to some of his devised writing systems.
|
7
|
+
#
|
8
|
+
# Copyright (C) 2015 Benjamin Babut (Talagan).
|
9
|
+
#
|
10
|
+
# This program is free software: you can redistribute it and/or modify
|
11
|
+
# it under the terms of the GNU Affero General Public License as published by
|
12
|
+
# the Free Software Foundation, either version 3 of the License, or
|
13
|
+
# any later version.
|
14
|
+
#
|
15
|
+
# This program is distributed in the hope that it will be useful,
|
16
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
17
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
18
|
+
# GNU Affero General Public License for more details.
|
19
|
+
#
|
20
|
+
# You should have received a copy of the GNU Affero General Public License
|
21
|
+
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
22
|
+
|
23
|
+
module Glaemscribe
|
24
|
+
module API
|
25
|
+
|
26
|
+
class ModeParser
|
27
|
+
|
28
|
+
def initialize
|
29
|
+
@mode = nil
|
30
|
+
end
|
31
|
+
|
32
|
+
def validate_presence_of_args(node, arg_count = nil)
|
33
|
+
if(arg_count)
|
34
|
+
if(node.args.count != arg_count)
|
35
|
+
@mode.errors << Glaeml::Error.new(node.line,"Element '#{node.name}' should have #{arg_count} arguments.")
|
36
|
+
end
|
37
|
+
end
|
38
|
+
end
|
39
|
+
|
40
|
+
def validate_presence_of_children(parent_node, elt_name, elt_count = nil, arg_count = nil)
|
41
|
+
res = parent_node.gpath(elt_name)
|
42
|
+
if(elt_count)
|
43
|
+
if(res.count != elt_count)
|
44
|
+
@mode.errors << Glaeml::Error.new(parent_node.line,"Element '#{parent_node.name}' should have exactly #{elt_count} children of type '#{elt_name}'.")
|
45
|
+
end
|
46
|
+
end
|
47
|
+
if(arg_count)
|
48
|
+
res.each{ |child_node|
|
49
|
+
validate_presence_of_args(child_node, arg_count)
|
50
|
+
}
|
51
|
+
end
|
52
|
+
end
|
53
|
+
|
54
|
+
# Very simplified 'dtd' like verification
|
55
|
+
def verify_mode_glaeml(doc)
|
56
|
+
validate_presence_of_children(doc.root_node, "language", 1, 1)
|
57
|
+
validate_presence_of_children(doc.root_node, "writing", 1, 1)
|
58
|
+
validate_presence_of_children(doc.root_node, "mode", 1, 1)
|
59
|
+
validate_presence_of_children(doc.root_node, "authors", 1, 1)
|
60
|
+
validate_presence_of_children(doc.root_node, "version", 1, 1)
|
61
|
+
|
62
|
+
doc.root_node.gpath("charset").each{ |charset_element|
|
63
|
+
validate_presence_of_args(charset_element, 2)
|
64
|
+
}
|
65
|
+
|
66
|
+
doc.root_node.gpath("options.option").each{ |option_element|
|
67
|
+
validate_presence_of_args(option_element, 2)
|
68
|
+
option_element.gpath("value").each{ |value_element|
|
69
|
+
validate_presence_of_args(value_element, 2)
|
70
|
+
}
|
71
|
+
}
|
72
|
+
|
73
|
+
doc.root_node.gpath("processor.outspace").each{ |outspace_element|
|
74
|
+
validate_presence_of_args(outspace_element, 1)
|
75
|
+
}
|
76
|
+
|
77
|
+
doc.root_node.gpath("processor.rules").each{ |rules_element|
|
78
|
+
validate_presence_of_args(rules_element, 1)
|
79
|
+
}
|
80
|
+
end
|
81
|
+
|
82
|
+
def create_if_cond_for_if_term(line, if_term, cond)
|
83
|
+
ifcond = IfTree::IfCond.new(line, if_term, cond)
|
84
|
+
child_code_block = IfTree::CodeBlock.new(ifcond)
|
85
|
+
ifcond.child_code_block = child_code_block
|
86
|
+
if_term.if_conds << ifcond
|
87
|
+
ifcond
|
88
|
+
end
|
89
|
+
|
90
|
+
def traverse_if_tree(root_code_block, root_element, text_procedure, element_procedure)
|
91
|
+
current_parent_code_block = root_code_block
|
92
|
+
|
93
|
+
root_element.children.each{ |child|
|
94
|
+
if(child.text?)
|
95
|
+
# Do something with this text
|
96
|
+
text_procedure.call(current_parent_code_block, child)
|
97
|
+
elsif(child.element?)
|
98
|
+
case child.name
|
99
|
+
when 'if'
|
100
|
+
cond_attribute = child.args.first
|
101
|
+
if_term = IfTree::IfTerm.new(current_parent_code_block)
|
102
|
+
current_parent_code_block.terms << if_term
|
103
|
+
if_cond = create_if_cond_for_if_term(child.line, if_term, cond_attribute)
|
104
|
+
current_parent_code_block = if_cond.child_code_block
|
105
|
+
|
106
|
+
when 'elsif'
|
107
|
+
|
108
|
+
cond_attribute = child.args.first
|
109
|
+
if_term = current_parent_code_block.parent_if_cond.parent_if_term
|
110
|
+
|
111
|
+
if !if_term
|
112
|
+
@mode.errors << Glaeml::Error.new(child.line, " 'elsif' without a 'if'.")
|
113
|
+
return
|
114
|
+
end
|
115
|
+
|
116
|
+
# TODO : check that precendent one is a elsif, not a else
|
117
|
+
if_cond = create_if_cond_for_if_term(child.line, if_term, cond_attribute)
|
118
|
+
current_parent_code_block = if_cond.child_code_block
|
119
|
+
|
120
|
+
when 'else'
|
121
|
+
|
122
|
+
if_term = current_parent_code_block.parent_if_cond.parent_if_term
|
123
|
+
|
124
|
+
if !if_term
|
125
|
+
@mode.errors << Glaeml::Error.new(child.line, " 'else' without a 'if'.")
|
126
|
+
return
|
127
|
+
end
|
128
|
+
|
129
|
+
if_cond = create_if_cond_for_if_term(child.line, if_term, "true")
|
130
|
+
current_parent_code_block = if_cond.child_code_block
|
131
|
+
|
132
|
+
when 'endif'
|
133
|
+
if_term = current_parent_code_block.parent_if_cond.parent_if_term
|
134
|
+
|
135
|
+
if !if_term
|
136
|
+
@mode.errors << Glaeml::Error.new(child.line, " 'endif' without a 'if'.")
|
137
|
+
return
|
138
|
+
end
|
139
|
+
|
140
|
+
current_parent_code_block = if_term.parent_code_block
|
141
|
+
|
142
|
+
else
|
143
|
+
# Do something with this child element
|
144
|
+
element_procedure.call(current_parent_code_block, child)
|
145
|
+
end
|
146
|
+
end
|
147
|
+
}
|
148
|
+
|
149
|
+
if(current_parent_code_block.parent_if_cond)
|
150
|
+
@mode.errors << Glaeml::Error.new(root_element.line, "Unended 'if' at the end of this '#{root_element.name}' element.")
|
151
|
+
end
|
152
|
+
end
|
153
|
+
|
154
|
+
def parse_pre_post_processor(processor_element, pre_not_post)
|
155
|
+
# Do nothing with text elements
|
156
|
+
text_procedure = Proc.new { |current_parent_code_block, element| }
|
157
|
+
|
158
|
+
element_procedure = Proc.new { |current_parent_code_block, element|
|
159
|
+
|
160
|
+
# A block of code lines. Put them in a codelinesterm.
|
161
|
+
term = current_parent_code_block.terms.last
|
162
|
+
if(!term || !term.is_pre_post_processor_operators?)
|
163
|
+
term = IfTree::PrePostProcessorOperatorsTerm.new(current_parent_code_block)
|
164
|
+
current_parent_code_block.terms << term
|
165
|
+
end
|
166
|
+
|
167
|
+
operator_name = element.name
|
168
|
+
operator_class = if pre_not_post
|
169
|
+
ResourceManager::class_for_pre_processor_operator_name(operator_name)
|
170
|
+
else
|
171
|
+
ResourceManager::class_for_post_processor_operator_name(operator_name)
|
172
|
+
end
|
173
|
+
|
174
|
+
if !operator_class
|
175
|
+
@mode.errors << Glaeml::Error.new(element.line,"Operator #{operator_name} is unknown.")
|
176
|
+
else
|
177
|
+
arg0 = element.args[0]
|
178
|
+
arg1 = element.args[1]
|
179
|
+
arg2 = element.args[2]
|
180
|
+
arg3 = element.args[3]
|
181
|
+
|
182
|
+
term.operators << operator_class.new([arg0,arg1,arg2,arg3])
|
183
|
+
end
|
184
|
+
}
|
185
|
+
|
186
|
+
root_code_block = ((pre_not_post)?(@mode.pre_processor.root_code_block):(@mode.post_processor.root_code_block))
|
187
|
+
|
188
|
+
self.traverse_if_tree(root_code_block, processor_element, text_procedure, element_procedure )
|
189
|
+
end
|
190
|
+
|
191
|
+
def parse(file_path, mode_options = {})
|
192
|
+
|
193
|
+
@mode = Mode.new(ResourceManager::mode_name_from_file_path(file_path))
|
194
|
+
|
195
|
+
raw = File.open(file_path,"rb:utf-8").read
|
196
|
+
doc = Glaeml::Parser.new.parse(raw)
|
197
|
+
|
198
|
+
if(doc.errors.any?)
|
199
|
+
@mode.errors = doc.errors
|
200
|
+
return @mode
|
201
|
+
end
|
202
|
+
|
203
|
+
verify_mode_glaeml(doc)
|
204
|
+
|
205
|
+
return @mode if(@mode.errors.any?)
|
206
|
+
|
207
|
+
# Get the attributes of the mode
|
208
|
+
@mode.language = doc.root_node.gpath('language').first.args.first
|
209
|
+
@mode.writing = doc.root_node.gpath('writing').first.args.first
|
210
|
+
@mode.human_name = doc.root_node.gpath('mode').first.args.first
|
211
|
+
@mode.authors = doc.root_node.gpath('authors').first.args.first
|
212
|
+
@mode.version = doc.root_node.gpath('version').first.args.first
|
213
|
+
|
214
|
+
doc.root_node.gpath("options.option").each{ |option_element|
|
215
|
+
values = {}
|
216
|
+
option_element.gpath("value").each{ |value_element|
|
217
|
+
value_name = value_element.args.first
|
218
|
+
values[value_name] = value_element.args.last.to_i
|
219
|
+
}
|
220
|
+
|
221
|
+
option_name_at = option_element.args[0]
|
222
|
+
option_default_val_at = option_element.args[1]
|
223
|
+
# TODO: check syntax of the option name
|
224
|
+
|
225
|
+
if(option_default_val_at.nil?)
|
226
|
+
@mode.errors << Glaeml::Error.new(option_element.line, "Missing option default value.")
|
227
|
+
end
|
228
|
+
|
229
|
+
option = Option.new(option_name_at, option_default_val_at, values)
|
230
|
+
@mode.options[option.name] = option
|
231
|
+
}
|
232
|
+
|
233
|
+
# Read the supported font list
|
234
|
+
doc.root_node.gpath("charset").each { |charset_element|
|
235
|
+
charset_name = charset_element.args.first
|
236
|
+
charset = ResourceManager::charset(charset_name)
|
237
|
+
|
238
|
+
# Load the charset if we don't have it
|
239
|
+
if !charset
|
240
|
+
ResourceManager::load_charsets([charset_name])
|
241
|
+
charset = ResourceManager::charset(charset_name)
|
242
|
+
end
|
243
|
+
if charset
|
244
|
+
if charset.errors.any?
|
245
|
+
charset.errors.each{ |e|
|
246
|
+
@mode.errors << Glaeml::Error.new(charset_element.line, "#{charset_name}:#{e.line}:#{e.text}")
|
247
|
+
}
|
248
|
+
return @mode
|
249
|
+
end
|
250
|
+
|
251
|
+
@mode.supported_charsets[charset_name] = charset
|
252
|
+
@mode.default_charset = charset if charset_element.args[1] && charset_element.args[1] == "true"
|
253
|
+
else
|
254
|
+
@mode.warnings << Glaeml::Error.new(charset_element.line,"Failed to load charset '#{charset_name}'.")
|
255
|
+
end
|
256
|
+
}
|
257
|
+
|
258
|
+
if !@mode.default_charset
|
259
|
+
@mode.warnings << Glaeml::Error.new(0,"No default charset defined!!")
|
260
|
+
end
|
261
|
+
|
262
|
+
# Read the preprocessor
|
263
|
+
doc.root_node.gpath("preprocessor").each{ |preprocessor_element|
|
264
|
+
self.parse_pre_post_processor(preprocessor_element, true)
|
265
|
+
}
|
266
|
+
|
267
|
+
# Read the postprocessor
|
268
|
+
doc.root_node.gpath("postprocessor").each{ |postprocessor_element|
|
269
|
+
self.parse_pre_post_processor(postprocessor_element, false)
|
270
|
+
}
|
271
|
+
|
272
|
+
# Read the processor
|
273
|
+
doc.root_node.gpath("processor.outspace").each{ |outspace_element|
|
274
|
+
val = outspace_element.args[0]
|
275
|
+
@mode.processor.out_space = val.split.reject{|token| token.empty? }
|
276
|
+
}
|
277
|
+
|
278
|
+
doc.root_node.gpath("processor.rules").each{ |rules_element|
|
279
|
+
|
280
|
+
rule_group_name = rules_element.args.first
|
281
|
+
rule_group = RuleGroup.new(@mode,rule_group_name)
|
282
|
+
@mode.processor.rule_groups[rule_group_name] = rule_group
|
283
|
+
|
284
|
+
text_procedure = Proc.new { |current_parent_code_block, element|
|
285
|
+
# A block of code lines. Put them in a codelinesterm.
|
286
|
+
term = current_parent_code_block.terms.last
|
287
|
+
if(!term || !term.is_code_lines?)
|
288
|
+
term = IfTree::CodeLinesTerm.new(current_parent_code_block)
|
289
|
+
current_parent_code_block.terms << term
|
290
|
+
end
|
291
|
+
|
292
|
+
lcount = element.line
|
293
|
+
element.args[0].lines.to_a.each{ |l|
|
294
|
+
l = l.strip
|
295
|
+
term.code_lines << IfTree::CodeLine.new(l, lcount)
|
296
|
+
lcount += 1
|
297
|
+
}
|
298
|
+
}
|
299
|
+
|
300
|
+
element_procedure = Proc.new { |current_parent_code_block, element|
|
301
|
+
# This is fatal.
|
302
|
+
@mode.errors << Glaeml::Error.new(element.line, "Unknown directive #{element.name}.")
|
303
|
+
}
|
304
|
+
|
305
|
+
self.traverse_if_tree( rule_group.root_code_block, rules_element, text_procedure, element_procedure )
|
306
|
+
}
|
307
|
+
|
308
|
+
@mode.finalize(mode_options) if !@mode.errors.any?
|
309
|
+
|
310
|
+
@mode
|
311
|
+
end
|
312
|
+
end
|
313
|
+
end
|
314
|
+
end
|