Commit | Line | Data |
---|---|---|
1528431b MW |
1 | Notes on Lisp style |
2 | ||
3 | * Language subset and extensions | |
4 | ||
5 | None of ANSI Common Lisp is off-limits. | |
6 | ||
828cb3b1 MW |
7 | I think my Lisp style is rather more imperative in flavour than most |
8 | modern Lisp programmers. It's probably closer to historical Lisp | |
9 | practice in that regard, even though I wasn't writing Lisp back then. | |
10 | ||
1528431b MW |
11 | I make extensive use of CLOS, and macros. On a couple of occasions I've |
12 | made macros which use CLOS generic function dispatch to compute their | |
13 | expansions. The parser language is probably the best example of this in | |
828cb3b1 MW |
14 | the codebase. |
15 | ||
16 | I like hairy ~format~ strings. | |
1528431b MW |
17 | |
18 | I've avoided hairy ~loop~ for the most part, not because I dislike it | |
19 | strongly but because others do and I don't find that it wins big enough | |
20 | for the fight to be worthwhile. | |
21 | ||
22 | I only use ~&aux~ lambda-list parameters in ~defstruct~ BOA | |
23 | constructors, for special effects. | |
24 | ||
25 | I use ~car~, not ~first~, and ~cdr~, not ~rest~. Similarly, I use | |
26 | ~cadr~, not ~second~, and I'm not afraid to use ~cddr~ or ~cadar~. | |
27 | ||
28 | Similarly, I've not used ~elt~, preferring to know what kind of sequence | |
29 | I'm dealing with, or using the built-in sequence functions. | |
30 | ||
31 | I'm happy to use ~1+~, and I like the brevity of ~1-~ enough to use it | |
32 | despite its terrible name. | |
33 | ||
34 | There are no reader syntax extensions in the code. This is because I | |
35 | couldn't think of any way they'd be especially helpful, and not because | |
36 | I'm in any way opposed to them. | |
37 | ||
38 | The main translator, in the ~SOD~ package, tries to assume very little | |
39 | beyond ANSI Common Lisp and what's included in just about every serious | |
40 | implementation: specifically, MOP introspection, and Gray streams. | |
41 | There's intentionally no MOP intercession. | |
42 | ||
7e55d099 MW |
43 | The frontend additionally makes use of ~cl-launch~, but the dependency |
44 | is actually quite weak, and it could be replaced with a different, maybe | |
1528431b MW |
45 | implementation-specific, mechanism fairly easily. I'm keen to take |
46 | patches which improve frontend portability. | |
47 | ||
48 | I'm more tolerant of extensions and external dependencies in the test | |
49 | suite, which makes additional use of ~xlunit~. Running the test suite | |
50 | isn't essential to getting the translator built, so this isn't as much | |
51 | of a problem. | |
52 | ||
53 | ||
54 | * Layout | |
55 | ||
56 | I pretty much let Emacs indent my code for me, based on information | |
57 | collected by SLIME. Some exceptions: | |
58 | ||
59 | + DSLs (e.g., the parser language) have their own space of macros | |
60 | which Emacs doesn't understand and for the most part I haven't | |
61 | bothered to teach it. | |
62 | ||
63 | + Emacs sometimes does a bad job with hairy ~loop~ and requires manual | |
64 | fixing. Since I don't use hairy ~loop~ much, this isn't a major | |
65 | problem. | |
66 | ||
67 | Lines are 77 characters at most, except for strange special effects. | |
68 | Don't ask. This is not negotiable, though. Don't try to tell me that | |
69 | your monitor is very wide so you can read longer lines. My monitor is | |
70 | likely at least as wide. On the other hand, most lines are easily short | |
71 | enough to fit in my narrow columns, so the right hand side of a wide | |
72 | window would be mostly blank. This seems wasteful to me, when I could | |
73 | fill that space with more code. | |
74 | ||
75 | Lisp code does have a tendency to march across to the right quite | |
76 | rapidly given a chance. I have a number of strategies for dealing with | |
77 | this. | |
78 | ||
79 | + Break a long nested calculation into pieces, giving names to the | |
80 | intermediate results, in a ~let*~ form. | |
81 | ||
7e55d099 MW |
82 | + Hoist deeply nested complex computations out into ~flet~ or |
83 | ~labels~, and then invoke them from inside whatever complicated | |
1528431b MW |
84 | conditional mess was needed to decide what to do. |
85 | ||
86 | + Shrug my shoulders and let code dribble down the right hand side for | |
87 | a bit. | |
88 | ||
89 | ||
90 | * Packages and exporting | |
91 | ||
92 | A package collects symbols which are given meanings in one or more | |
93 | source files. If a package's code is all in one file, then the package | |
94 | definition can be put in that file too; otherwise I put it in its own | |
95 | file. | |
96 | ||
97 | I don't put ~:export~ in package definitions. Instead, I scatter calls | |
98 | to the ~export~ function throughout the code, right next to where the | |
99 | relevant symbol is defined. This has three important advantages. | |
100 | ||
101 | + You can tell, when you're reading the code which defines ~foo~, | |
102 | whether ~foo~ is exported and therefore a defined part of the | |
103 | package interface. | |
104 | ||
105 | + When you know that you're writing a thing which will form part of | |
106 | the package interface, you don't have to go off and edit some other | |
107 | file to export it. | |
108 | ||
109 | + A master list of exported symbols becomes a merge hazard: if two | |
110 | different branches add symbols to nearby pieces of the master list | |
111 | then you get a merge conflict for no especially good reason. | |
112 | ||
113 | There's an apparent disadvantage: there's no immediately visible master | |
114 | list of exported symbols. But that's not a big problem: | |
115 | ||
116 | : (loop for s being the external-symbols of pkg collect s) | |
117 | ||
118 | See ~doc/list-symbols.lisp~ for more sophisticated reporting. (In | |
119 | particular, this identifies what kind of thing(s) each external symbol | |
120 | names.) | |
121 | ||
122 | ||
123 | * Comments and file structuring | |
124 | ||
125 | A file starts with a big ~;;;~ comment bearing the Emacs ~-*-lisp-*-~ | |
126 | marker, a quick description, and copyright and licensing boilerplate. I | |
127 | don't use four-semicolon comments, and I only use ~#|~ ... ~|#~ for | |
128 | special effects. | |
129 | ||
130 | Then there's package stuff. There may be a ~cl:defpackage~ form (with | |
131 | explicit package qualifier) if the relevant package doesn't have its own | |
132 | package definition file. | |
133 | ||
134 | Then there's ~cl:in-package~. Like ~defpackage~, I use a gensym to name | |
135 | the package. I can't think offhand of a good reason to have a file with | |
7e55d099 MW |
136 | sections `in' more than one package. So, the ~in-package~ form goes at |
137 | the top of the file, before the first section header. If sections are | |
138 | going to end up in separate packages, I think I'd put a ~cl:in-package~ | |
139 | at the top of each section in case I wanted to reorder them. | |
1528431b MW |
140 | |
141 | The rest of the file consists of Lisp code. I don't use page boundaries | |
142 | ~^L~ to split files up. Instead, I use big banner comments for this: | |
143 | ||
144 | : ;;;-------------------------------------------------------------------------- | |
145 | : ;;; Section title. | |
146 | ||
147 | Sections don't usually have internal comments, but if they did they'd | |
148 | also be ~;;;~ comments. | |
149 | ||
150 | Almost all definitions get documentation strings. I've tried to be | |
151 | consistent about formatting. | |
152 | ||
153 | + Docstring lines are 77 characters or less. | |
154 | ||
155 | + The first line gives a summary of what the thing does. The summary, | |
156 | together with the SLIME-generated synopsis, is likely enough to | |
157 | remind you what the thing does. | |
158 | ||
159 | + The rest of the lines are indented by three spaces, and explain | |
160 | carefully what the thing does and what all the parameters mean. | |
161 | ||
162 | Smallish functions and macros don't usually need any further | |
163 | commentary. Big functions often need to be split into bitesize pieces | |
164 | with their own internal ~;;~ comments. The idea is that these comments | |
165 | should explain the code's overall strategy to the reader, and help them | |
166 | figure out how a piece fits into that strategy. | |
167 | ||
168 | Winged, single ~;~ comments are very rare. | |
169 | ||
170 | Files end, as a result of long tradition, with a comment | |
171 | ||
172 | : ;;;----- That's all, folks -------------------------------------------------- | |
173 | ||
174 | ||
175 | * Macro style | |
176 | ||
177 | I don't mind complicated macros if they're doing something worthwhile. | |
178 | They need to have good documentation strings, though. | |
179 | ||
180 | That said, where possible I've tried to factor macros into an actual | |
181 | macro providing the syntactic sugar, and a function which receives the | |
182 | parameters and $\eta$-expanded forms, and does the actual work. | |
183 | ||
184 | It's extremely bad taste for a macro to evaluate its evaluable | |
185 | parameters in any order other than strictly left to right, or to | |
186 | evaluate them more than once. | |
187 | ||
188 | ||
189 | * Data structures | |
190 | ||
191 | I've tended to be happy with plain lists for homogeneous-ish | |
192 | collections. Strongly heterogeneous collections (other than input | |
193 | syntax, destructured using ~defmacro~ or ~destructuring-bind~) I've | |
194 | tended to make a proper data type for. | |
195 | ||
196 | My first instinct when defining a new structure is to use ~defclass~. | |
197 | While it's annoyingly verbose, it has the immense benefit over | |
198 | ~defstruct~ that it's safe to redefine CLOS classes in a running image | |
199 | without the world breaking, and I usually find it necessary to add or | |
200 | change slots while I'm working on new code. Once a piece of code has | |
201 | settled down and I have a good feel for what my structure is actually | |
202 | doing, I might switch the ~defclass~ for a ~defstruct~. Several | |
203 | questions influence my decision. | |
204 | ||
205 | + Do slot accesses need to be really fast? My usual Lisp | |
206 | implementations aggressively optimize ~defstruct~ accessor | |
207 | functions. | |
208 | ||
7e55d099 | 209 | + Have I subclassed my class? While I can move over a |
1528431b MW |
210 | single-inheritance tree using ~:include~, it seems wrong to do this |
211 | most of the time. Also, I'd be precluding subclasses from multiple | |
212 | inheritance, and I'd either have to prohibit subclassing by | |
213 | extensions or have to commit to ~defstruct~ in the documentation. | |
214 | In general, I'm much happier committing to ~defclass~. | |
215 | ||
216 | + Are there methods specialized on my class? Again, structure classes | |
217 | make fine method specializers, but it doesn't seem right. | |
218 | ||
219 | Apart from being hard to redefine, ~defstruct~ does a pretty good job of | |
220 | making a new structure type. I tend to tidy up a few rough edges. | |
221 | ||
222 | + The default predicate always has ~-p~ appended. If the class name | |
223 | is a single word, then I'll explicitly name the predicate with a | |
224 | simple ~p~ suffix. For example, ~ship~ would have the predicate | |
a51bf71a | 225 | ~shipp~, rather than ~ship-p~. |
1528431b MW |
226 | |
227 | + If there are slots I can't default then I'll usually provide a BOA | |
228 | constructor which sets them from required parameters; other slots | |
229 | I'll set from optional or keyword parameters according to my taste | |
230 | and judgement. | |
231 | ||
232 | + Slots mustn't be given names which are external in any package. | |
233 | Unfortunately, slot names are used in constructing accessor names, | |
234 | and sometimes the right accessor name involves a prohibited symbol. | |
235 | I've mostly addressed this by naming the slot ~%foo~, and then | |
236 | providing inline reader and writer functions. (CLOS class | |
237 | definitions don't have this problem because you get to set the | |
238 | accessor function names independently of the slot names.) | |
239 | ||
240 | + BOA constructors are strange. You can set the initial slots based | |
241 | on an arbitrary computation on the provided parameters, but you have | |
242 | to roll up your sleeves and mess with ~&aux~ parameters to pull it | |
243 | off. | |
244 | ||
245 | ||
246 | * Naming | |
247 | ||
248 | I'm a traditionalist in some ways, and one of the reasons I like Lisp is | |
249 | the richness of its history and tradition. | |
250 | ||
251 | In other languages, I tend to use single- or two-letter names for | |
252 | variables and structure slots; not so much in Lisp. Other languages | |
253 | express more using punctuation, so the names stand out easily; I find | |
254 | that short names can be lost more easily in Lisp. | |
255 | ||
256 | I've also tended to go for fairly prosaic names, taking my inspiration | |
257 | from the CLOS MOP. While I mourn the loss of whimsical names like | |
258 | ~haulong~ and ~haipart~, I've tried to avoid inventing more of them. | |
259 | ||
260 | There's a convention, which I think comes from ML, of using ~_~ in a | |
261 | where a binding occurrence of a variable name is expected, to signify | |
262 | that that the corresponding value is to be discarded. Common Lisp, | |
263 | alas, doesn't have such a convention. Instead, there's a sequence of | |
264 | silly names used with the same intention, and the bindings are then | |
265 | explicitly ignored with a declaration. The names begin ~hunoz~, | |
266 | ~hukairz~, and (I think) ~huaskt~. | |
267 | ||
268 | ||
269 | * Declarations | |
270 | ||
271 | The code is light on declarations, other than ~ignore~ and similar used | |
272 | to muffle warnings. The macros try to do sensible things with | |
273 | declarations, and I think they succeed fairly well, but there might be | |
274 | bugs and rough edges. I know that some are just broken because, for | |
275 | actual correctness, declarations provided by the caller need to be split | |
276 | up into a number of different parts of the expansion, which in turn | |
277 | requires figuring out what the declarations mean and which bindings | |
278 | they're referring to. That's not completely impossible, assuming that | |
279 | there aren't implementation-specific declarations which crazy syntax | |
280 | mixed in there, but it's more work than seems worthwhile. | |
281 | ||
282 | ||
283 | * COMMENT Emacs cruft | |
284 | ||
285 | #+LATEX_CLASS: strayman | |
286 | ||
287 | ## LocalWords: CLOS ish destructure destructured accessor specializers | |
288 | ## LocalWords: accessors DSLs gensym | |
289 | ||
290 | ## Local variables: | |
291 | ## mode: org | |
292 | ## End: |