@@ -54,6 +54,7 @@ def parse(str)
5454 T_STAR = :STAR # atom special; list wildcard
5555 T_PERCENT = :PERCENT # atom special; list wildcard
5656 T_LITERAL = :LITERAL # starts with atom special
57+ T_LITERAL8 = :LITERAL8 # starts with atom char "~"
5758 T_CRLF = :CRLF # atom special; text special; quoted special
5859 T_TEXT = :TEXT # any char except CRLF
5960 T_EOF = :EOF # end of response string
@@ -279,6 +280,16 @@ module RFC3629
279280 # ; sent from server to the client.
280281 LITERAL = /\{ (\d +)\} \r \n /n
281282
283+ # RFC3516 (BINARY):
284+ # literal8 = "~{" number "}" CRLF *OCTET
285+ # ; <number> represents the number of OCTETs
286+ # ; in the response string.
287+ # RFC9051:
288+ # literal8 = "~{" number64 "}" CRLF *OCTET
289+ # ; <number64> represents the number of OCTETs
290+ # ; in the response string.
291+ LITERAL8 = /~\{ (\d +)\} \r \n /n
292+
282293 module_function
283294
284295 def unescape_quoted! ( quoted )
@@ -298,27 +309,28 @@ def unescape_quoted(quoted)
298309 # the default, used in most places
299310 BEG_REGEXP = /\G (?:\
300311\
301- \
312+ #{ Patterns ::LITERAL8 } \
313+ \
302314\
303- 3 : NIL )(NIL)|\
304- 4 : NUMBER )(\d +)|\
305- 5 : PLUS )(\+ ))\
306- 6 : ATOM remaining after prefix )(#{ Patterns ::ATOMISH } \
315+ 4 : NIL )(NIL)|\
316+ 5 : NUMBER )(\d +)|\
317+ 6 : PLUS )(\+ ))\
318+ 7 : ATOM remaining after prefix )(#{ Patterns ::ATOMISH } \
307319\
308320\
309321\
310- 7 : ATOM )(#{ Patterns ::ATOMISH } \
311- 8 : QUOTED )#{ Patterns ::QUOTED_rev2 } \
312- 9 : LPAR )(\( )|\
313- 10 : RPAR )(\) )|\
314- 11 : BSLASH )(\\ )|\
315- 12 : STAR )(\* )|\
316- 13 : LBRA )(\[ )|\
317- 14 : RBRA )(\] )|\
318- 15 : LITERAL )#{ Patterns ::LITERAL } \
319- 16 : PERCENT )(%)|\
320- 17 : CRLF )(\r \n )|\
321- 18 : EOF )(\z ))/ni
322+ 8 : ATOM )(#{ Patterns ::ATOMISH } \
323+ 9 : QUOTED )#{ Patterns ::QUOTED_rev2 } \
324+ 10 : LPAR )(\( )|\
325+ 11 : RPAR )(\) )|\
326+ 12 : BSLASH )(\\ )|\
327+ 13 : STAR )(\* )|\
328+ 14 : LBRA )(\[ )|\
329+ 15 : RBRA )(\] )|\
330+ 16 : LITERAL )#{ Patterns ::LITERAL } \
331+ 17 : PERCENT )(%)|\
332+ 18 : CRLF )(\r \n )|\
333+ 19 : EOF )(\z ))/ni
322334
323335 # envelope, body(structure), namespaces
324336 DATA_REGEXP = /\G (?:\
@@ -359,6 +371,9 @@ def unescape_quoted(quoted)
359371 # string = quoted / literal
360372 def_token_matchers :string , T_QUOTED , T_LITERAL
361373
374+ # used by nstring8 = nstring / literal8
375+ def_token_matchers :string8 , T_QUOTED , T_LITERAL , T_LITERAL8
376+
362377 # use where string represents "LABEL" values
363378 def_token_matchers :case_insensitive__string ,
364379 T_QUOTED , T_LITERAL ,
@@ -460,6 +475,10 @@ def nstring
460475 NIL? ? nil : string
461476 end
462477
478+ def nstring8
479+ NIL? ? nil : string8
480+ end
481+
463482 def nquoted
464483 NIL? ? nil : quoted
465484 end
@@ -740,6 +759,8 @@ def msg_att(n)
740759 when "ENVELOPE" then envelope
741760 when "INTERNALDATE" then date_time
742761 when "RFC822.SIZE" then number64
762+ when /\A BINARY\[ /ni then nstring8 # BINARY, IMAP4rev2
763+ when /\A BINARY\. SIZE\[ /ni then number # BINARY, IMAP4rev2
743764 when "RFC822" then nstring # not in rev2
744765 when "RFC822.HEADER" then nstring # not in rev2
745766 when "RFC822.TEXT" then nstring # not in rev2
@@ -762,11 +783,18 @@ def msg_att__label
762783 lbra? and rbra
763784 when "BODY"
764785 peek_lbra? and name << section and
765- peek_str? ( "<" ) and name << atom # partial
786+ peek_str? ( "<" ) and name << gt__number__lt # partial
787+ when "BINARY" , "BINARY.SIZE"
788+ name << section_binary
789+ # see https://www.rfc-editor.org/errata/eid7246 and the note above
790+ peek_str? ( "<" ) and name << gt__number__lt # partial
766791 end
767792 name
768793 end
769794
795+ # this represents the partial size for BODY or BINARY
796+ alias gt__number__lt atom
797+
770798 def envelope
771799 @lex_state = EXPR_DATA
772800 token = lookahead
@@ -1070,6 +1098,13 @@ def section
10701098 str << rbra
10711099 end
10721100
1101+ # section-binary = "[" [section-part] "]"
1102+ def section_binary
1103+ str = +lbra
1104+ str << section_part unless peek_rbra?
1105+ str << rbra
1106+ end
1107+
10731108 # section-spec = section-msgtext / (section-part ["." section-text])
10741109 # section-msgtext = "HEADER" /
10751110 # "HEADER.FIELDS" [".NOT"] SP header-list /
@@ -1100,6 +1135,11 @@ def header_list
11001135 str << rpar
11011136 end
11021137
1138+ # section-part = nz-number *("." nz-number)
1139+ # ; body part reference.
1140+ # ; Allows for accessing nested body parts.
1141+ alias section_part atom
1142+
11031143 # RFC3501 & RFC9051:
11041144 # header-fld-name = astring
11051145 #
@@ -1789,42 +1829,47 @@ def next_token
17891829 @pos = $~. end ( 0 )
17901830 if $1
17911831 return Token . new ( T_SPACE , $+)
1792- elsif $2 && $6
1832+ elsif $2
1833+ len = $+. to_i
1834+ val = @str [ @pos , len ]
1835+ @pos += len
1836+ return Token . new ( T_LITERAL8 , val )
1837+ elsif $3 && $7
17931838 # greedily match ATOM, prefixed with NUMBER, NIL, or PLUS.
1794- return Token . new ( T_ATOM , $2)
1795- elsif $3
1796- return Token . new ( T_NIL , $+)
1839+ return Token . new ( T_ATOM , $3)
17971840 elsif $4
1798- return Token . new ( T_NUMBER , $+)
1841+ return Token . new ( T_NIL , $+)
17991842 elsif $5
1843+ return Token . new ( T_NUMBER , $+)
1844+ elsif $6
18001845 return Token . new ( T_PLUS , $+)
1801- elsif $7
1846+ elsif $8
18021847 # match ATOM, without a NUMBER, NIL, or PLUS prefix
18031848 return Token . new ( T_ATOM , $+)
1804- elsif $8
1805- return Token . new ( T_QUOTED , Patterns . unescape_quoted ( $+) )
18061849 elsif $9
1807- return Token . new ( T_LPAR , $+ )
1850+ return Token . new ( T_QUOTED , Patterns . unescape_quoted ( $+ ) )
18081851 elsif $10
1809- return Token . new ( T_RPAR , $+)
1852+ return Token . new ( T_LPAR , $+)
18101853 elsif $11
1811- return Token . new ( T_BSLASH , $+)
1854+ return Token . new ( T_RPAR , $+)
18121855 elsif $12
1813- return Token . new ( T_STAR , $+)
1856+ return Token . new ( T_BSLASH , $+)
18141857 elsif $13
1815- return Token . new ( T_LBRA , $+)
1858+ return Token . new ( T_STAR , $+)
18161859 elsif $14
1817- return Token . new ( T_RBRA , $+)
1860+ return Token . new ( T_LBRA , $+)
18181861 elsif $15
1862+ return Token . new ( T_RBRA , $+)
1863+ elsif $16
18191864 len = $+. to_i
18201865 val = @str [ @pos , len ]
18211866 @pos += len
18221867 return Token . new ( T_LITERAL , val )
1823- elsif $16
1824- return Token . new ( T_PERCENT , $+)
18251868 elsif $17
1826- return Token . new ( T_CRLF , $+)
1869+ return Token . new ( T_PERCENT , $+)
18271870 elsif $18
1871+ return Token . new ( T_CRLF , $+)
1872+ elsif $19
18281873 return Token . new ( T_EOF , $+)
18291874 else
18301875 parse_error ( "[Net::IMAP BUG] BEG_REGEXP is invalid" )
0 commit comments