| blob | 7c2f106ac31de153d687efd00c376d1676a89079 |
2 <!DOCTYPE html
3 PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
7 <head>
23 </head>
24 <body>
29 transforming lemmings into gold.
30 </p>
33 <!-- ####################################################### -->
34 <hr />
36 <ul>
39 </ul>
41 <hr />
43 <!-- ####################################################### -->
47 are not implemented as pointers. They are a generalization of
48 pointers, but they are implemented in libstdc++ as separate classes.
49 </p>
50 <p>Keeping that simple fact in mind as you design your code will
51 prevent a whole lot of difficult-to-understand bugs.
52 </p>
53 <p>You can think of it the other way 'round, even. Since iterators
56 iterator would be. All those functions in the Algorithms chapter
57 of the Standard will work just as well on plain arrays and their
58 pointers.
59 </p>
60 <p>That doesn't mean that when you pass in a pointer, it gets wrapped
61 into some special delegating iterator-to-pointer class with a layer
62 of overhead. (If you think that's the case anywhere, you don't
63 understand templates to begin with...) Oh, no; if you pass
64 in a pointer, then the compiler will instantiate that template
65 using T* as a type, and good old high-speed pointer arithmetic as
66 its operations, so the resulting code will be doing exactly the same
67 things as it would be doing if you had hand-coded it yourself (for
68 the 273rd time).
69 </p>
71 Very little. Most of the layering classes contain nothing but
73 tell the compiler some nicknames; they don't create code. That
74 information gets passed down through inheritance, so while the
75 compiler has to do work looking up all the names, your runtime code
76 does not. (This has been a prime concern from the beginning.)
77 </p>
80 </p>
82 <hr />
84 <p>This starts off sounding complicated, but is actually very easy,
85 especially towards the end. Trust me.
86 </p>
87 <p>Beginners usually have a little trouble understand the whole
88 'past-the-end' thing, until they remember their early algebra classes
90 the concept of half-open ranges.
91 </p>
92 <p>First, some history, and a reminder of some of the funkier rules in
93 C and C++ for builtin arrays. The following rules have always been
94 true for both languages:
95 </p>
96 <ol>
98 past the end of the array</em>. A pointer that points to one
99 past the end of the array is guaranteed to be as unique as a
100 pointer to somewhere inside the array, so that you can compare
101 such pointers safely.
102 </li>
103 <li>You can only dereference a pointer that points into an array.
104 If your array pointer points outside the array -- even to just
105 one past the end -- and you dereference it, Bad Things happen.
106 </li>
107 <li>Strictly speaking, simply pointing anywhere else invokes
108 undefined behavior. Most programs won't puke until such a
109 pointer is actually dereferenced, but the standards leave that
110 up to the platform.
111 </li>
112 </ol>
113 <p>The reason this past-the-end addressing was allowed is to make it
114 easy to write a loop to go over an entire array, e.g.,
115 while (*d++ = *s++);.
116 </p>
117 <p>So, when you think of two pointers delimiting an array, don't think
119 markers</em>:
120 </p>
121 <pre>
123 beginning end
124 | |
125 | | This is bad. Always having to
126 | | remember to add or subtract one.
127 | | Off-by-one bugs very common here.
128 V V
129 array of N elements
130 |---|---|--...--|---|---|
132 |---|---|--...--|---|---|
134 ^ ^
135 | |
136 | | This is good. This is safe. This
137 | | is guaranteed to work. Just don't
138 | | dereference 'end'.
139 beginning end
141 </pre>
142 <p>See? Everything between the boundary markers is part of the array.
143 Simple.
144 </p>
145 <p>Now think back to your junior-high school algebra course, when you
146 were learning how to draw graphs. Remember that a graph terminating
148 and a graph terminating with an open dot meant, "Everything up
149 to, but not including, this point," respectively called closed
150 and open ranges? Remember how closed ranges were written with
153 </p>
155 starting with (and including) the first element, and ending with (but
157 told you it would be simple in the end.
158 </p>
159 <p>Iterators, and everything working with iterators, follows this same
162 method returns a past-the-end iterator, which is guaranteed to be
163 unique and comparable against any other iterator pointing into the
164 middle of the container.
165 </p>
166 <p>Container constructors, container methods, and algorithms, all take
167 pairs of iterators describing a range of values on which to operate.
168 All of these ranges are half-open ranges, so you pass the beginning
169 iterator as the starting parameter, and the one-past-the-end iterator
170 as the finishing parameter.
171 </p>
172 <p>This generalizes very well. You can operate on sub-ranges quite
174 don't know or care whether they are the boundaries of an entire {array,
175 sequence, container, whatever}, or whether they only enclose a few
176 elements from the center. This approach also makes zero-length
177 sequences very simple to recognize: if the two endpoints compare
178 equal, then the {array, sequence, container, whatever} is empty.
179 </p>
181 </p>
184 </p>
189 <!-- ####################################################### -->
191 <hr />
194 Comments and suggestions are welcome, and may be sent to
196 </em></p>
199 </body>
200 </html>