1Marpa::XS::Advanced::BiUbsleirogCroanpthryi(b3u)ted PerlMaDropcau:m:eXnSt:a:tAidovnanced::Bibliography(3)
2
6 Marpa::XS::Advanced::Bibliography - A Marpa Bibliography
7
9 1970
10 Jay Earley invents the algorithm that now bears his name.
11 1991
13 Joop Leo describes a way to modify Earley's algorithm so that it
14 runs in O(n) time for all LR-regular grammars. LR-regular is a
15 vast class of grammars, including all the LR(k) grammars, all
16 grammars parseable with recursive descent, and regular expressions.
17 LR-regular can safely be thought of as including all grammars in
18 practical use today, and then some.
19 2002
21 Aycock and Horspool describe a way to do LR(0) precomputation for
22 Earley's algorithm. Their method makes Earley's faster in most
23 practical situations, but not all. In particular, right-recursion
24 remains quadratic in the Aycock and Horspool algorithm. Worst case
25 is no better than Earley's. Leo is unaware of Aycock and
26 Horspool's work and Aycock and Horspool seem unaware of Leo.
27 2010
29 Marpa combines the Leo and Aycock-Horspool algorithms, in the
30 process making significant changes to both of them. The result
31 preserves the best features of both. Marpa also tackles the many
32 remaining implementation issues.
33
35 Aho and Ullman 1972
36 The Theory of Parsing, Translation and Compiling, Volume I: Parsing by
37 Alfred Aho and Jeffrey Ullman (Prentice-Hall: Englewood Cliffs, New
38 Jersey, 1972). I think this was the standard source for Earley's
39 algorithm for decades. It certainly was my standard source. The
40 account of Earley's algorithm is on pages 320-330.
41 Aycock and Horspool 2002
43 Marpa is based on ideas from John Aycock and R. Nigel Horspool's
44 "Practical Earley Parsing", The Computer Journal, Vol. 45, No. 6, 2002,
45 pp. 620-630. The idea of doing LR(0) precomputation for Earley's
46 general parsing algorithm, and Marpa's approach to handling nullable
47 symbols and rules, both came from this article.
48 The Aycock and Horspool paper summarizes Earley's very nicely and is
50 available on the web:
51 <http://www.cs.uvic.ca/~nigelh/Publications/PracticalEarleyParsing.pdf>.
52 Unlike "Earley 1970", Aycock and Horspool 2002 is not easy reading. I
53 have been following this particular topic on and off for years and
54 nonetheless found this paper very heavy going.
55 Dominus 2005
57 Although my approach to parsing is not influenced by Mark Jason
58 Dominus's Higher Order Perl, Mark's treatment of parsing is an
59 excellent introduction to parsing, especially in a Perl context. His
60 focus on just about every other technique except general BNF parsing is
61 pretty much standard, and will help a beginner understand how
62 unconventional Marpa's approach is.
63 Mark's book opened my eyes to many new ideas. Both Mark's Perl and his
65 English are examples of good writing, and the book is dense with
66 insights. Mark's discussion on memoization in Chapter 3 is the best
67 I've seen. I wish I'd bought his book earlier in my coding.
68 Mark's book is available on-line. You can download chapter-by-chapter
70 or the whole thing at once, and you can take your pick of his original
71 sources or PDF, at <http://hop.perl.plover.com/book/>. A PDF of the
72 parsing chapter is at
73 <http://hop.perl.plover.com/book/pdf/08Parsing.pdf>.
74 Earley 1970
76 Of Jay Earley's papers on his general parsing algorithm, the most
77 readily available is "An efficient context-free parsing algorithm",
78 Communications of the Association for Computing Machinery, 13:2:94-102,
79 1970.
80 Ordinarily, I'd not bother pointing out 35-year old nits in a brilliant
82 and historically important article. But more than a few people treat
83 this article as not just the first word in Earley parsing, but the last
84 as well. Many implementations of Earley's algorithm come, directly and
85 unaltered, from his paper. These implementers and their users need to
86 be aware of two issues.
87 First, the recognition engine itself, as described, has a serious bug.
89 There's an easy fix, but one that greatly slows down an algorithm whose
90 main problem, in its original form, was speed. This issue is well laid
91 out by Aycock and Horspool in their article.
92 Second, according to Tomita there is a mistake in the parse tree
94 representation. See page 153 of "Grune and Jacobs 1990", page 210 of
95 "Grune and Jacobs 2008", and the bibliography entry for Earley 1970 in
96 "Grune and Jacobs 2008". In the printed edition of the 2008
97 bibliography, the entry is on page 578, and on the web
98 (<ftp://ftp.cs.vu.nl/pub/dick/PTAPG_2nd_Edition/CompleteList.pdf>),
99 it's on pp. 583-584. My methods for producing parse results from
100 Earley sets do not come from Earley 1970, so I am taking Tomita's word
101 on this one.
102 Grune and Jacobs 1990
104 Parsing Techniques: A Practical Guide, by Dick Grune and Ceriel Jacobs,
105 (Ellis Horwood Limited: Chichester, West Sussex, England, 1990). This
106 book is available on the Web: <http://www.cs.vu.nl/~dick/PTAPG.html>
107 Grune and Jacobs 2008
109 Parsing Techniques: A Practical Guide, by Dick Grune and Ceriel Jacobs,
110 2nd Edition. (Springer: New York NY, 2008). This is the most
111 authoritative and comprehensive introduction to parsing I know of. In
112 theory it requires no mathematics, only a programming background, but
113 even so it is moderately difficult reading.
114 This is "Grune and Jacobs 1990" updated. The bibliography for this
116 book is available in enlarged form on the web:
117 <ftp://ftp.cs.vu.nl/pub/dick/PTAPG_2nd_Edition/CompleteList.pdf>.
118 Leo 1991
120 Marpa's handling of right-recursion uses the ideas in Joop M.I.M. Leo's
121 "A General Context-Free Parsing Algorithm Running in Linear Time on
122 Every LR(k) Grammar Without Using Lookahead", Theoretical Computer
123 Science, Vol. 82, No. 1, 1991, pp 165-176. This is a difficult paper.
124 Unfortunately, there is no copy of it on-line.
125 Wikipedia
127 Wikipedia's article on Backus-Naur form is
128 <http://en.wikipedia.org/wiki/Backus-Naur_form>. It's a great place to
129 start if you don't know the basics of grammars and parsing. As
130 Wikipedia points out, BNF might better be called Panini-Backus Form.
131 The grammarian Panini gave a precise description of Sanskirt more than
132 23 centuries earlier in India using a similar notation.
133
135 Copyright 2012 Jeffrey Kegler
136 This file is part of Marpa::XS. Marpa::XS is free software: you can
137 redistribute it and/or modify it under the terms of the GNU Lesser
138 General Public License as published by the Free Software Foundation,
139 either version 3 of the License, or (at your option) any later version.
140 Marpa::XS is distributed in the hope that it will be useful,
142 but WITHOUT ANY WARRANTY; without even the implied warranty of
143 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
144 Lesser General Public License for more details.
145 You should have received a copy of the GNU Lesser
147 General Public License along with Marpa::XS. If not, see
148 http://www.gnu.org/licenses/.
149
151 Hey! The above document had some coding errors, which are explained
152 below:
153 Around line 29:
155 Expected text after =item, not a number
156 Around line 40:
158 Expected text after =item, not a number
159 Around line 53:
161 Expected text after =item, not a number
162perl v5.32.1 2021-01-2M7arpa::XS::Advanced::Bibliography(3)