Skip to content

Fix heredoc #479

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Aug 8, 2017
Merged

Fix heredoc #479

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7be9763
Fix heredoc implementation
aycabta Aug 2, 2017
b0a0554
Fix tests for heredoc tokens
aycabta Aug 3, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
118 changes: 64 additions & 54 deletions lib/rdoc/ruby_lex.rb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,6 +106,8 @@ def initialize(content, options)
@rests = []
@seek = 0

@heredoc_queue = []

@indent = 0
@indent_stack = []
@lex_state = :EXPR_BEG
Expand Down Expand Up @@ -462,21 +464,43 @@ def lex_init()

@OP.def_rule("\n") do |op, io|
print "\\n\n" if RDoc::RubyLex.debug?
case @lex_state
when :EXPR_BEG, :EXPR_FNAME, :EXPR_DOT
@continue = true
else
@continue = false
@lex_state = :EXPR_BEG
until (@indent_stack.empty? ||
[TkLPAREN, TkLBRACK, TkLBRACE,
TkfLPAREN, TkfLBRACK, TkfLBRACE].include?(@indent_stack.last))
@indent_stack.pop
unless @heredoc_queue.empty?
info = @heredoc_queue[0]
if !info[:started] # "\n"
info[:started] = true
ungetc "\n"
elsif info[:heredoc_end].nil? # heredoc body
tk, heredoc_end = identify_here_document_body(info[:quoted], info[:lt], info[:indent])
info[:heredoc_end] = heredoc_end
ungetc "\n"
else # heredoc end
@heredoc_queue.shift
@lex_state = :EXPR_BEG
tk = Token(TkHEREDOCEND, info[:heredoc_end])
if !@heredoc_queue.empty?
@heredoc_queue[0][:started] = true
ungetc "\n"
end
end
end
@current_readed = @readed
@here_readed.clear
Token(TkNL)
unless tk
case @lex_state
when :EXPR_BEG, :EXPR_FNAME, :EXPR_DOT
@continue = true
else
@continue = false
@lex_state = :EXPR_BEG
until (@indent_stack.empty? ||
[TkLPAREN, TkLBRACK, TkLBRACE,
TkfLPAREN, TkfLBRACK, TkfLBRACE].include?(@indent_stack.last))
@indent_stack.pop
end
end
@current_readed = @readed
@here_readed.clear
tk = Token(TkNL)
end
tk
end

@OP.def_rules("*", "**",
Expand Down Expand Up @@ -506,8 +530,8 @@ def lex_init()
if @lex_state != :EXPR_END && @lex_state != :EXPR_CLASS &&
(@lex_state != :EXPR_ARG || @space_seen)
c = peek(0)
if /\S/ =~ c && (/["'`]/ =~ c || /\w/ =~ c || c == "-")
tk = identify_here_document
if /\S/ =~ c && (/["'`]/ =~ c || /\w/ =~ c || c == "-" || c == "~")
tk = identify_here_document(op)
end
end
unless tk
Expand Down Expand Up @@ -985,77 +1009,63 @@ def identify_identifier
end
end

def identify_here_document
def identify_here_document(op)
ch = getc
start_token = op
# if lt = PERCENT_LTYPE[ch]
if ch == "-"
if ch == "-" or ch == "~"
start_token.concat ch
ch = getc
indent = true
end
if /['"`]/ =~ ch
start_token.concat ch
user_quote = lt = ch
quoted = ""
while (c = getc) && c != lt
quoted.concat c
end
start_token.concat quoted
start_token.concat lt
else
user_quote = nil
lt = '"'
quoted = ch.dup
while (c = getc) && c =~ /\w/
quoted.concat c
end
start_token.concat quoted
ungetc
end

ltback, @ltype = @ltype, lt
reserve = []
while ch = getc
reserve.push ch
if ch == "\\"
reserve.push ch = getc
elsif ch == "\n"
break
end
end

output_heredoc = reserve.join =~ /\A\r?\n\z/
@heredoc_queue << {
quoted: quoted,
lt: lt,
indent: indent,
started: false
}
@lex_state = :EXPR_BEG
Token(RDoc::RubyLex::TkHEREDOCBEG, start_token)
end

if output_heredoc then
doc = '<<'
doc << '-' if indent
doc << "#{user_quote}#{quoted}#{user_quote}\n"
else
doc = '"'
end
def identify_here_document_body(quoted, lt, indent)
ltback, @ltype = @ltype, lt

@current_readed = @readed
doc = ""
heredoc_end = nil
while l = gets
l = l.sub(/(:?\r)?\n\z/, "\n")
if (indent ? l.strip : l.chomp) == quoted
heredoc_end = l
break
end
doc << l
end
raise Error, "Missing terminating #{quoted} for string" unless heredoc_end

if output_heredoc then
raise Error, "Missing terminating #{quoted} for string" unless l

doc << l.chomp
else
doc << '"'
end

@current_readed = @here_readed
@here_readed.concat reserve
while ch = reserve.pop
ungetc ch
end

token_class = output_heredoc ? RDoc::RubyLex::TkHEREDOC : Ltype2Token[lt]
@ltype = ltback
@lex_state = :EXPR_END
Token(token_class, doc)
@lex_state = :EXPR_BEG
[Token(RDoc::RubyLex::TkHEREDOC, doc), heredoc_end]
end

def identify_quotation
Expand Down
2 changes: 2 additions & 0 deletions lib/rdoc/ruby_token.rb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -329,6 +329,8 @@ def Token(token, value = nil)
[:TkCVAR, TkId],
[:TkIVAR, TkId],
[:TkCONSTANT, TkId],
[:TkHEREDOCBEG, TkId],
[:TkHEREDOCEND, TkId],

[:TkINTEGER, TkVal],
[:TkFLOAT, TkVal],
Expand Down
43 changes: 27 additions & 16 deletions test/test_rdoc_ruby_lex.rb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,9 +89,11 @@ def x
@TK::TkIDENTIFIER.new( 4, 1, 4, 'x'),
@TK::TkNL .new( 5, 1, 5, "\n"),
@TK::TkSPACE .new( 6, 2, 0, ' '),
@TK::TkHEREDOC .new( 8, 2, 2,
%Q{<<E\nLine 1\nLine 2\nE}),
@TK::TkNL .new(27, 5, 28, "\n"),

@TK::TkHEREDOCBEG.new( 8, 2, 2, '<<E'),
@TK::TkNL .new(11, 2, 6, "\n"),
@TK::TkHEREDOC .new(11, 2, 6, "Line 1\nLine 2\n"),
@TK::TkHEREDOCEND.new(27, 5, 26, "E\n"),
@TK::TkEND .new(28, 6, 0, 'end'),
@TK::TkNL .new(31, 6, 28, "\n"),
]
Expand Down Expand Up @@ -147,10 +149,12 @@ def test_class_tokenize_heredoc_CR_NL
@TK::TkSPACE .new( 6, 1, 6, ' '),
@TK::TkASSIGN .new( 7, 1, 7, '='),
@TK::TkSPACE .new( 8, 1, 8, ' '),
@TK::TkHEREDOC .new( 9, 1, 9,
%Q{<<-STRING\nLine 1\nLine 2\n STRING}),
@TK::TkSPACE .new(44, 4, 45, "\r"),
@TK::TkNL .new(45, 4, 46, "\n"),
@TK::TkHEREDOCBEG.new( 9, 1, 9, '<<-STRING'),
@TK::TkSPACE .new(18, 1, 18, "\r"),
@TK::TkNL .new(19, 1, 19, "\n"),
@TK::TkHEREDOC .new(19, 1, 19,
%Q{Line 1\nLine 2\n}),
@TK::TkHEREDOCEND.new(45, 4, 36, " STRING\n"),
]

assert_equal expected, tokens
Expand All @@ -169,10 +173,12 @@ def test_class_tokenize_heredoc_call
@TK::TkSPACE .new( 6, 1, 6, ' '),
@TK::TkASSIGN .new( 7, 1, 7, '='),
@TK::TkSPACE .new( 8, 1, 8, ' '),
@TK::TkSTRING .new( 9, 1, 9, %Q{"Line 1\nLine 2\n"}),
@TK::TkDOT .new(41, 4, 42, '.'),
@TK::TkIDENTIFIER.new(42, 4, 43, 'chomp'),
@TK::TkNL .new(47, 4, 48, "\n"),
@TK::TkHEREDOCBEG.new( 9, 1, 9, '<<-STRING'),
@TK::TkDOT .new(18, 1, 18, '.'),
@TK::TkIDENTIFIER.new(19, 1, 19, 'chomp'),
@TK::TkNL .new(24, 1, 24, "\n"),
@TK::TkHEREDOC .new(24, 1, 24, "Line 1\nLine 2\n"),
@TK::TkHEREDOCEND.new(47, 4, 39, " STRING\n"),
]

assert_equal expected, tokens
Expand All @@ -191,9 +197,12 @@ def test_class_tokenize_heredoc_indent
@TK::TkSPACE .new( 6, 1, 6, ' '),
@TK::TkASSIGN .new( 7, 1, 7, '='),
@TK::TkSPACE .new( 8, 1, 8, ' '),
@TK::TkHEREDOC .new( 9, 1, 9,
%Q{<<-STRING\nLine 1\nLine 2\n STRING}),
@TK::TkNL .new(41, 4, 42, "\n"),


@TK::TkHEREDOCBEG.new( 9, 1, 9, '<<-STRING'),
@TK::TkNL .new(18, 1, 18, "\n"),
@TK::TkHEREDOC .new(18, 1, 18, "Line 1\nLine 2\n"),
@TK::TkHEREDOCEND.new(41, 4, 33, " STRING\n")
]

assert_equal expected, tokens
Expand Down Expand Up @@ -223,8 +232,10 @@ def test_class_tokenize_heredoc_percent_N
@TK::TkSPACE .new( 1, 1, 1, ' '),
@TK::TkIDENTIFIER.new( 2, 1, 2, 'b'),
@TK::TkSPACE .new( 3, 1, 3, ' '),
@TK::TkHEREDOC .new( 4, 1, 4, %Q{<<-U\n%N\nU}),
@TK::TkNL .new(13, 3, 14, "\n"),
@TK::TkHEREDOCBEG.new( 4, 1, 4, '<<-U'),
@TK::TkNL .new( 8, 1, 8, "\n"),
@TK::TkHEREDOC .new( 8, 1, 8, "%N\n"),
@TK::TkHEREDOCEND.new(13, 3, 12, "U\n")
]

assert_equal expected, tokens
Expand Down