lvc@danews.ATT.COM (Larry Cipriani) (01/03/87)
In article <4900@mimsy.UUCP>, mangoe@mimsy.UUCP (Charley Wingate) writes: >...It >seems to me that there's every advantage in the world to having the computer >catch stupid and careless errors wherever possible; I utterly fail to see >the advantage in deliberately maintaining a syntactic confusion between >mistakes and a "feature", especially when making the latter explicit makes >it vastly more obvious what the code really does. > > CGW I largely agree with this however consider that C has about 46 operators and 15 precedence levels. This complexity can lead to subtle bugs. The reason these 'bugs' slip by cc are that they are allowed in the language and *can be intentional*. The worse compiler is one that complains about something that isn't wrong. Perhaps lint will be changed - I'm trying to get it changed but until then you'll have to write a separate program to find them (I already have). Sorry, I can't give you a copy or tell you much more about it. If I get something published on it, I'll talk some more about it. -- Larry Cipriani AT&T Network Systems cbosgd!{danews,cbsck}!lvc
mangoe@mimsy.UUCP (Charley Wingate) (01/06/87)
Larry Cipriani writes: >I largely agree with this however consider that C has about 46 >operators and 15 precedence levels. This complexity can lead to >subtle bugs. The reason these 'bugs' slip by cc are that they are >allowed in the language and *can be intentional*. The worse compiler >is one that complains about something that isn't wrong. Perhaps lint >will be changed - I'm trying to get it changed but until then you'll >have to write a separate program to find them (I already have). Sorry, >I can't give you a copy or tell you much more about it. If I get >something published on it, I'll talk some more about it. OK. Now let's eliminate all these "features" that are just like "bugs". The probable requirement of VARARGS in the ANSI standard is a step in the right direction. (actually, I don't want the features eliminated. I want them to be made explicit.) C. Wingate