shodsdon@elrond.CalComp.COM (Steve V. Hodsdon) (07/20/90)
I just finished porting a software package that ran on a VAX 11/780
to a Sun SPARCserver. One of the programs would not run. (Could not find
the filename specified.). The program was looking for a certain extension,
but could not find it.
Finally, it dawned on me that the function call tolower was to blame.
On the VAX, the call did nothing if the char was already lowercase, while
the Sun converted it anyway.
I know, I know... RTFM. The FM does say this also.
What a pain in the ...
I wonder what other little gotchas there are...
--
Steve Hodsdon | Stay sane inside insanity.
shodsdon@elrond.CalComp.COM | Don't dream it, be it.
{decvax|harvard}!elrond!shodsdon |
(603) 885-8324 |diamond@tkou02.enet.dec.com (diamond@tkovoa) (07/23/90)
In article <2841@elrond.CalComp.COM> shodsdon@elrond.CalComp.COM (Steve V. Hodsdon) writes: >I just finished porting a software package that ran on a VAX 11/780 >to a Sun SPARCserver. One of the programs would not run. >Finally, it dawned on me that the function call tolower was to blame. >On the VAX, the call did nothing if the char was already lowercase, while >the Sun converted it anyway. >I know, I know... RTFM. The FM does say this also. Huh? File a bug report. Tell Sun to RTFS. It's nice to see that the VAX compiler conforms more closely to the standard than Sun does. -- Norman Diamond, Nihon DEC diamond@tkou02.enet.dec.com This is me speaking. If you want to hear the company speak, you need DECtalk.
bruce@seismo.gps.caltech.edu (Bruce Worden) (07/23/90)
In article <1863@tkou02.enet.dec.com> diamond@tkou02.enet.dec.com (diamond@tkovoa) writes: >In article <2841@elrond.CalComp.COM> shodsdon@elrond.CalComp.COM (Steve V. Hodsdon) writes: > >>I just finished porting a software package that ran on a VAX 11/780 >>to a Sun SPARCserver. One of the programs would not run. >>Finally, it dawned on me that the function call tolower was to blame. >>On the VAX, the call did nothing if the char was already lowercase, while >>the Sun converted it anyway. >>I know, I know... RTFM. The FM does say this also. >Huh? File a bug report. Tell Sun to RTFS. No don't file a bug report. The tolower() function (it is a function now) works as per the ANSI standard under SunOS 4.1. It did have the described behavior up to and including SunOS 4.0.3. I do believe however that, in the pre-ANSI standard version of K&R, it is not quite clear what behavior tolower() should have when the argument is already lower case. I can't find an explicit statement. Do note, however, that the example program given checks the case of the argument before calling tolower(). In the absence of an explicit statement Sun may have opted for the simplest implementation, if not the most benign. -------------------------------------------------------------------------- Bruce Worden bruce@seismo.gps.caltech.edu Seismological Laboratory Caltech --------------------------------------------------------------------------
boylanr@silver.ucs.indiana.edu (ross boylan) (07/24/90)
diamond@tkou02.enet.dec.com (diamond@tkovoa) writes: >In article <2841@elrond.CalComp.COM> shodsdon@elrond.CalComp.COM (Steve V. Hodsdon) writes: >>I just finished porting a software package that ran on a VAX 11/780 >>to a Sun SPARCserver. One of the programs would not run. >>Finally, it dawned on me that the function call tolower was to blame. >>On the VAX, the call did nothing if the char was already lowercase, while >>the Sun converted it anyway. >>I know, I know... RTFM. The FM does say this also. >Huh? File a bug report. Tell Sun to RTFS. I recall having a similar problem with tolower, but I think it's a "feature" not a bug. On some systems tolower is only defined for upper case laters--that is, it will do strange things to non-upper case letters. May be a Unix Sys V v. BSD difference. The world would
wht@n4hgf.Mt-Park.GA.US (Warren Tucker) (07/24/90)
>>>Finally, it dawned on me that the function call tolower was to blame. >>>On the VAX, the call did nothing if the char was already lowercase, while >>>the Sun converted it anyway. System 5-derived tolower() absolutely coverts, while BSD does the equivalent if isupper(ch) convert ANSI aside, I'll always code if(isupper(ch)) ch = tolower(ch); to be safe. ---------------------------------------------------------------------- Warren Tucker, TuckerWare emory!n4hgf!wht or wht@n4hgf.Mt-Park.GA.US "It was electrons that brought down the Berlin Wall." -- Timothy Leary
johnb@srchtec.UUCP (John Baldwin) (07/24/90)
In article <1863@tkou02.enet.dec.com> diamond@tkou02.enet.dec.com (diamond@tkovoa) writes: > >... Tell Sun to RTFS. ^^^^ I LOVE IT! Great new acronym! -- John T. Baldwin | Disclaimer: search technology, inc. | Some people claim I never existed. Norcross, Georgia | (real .sig under construction johnb@srchtec.uucp | at Starfleet Orbital Navy Yards ;-)
merriman@ccavax.camb.com (07/24/90)
In article <2841@elrond.CalComp.COM>, shodsdon@elrond.CalComp.COM (Steve V. Hodsdon) writes: > I just finished porting a software package that ran on a VAX 11/780 > to a Sun SPARCserver. One of the programs would not run. (Could not find > the filename specified.). The program was looking for a certain extension, > but could not find it. > > Finally, it dawned on me that the function call tolower was to blame. > On the VAX, the call did nothing if the char was already lowercase, while > the Sun converted it anyway. Converted it to what?
henry@zoo.toronto.edu (Henry Spencer) (07/24/90)
In article <1990Jul23.040004.12292@laguna.ccsf.caltech.edu> bruce@seismo.gps.caltech.edu (Bruce Worden) writes: >... in the pre-ANSI standard version of K&R, it is not >quite clear what behavior tolower() should have when the argument >is already lower case... The only explicit statement you will find about tolower()'s behavior in the early days was that given an uppercase letter, it produced the corresponding lowercase letter. Its behavior on any other input was in fact undefined. -- NFS: all the nice semantics of MSDOS, | Henry Spencer at U of Toronto Zoology and its performance and security too. | henry@zoo.toronto.edu utzoo!henry
shodsdon@elrond.CalComp.COM (Steve V. Hodsdon) (07/25/90)
In article <174@n4hgf.Mt-Park.GA.US> wht@n4hgf.UUCP (Warren Tucker) writes: > >System 5-derived tolower() absolutely coverts, while BSD ^^^^^^^^^^^^^^^^^^ Just aother little gotcha: Motorola's sysV68 R3.5 checks first before converting to lower (or upper). There is the macro's _tolower and _toupper which assume the 'proper case'. >---------------------------------------------------------------------- >Warren Tucker, TuckerWare emory!n4hgf!wht or wht@n4hgf.Mt-Park.GA.US >"It was electrons that brought down the Berlin Wall." -- Timothy Leary -- Steve Hodsdon | Stay sane inside insanity. shodsdon@elrond.CalComp.COM | Don't dream it - Be it. {decvax|harvard}!elrond!shodsdon | (603) 885-8324 |