View previous topic :: View next topic |
Author |
Message |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
SOLVED: #IGNORE_WARNINGS to ignore #include source files? |
Posted: Fri Apr 23, 2021 9:44 am |
|
|
The Wiznet library has a bunch of warnings in it, and I wanted to quiet them down by doing something like this:
Code: | #IGNORE_WARNINGS 203 // Condition always TRUE
#IGNORE_WARNINGS 204 // Condition always FALSE
#IGNORE_WARNINGS 209 // Assignment to enum is not of the correct type
#IGNORE_WARNINGS 214 // Operator precedance rules may not be as intended, use () to clarify
#include "socket.c"
#include "wizchip_conf.c"
#include "wizchip_conf.h"
#include "w5500/w5500.c"
#include "w5500/w5500.h"
#IGNORE_WARNINGS NONE
|
This does not seem to work. Ideally, we would not want to touch the original source so it can be updated without extra work each time.
Thoughts? _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ?
Last edited by allenhuffman on Fri Apr 30, 2021 1:12 pm; edited 1 time in total |
|
![](templates/subSilver/images/spacer.gif) |
temtronic
Joined: 01 Jul 2010 Posts: 9174 Location: Greensville,Ontario
|
|
Posted: Fri Apr 23, 2021 12:26 pm |
|
|
hmmm.. does ignorewarnings none defeat the previous commands ? |
|
![](templates/subSilver/images/spacer.gif) |
PCM programmer
Joined: 06 Sep 2003 Posts: 21708
|
|
Posted: Fri Apr 23, 2021 1:35 pm |
|
|
It works for me. I tried all combinations of the #ignore_warnings
lines below (commented out or not) and it works as desired.
This was with CCS vs. 5.103
Code: |
#include <18F46K22.h>
#use delay(internal=4M)
#IGNORE_WARNINGS 203 // Disable Condition always True warning
#include "test.h"
#IGNORE_WARNINGS NONE
//=================================
void main()
{
test_func();
while(1);
} |
test.h file:
Code: |
void test_func()
{
while(1);
} |
|
|
![](templates/subSilver/images/spacer.gif) |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
|
Posted: Fri Apr 23, 2021 9:13 pm |
|
|
temtronic wrote: | hmmm.. does ignorewarnings none defeat the previous commands ? |
You can turn it on and off that way. We always try to have zero warnings, and use this for places we want to ignore. Like this common one:
Code: |
Switch (x)
{
Case 1:
Function();
Break;
Case 2:
Default:
DefaultFunction();
Break;
}
|
Sorry about the caps. Typing in the tablet.
CCS will give a "missing break" warning which is a good warning in case someone truly did mean those to be two different case statements, but you work around it by muting the warning just around that bit of code:
Code: |
#ignore_warnings 204
Case 2:
Default:
DefaultFunction();
Break;
#ignore_warnings none
|
...or whatever that specific missing break warning is. We use that often when we have a default case like that.
Maybe the issue is including a .c file rather than a .h and if parses it differently. I’ll experiment next week when I am back in the office. _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ? |
|
![](templates/subSilver/images/spacer.gif) |
dyeatman
Joined: 06 Sep 2003 Posts: 1924 Location: Norman, OK
|
|
Posted: Sat Apr 24, 2021 5:11 am |
|
|
FWIW, when I have empty cases back to back I put a pair of empty braces after each case and then I don't get the warnings.
Code: | switch (testit)
{
case 1: {};
case 2: {};
case 3: {};
default: testit=testit+1;
} |
_________________ Google and Forum Search are some of your best tools!!!! |
|
![](templates/subSilver/images/spacer.gif) |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
|
Posted: Sat Apr 24, 2021 6:18 am |
|
|
dyeatman wrote: | FWIW, when I have empty cases back to back I put a pair of empty braces after each case and then I don't get the warnings.
Code: | switch (testit)
{
case 1: {};
case 2: {};
case 3: {};
default: testit=testit+1;
} |
|
Interesting. That must be a CCS thing, just like while(1) is a warning but using their while(TRUE) is not. Since it definitely is still missing a break. _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ? |
|
![](templates/subSilver/images/spacer.gif) |
Ttelmah
Joined: 11 Mar 2010 Posts: 19369
|
|
Posted: Sat Apr 24, 2021 7:24 am |
|
|
On the break, why not use the nobreak define given by CCS in their
readme, to avoid this?. |
|
![](templates/subSilver/images/spacer.gif) |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
|
Posted: Fri Apr 30, 2021 12:23 pm |
|
|
Ttelmah wrote: | On the break, why not use the nobreak define given by CCS in their
readme, to avoid this?. |
Cross-compatibility with other compilers. At a previous job, two of the compilers (one was GCC) had a way to flag this, but each used a different syntax so it made for non-portable code. #defines were used to do all that.
I'm doing the same for "_readonly" in the CCS code so I can still build it externally. We tend to write out routines in ANSI-C on a PC compiler to get all the logic worked out, then do any necessary updates for PIC (like not being able to pass in a constant string by default).
If this works when #including a .h file, it apparently is turned off when doing it for including a .c :( _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ? |
|
![](templates/subSilver/images/spacer.gif) |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
|
Posted: Fri Apr 30, 2021 12:24 pm |
|
|
Ttelmah wrote: | On the break, why not use the nobreak define given by CCS in their
readme, to avoid this?. |
Where is this? I just did a search and did not find a hit for "nobreak" in the help file. It may be easier than #define the IGNORE_WARNING macro. _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ? |
|
![](templates/subSilver/images/spacer.gif) |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
|
Posted: Fri Apr 30, 2021 1:11 pm |
|
|
#IGNORE_WARNINGS 203,204,209,214,215,237
Solved: All one one line, with commas, not separate. _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ? |
|
![](templates/subSilver/images/spacer.gif) |
PCM programmer
Joined: 06 Sep 2003 Posts: 21708
|
|
Posted: Fri Apr 30, 2021 1:23 pm |
|
|
It's in the latest readme.txt file:
A new warning has been added to the compiler to indicate when a
case inside a switch does not have a break. For those who know
there is no break and don't need the warning you can do something
like this:
Code: | #define nobreak
...
case 4: handle_four();
nobreak;
case 5: handle_five();
break; |
|
|
![](templates/subSilver/images/spacer.gif) |
allenhuffman
Joined: 17 Jun 2019 Posts: 540 Location: Des Moines, Iowa, USA
|
|
Posted: Fri Apr 30, 2021 1:29 pm |
|
|
PCM programmer wrote: | It's in the latest readme.txt file: |
I just found that in the PICC program directory. I do not remember if I even realized there was a readme with the updates! Thanks! _________________ Allen C. Huffman, Sub-Etha Software (est. 1990) http://www.subethasoftware.com
Embedded C, Arduino, MSP430, ESP8266/32, BASIC Stamp and PIC24 programmer.
http://www.whywouldyouwanttodothat.com ? |
|
![](templates/subSilver/images/spacer.gif) |
Ttelmah
Joined: 11 Mar 2010 Posts: 19369
|
|
Posted: Fri Apr 30, 2021 10:18 pm |
|
|
Apologies for not seeing that you had asked where this was...
The readme.txt, just occasionally has some absolutely brilliant little bits
like this. Like you, I had hit the case with no break problem, and found
this. I do now make it a 'point' to look in the readme when a change
happens in the compiler. ![Very Happy](images/smiles/icon_biggrin.gif) |
|
![](templates/subSilver/images/spacer.gif) |
|