![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to error.1 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Power 6/32 Unix Tahoe 4.2BSD] / cci / usr / src / man / man1|
Return to error.1 CVS log | Up to [Power 6/32 Unix Tahoe 4.2BSD] / cci / usr / src / man / man1 |
1.1 root 1: .TH ERROR 1 "18 January 1983"
2: .UC 4
3: .SH NAME
4: error \- analyze and disperse compiler error messages
5: .SH SYNOPSIS
6: .B error
7: [
8: .B \-n
9: ] [
10: .B \-s
11: ] [
12: .B \-q
13: ] [
14: .B \-v
15: ] [
16: .B \-t
17: suffixlist
18: ] [
19: .B \-I
20: ignorefile
21: ] [ name ]
22: .SH DESCRIPTION
23: .I Error
24: analyzes and optionally disperses the diagnostic error messages
25: produced by a number of compilers and language processors to the source
26: file and line where the errors occurred. It can replace the painful,
27: traditional methods of scribbling abbreviations of errors on paper, and
28: permits error messages and source code to be viewed simultaneously
29: without machinations of multiple windows in a screen editor.
30: .PP
31: .I Error
32: looks at the error messages,
33: either from the specified file \fIname\fR
34: or from the standard input,
35: and attempts to determine which
36: language processor produced each error message,
37: determines the source file and line number to which the error message refers,
38: determines if the error message is to be ignored or not,
39: and inserts the (possibly slightly modified) error message into
40: the source file as a comment on the line preceding to which the
41: line the error message refers.
42: Error messages which can't be categorized by language processor
43: or content are not inserted into any file,
44: but are sent to the standard output.
45: .I Error
46: touches source files only after all input has been read.
47: By specifying the
48: .B \-q
49: query option,
50: the user is asked to confirm any potentially
51: dangerous (such as touching a file) or verbose action.
52: Otherwise
53: .I error
54: proceeds on its merry business.
55: If the
56: .B \-t
57: touch option and associated suffix list is given,
58: .I error
59: will restrict itself to touch only those files with suffices
60: in the suffix list.
61: Error also can be asked (by specifying
62: .B \-v)
63: to invoke
64: .IR vi (1)
65: on the files in which error messages were inserted; this obviates
66: the need to remember the names of the files with errors.
67: .PP
68: .I Error
69: is intended to be run
70: with its standard input
71: connected via a pipe to the error message source.
72: Some language processors put error messages on their standard error file;
73: others put their messages on the standard output.
74: Hence, both error sources should be piped together into
75: .I error.
76: For example, when using the \fIcsh\fP syntax,
77: .IP
78: make \-s lint |\|& error \-q \-v
79: .LP
80: will analyze all the error messages produced
81: by whatever programs
82: .I make
83: runs when making lint.
84: .PP
85: .I Error
86: knows about the error messages produced by:
87: .I make,
88: .I cc,
89: .I cpp,
90: .I ccom,
91: .I as,
92: .I ld,
93: .I lint,
94: .I pi,
95: .I pc
96: and
97: .I f77.
98: .I Error
99: knows a standard format for error messages produced by
100: the language processors,
101: so is sensitive to changes in these formats.
102: For all languages except
103: .I Pascal,
104: error messages are restricted to be on one line.
105: Some error messages refer to more than one line in more than
106: one files;
107: .I error
108: will duplicate the error message and insert it at
109: all of the places referenced.
110: .PP
111: .I Error
112: will do one of six things with error messages.
113: .TP 10
114: .I synchronize
115: Some language processors produce short errors describing
116: which file it is processing.
117: .I Error
118: uses these to determine the file name for languages that
119: don't include the file name in each error message.
120: These synchronization messages are consumed entirely by
121: .I error.
122: .TP 10
123: .I discard
124: Error messages from
125: .I lint
126: that refer to one of the two
127: .I lint
128: libraries,
129: .I /usr/lib/llib-lc
130: and
131: .I /usr/lib/llib-port
132: are discarded,
133: to prevent accidently touching these libraries.
134: Again, these error messages are consumed entirely by
135: .I error.
136: .TP 10
137: .I nullify
138: Error messages from
139: .I lint
140: can be nullified if they refer to a specific function,
141: which is known to generate diagnostics which are not interesting.
142: Nullified error messages are not inserted into the source file,
143: but are written to the standard output.
144: The names of functions to ignore are taken from
145: either the file named
146: .I .errorrc
147: in the users's home directory,
148: or from the file named by the
149: .B \-I
150: option.
151: If the file does not exist,
152: no error messages are nullified.
153: If the file does exist, there must be one function
154: name per line.
155: .TP 10
156: .I not file specific
157: Error messages that can't be intuited are grouped together,
158: and written to the standard output before any files are touched.
159: They will not be inserted into any source file.
160: .TP 10
161: .I file specific
162: Error message that refer to a specific file,
163: but to no specific line,
164: are written to the standard output when
165: that file is touched.
166: .TP 10
167: .I true errors
168: Error messages that can be intuited are candidates for
169: insertion into the file to which they refer.
170: .PP
171: Only true error messages are candidates for inserting into
172: the file they refer to.
173: Other error messages are consumed entirely by
174: .I error
175: or are written to the standard output.
176: .I Error
177: inserts the error messages into the source file on the line
178: preceding the line the language processor found in error.
179: Each error message is turned into a one line comment for the
180: language,
181: and is internally flagged
182: with the string ``###'' at
183: the beginning of the error,
184: and ``%%%'' at the end of the error.
185: This makes pattern searching for errors easier with an editor,
186: and allows the messages to be easily removed.
187: In addition, each error message contains the source line number
188: for the line the message refers to.
189: A reasonably formatted source program can be recompiled
190: with the error messages still in it,
191: without having the error messages themselves cause future errors.
192: For poorly formatted source programs in free format languages,
193: such as C or Pascal,
194: it is possible to insert a comment into another comment,
195: which can wreak havoc with a future compilation.
196: To avoid this, programs with comments and source
197: on the same line should be formatted
198: so that language statements appear before comments.
199: .PP
200: Options available with
201: .I error
202: are:
203: .TP 5
204: .B \-n
205: Do
206: .I not
207: touch any files; all error messages are sent to the
208: standard output.
209: .TP 5
210: .B \-q
211: The user is
212: .I queried
213: whether s/he wants to touch the file.
214: A ``y'' or ``n'' to the question is necessary to continue.
215: Absence of the
216: .B \-q
217: option implies that all referenced files
218: (except those referring to discarded error messages)
219: are to be touched.
220: .TP 5
221: .B \-v
222: After all files have been touched,
223: overlay the visual editor
224: .I vi
225: with it set up to edit all files touched,
226: and positioned in the first touched file at the first error.
227: If
228: .I vi
229: can't be found, try
230: .I ex
231: or
232: .I ed
233: from standard places.
234: .TP 5
235: .B \-t
236: Take the following argument as a suffix list.
237: Files whose suffixes do not appear in the suffix list are not touched.
238: The suffix list is dot separated, and ``*'' wildcards work.
239: Thus the suffix list:
240: .IP
241: \& ".c.y.foo*.h"
242: .IP
243: allows
244: .I error
245: to touch files ending with ``.c'', ``.y'', ``.foo*'' and ``.y''.
246: .TP 5
247: .B \-s
248: Print out
249: .I statistics
250: regarding the error categorization.
251: Not too useful.
252: .PP
253: .I Error
254: catches interrupt and terminate signals,
255: and if in the insertion phase,
1.1.1.2 ! root 256: terminates what it is doing in an orderly way.
1.1 root 257: .SH AUTHOR
258: Robert Henry
259: .SH FILES
260: .ta 2i
261: ~/.errorrc function names to ignore for \fIlint\fP error messages
262: .br
263: /dev/tty user's teletype
264: .SH BUGS
265: .PP
266: Opens the teletype directly to do user querying.
267: .PP
268: Source files with links make a new copy of the file with
269: only one link to it.
270: .PP
271: Changing a language processor's format of error messages
272: may cause
273: .I error
274: to not understand the error message.
275: .PP
276: .I Error,
277: since it is purely mechanical,
278: will not filter out subsequent errors caused by `floodgating'
279: initiated by one syntactically trivial error.
280: .PP
281: Pascal error messages belong after the lines affected
282: (error puts them before). The alignment of the `\||\|' marking
283: the point of error is also disturbed by
284: .I error.
285: .PP
286: .I Error
287: was designed for work on CRT's at reasonably high speed.
288: It is less pleasant on slow speed terminals, and has never been
289: used on hardcopy terminals.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.