gwyn@BRL.ARPA (VLD/VMB) (05/27/88)
I previously stated that I didn't think trigraphs broke much existing code. This was challenged, with reference to UNIX source code (which is AT&T's problem, and they bought into this). I decided to search my branch's existing C application sources, with the following results: INSTALLED APPLICATIONS: 298 source files 145,540 source lines 3 lines containing ??, in comments, unaffected by trigraphs. APPLICATION UNDER DEVELOPMENT: 124 source files 36,393 source lines 11 lines containing ??, in comments, unaffected by trigraphs. (Obviously there are more unresolved questions in the code under development.) No special care was taken when writing this code to worry about trigraphs; also, several people contributed to this code. This supports my claim about lack of code breakage by trigraphs.
tanner@ki4pv.UUCP (05/31/88)
In article <14830@brl-adm.ARPA>, gwyn@BRL.ARPA (VLD/VMB) writes:
) [ stats of his current progs; no non-comment instances of "??" ] ....
) This supports my claim about lack of code breakage by trigraphs.
For contrast, I just processed the body of the most important prog
in our major application.
total 97 files 25758 lines
"??" 4 files 6 lines
All of the "??" instances were in strings to be output in case of
exception conditions.
--
{allegra codas killer decvax!ucf-cs}!ki4pv!tannergwyn@brl-smoke.ARPA (Doug Gwyn ) (06/01/88)
In article <6944@ki4pv.uucp> tanner@ki4pv.UUCP writes: > total 97 files 25758 lines > "??" 4 files 6 lines >All of the "??" instances were in strings to be output in case of >exception conditions. So, would any of them be adversely impacted by the trigraph mappings? For those of you who haven't bothered to find out what trigraphs really are, only the following sequences are mapped as trigraphs: ??= ??( ??/ ??) ??' ??< ??! ??> ??- Other occurrences of ?? are not affected.
jfh@rpp386.UUCP (John F. Haugh II) (06/04/88)
I didn't run SCCS-gets on all of the files, so this includes all of the past versions as well as the current. Out of 245 files covering about 45,000 lines of source, there were no ??'s. And who uses two question marks in a string or whatever anyway?? - John.
g-rh@cca.CCA.COM (Richard Harter) (06/04/88)
>And who uses two question marks in a string or whatever anyway??
Actually, we have one, for a very obvious reason. In a help program
? gives you a simple summary of commands, including the information that
?? gives you a complete listing of topics. The list of predefined tokens
contains "??".
--
In the fields of Hell where the grass grows high
Are the graves of dreams allowed to die.
Richard Harter, SMDS Inc.tanner@ki4pv.UUCP (06/04/88)
In article <8003@brl-smoke.ARPA>, gwyn@brl-smoke.ARPA (Doug Gwyn ) writes: ) [cites my stats on "??" presence] [includes list of trigraphs] > So, would any of them be adversely impacted by the trigraph mappings? Well, now that you mention it, yes. Some of these WOULD be replaced by a trigraph-swallowing compiler. -- {allegra codas killer decvax!ucf-cs}!ki4pv!tanner