~ubuntu-branches/ubuntu/vivid/inform/vivid

<TR><TD Valign="top"><A HREF="contents.html">Contents</A> <A HREF="section14.html">Back</A> <A HREF="section16.html">Forward</A><TD bgcolor="#F5DEB3"><BLOCKQUOTE><H3>15. Reading matter and consultation</H3></BLOCKQUOTE><TR><TD><TD>

Even at present... we still know very little about how access

to printed materials affects human behaviour.

...Elizabeth Eisenstein, The Printing Revolution

in Early Modern Europe</BLOCKQUOTE>

look up figure 18 in the engineering textbook

</BLOCKQUOTE>

is a difficult line for Inform to understand, because almost anything

could appear in the first part: even its format depends on what the

second part is. This kind of request, and more generally

look up <any words here> in <the object>

read about <any words here> in <the object>

consult <the object> about <any words here>

</BLOCKQUOTE>

cause the <TT>Consult object</TT> action. Note that <TT>second</TT> is just zero:

formally, there is no second noun attached to a <TT>Consult</TT> action.

The object has to parse the <any words here> part itself, in a <TT>before</TT>

rule for <TT>Consult</TT>. The following variables are set up to make this

possible:

<TT>consult_from</TT> holds the number of the first word in the <any...>

clause;

<TT>consult_words</TT> holds the number of words in the <any...> clause

(at least 1).

The words given are parsed using library routines like

<TT>NextWord()</TT>, <TT>TryNumber(word-number)</TT> and so on: see <A HREF="section24.html">Section 24</A> for

full details. As usual, the <TT>before</TT> routine should return true

if it has managed to deal with the action; returning false will

make the library print "You discover nothing of interest in...''.

Little hints are placed here and there in the 'Ruins', written

in the glyphs of an ancient dialect of Mayan. Our explorer has,

of course, come equipped with the latest and finest scholarship

on the subject:

<PRE>

Object dictionary "Waldeck's Mayan dictionary"

with name "dictionary" "local" "guide" "book" "mayan"

"waldeck" "waldeck^s",

description "Compiled from the unreliable lithographs of the

legendary raconteur and explorer ~Count~ Jean Frederic

Maximilien Waldeck (1766??-1875), this guide contains

what little is known of the glyphs used in the local

ancient dialect.",

before

[ w1 w2 glyph other; Consult:

if (consult_words>2) jump GlyphHelp;

wn=consult_from;

w1 = NextWord(); ! First word of subject

w2 = NextWord(); ! Second word (if any) of subject

if (consult_words==1 && w1=='glyph' or 'glyphs')

jump GlyphHelp;

! We want to recognise both "glyph q1" and "q1 glyph":

glyph=w1; other=w2;

if (w1=='glyph') { glyph=w2; other=w1; }

! So now glyph holds the name, and other the other word

if (consult_words==2 && other~='glyph') jump GlyphHelp;

switch(glyph)

{ 'q1': "(This is one glyph you have memorised!)^^

Q1: ~sacred site~.";

'circle': "Circle: ~the Sun; also life, lifetime~.";

...

default: "That glyph is so far unrecorded.";

}

! All three of the ways the text can go wrong lead to

! this message being produced:

.GlyphHelp; "Try ~look up <name of glyph> in book~.";

has proper;

</PRE>

Note that this understands any of the forms "q1'', "glyph q1'' or

"q1 glyph'' but is careful to reject, for instance, "glyph q1 glyph''.

(These aren't genuine Mayan glyphs, but some of the real ones have similar

names, dating from when their syllabic equivalents weren't known: G8,

the Lord of the Night, for instance.)

<TR><TD Valign="top"><IMG SRC="icons/exercise.gif" ALT="??"><TD bgcolor="#FBB9AC"><A NAME="ex17">EXERCISE 17: (link to <A HREF="answers1/answer17.html">the answer</A>)<TR><TD><TD> To mark the 500th anniversary of William Tyndale (the first

English translator of the New Testament), prepare an edition of the

100

four Gospels.

101

102

103

104

Ordinarily, a request by the player to "read'' something is translated

105

into an <TT>Examine</TT> action. But the "read'' verb is defined independently

106

of the "examine'' verb in order to make it easy to separate the two

107

requests. For instance:

108

<PRE>

109

Attribute legible;

110

...

111

Object textbook "textbook"

112

with name "engineering" "textbook" "text" "book",

113

description "What beautiful covers and spine!",

114

before

115

[; Consult, Read:

116

"The pages are full of senseless equations.";

117

118

has legible;

119

...

120

[ ReadSub; <<Examine noun>>; ];

121

Extend "read" first * legible -> Read;

122

</PRE>

123

124

125

Note that "read" causes a <TT>Read</TT> action only for <TT>legible</TT> objects,

126

and otherwise causes <TT>Examine</TT> in the usual way. <TT>ReadSub</TT> is coded

127

as a translation to <TT>Examine</TT> as well, so that if a <TT>legible</TT> object

128

doesn't provide a <TT>Read</TT> rule then an <TT>Examine</TT> happens after all.

129

130

131

132

133

<TR><TD Valign="top"><IMG SRC="icons/refs.gif" ALT="*"><TD bgcolor="#EEEEEE">REFERENCES: If you really need more elaborate topic-parsing (for, e.g.,

134

"look up <something> in the catalogue'', where any object name might

135

appear) then extending the grammar for <TT>look</TT> may be less trouble.

136

For a good implementation see 'Encyclopaedia Frobozzica', by

137

Gareth Rees.

138

</TABLE>

139

<HR><A HREF="contents.html">Contents</A> / <A HREF="section14.html">Back</A> / <A HREF="section16.html">Forward</A>

140

<A HREF="chapter1.html">Chapter I</A> / <A HREF="chapter2.html">Chapter II</A> / <A HREF="chapter3.html">Chapter III</A> / <A HREF="chapter4.html">Chapter IV</A> / <A HREF="chapter5.html">Chapter V</A> / <A HREF="chapter6.html">Chapter VI</A> / <A HREF="chapterA.html">Appendix</A><HR>Mechanically translated to HTML from third edition as revised 16 May 1997. Copyright © Graham Nelson 1993, 1994, 1995, 1996, 1997: all rights reserved.</BODY></HTML>

Older »