I'm in the process converting interpreted code to YottaDB, and it is great that object files are still produced, even if the compiler finds a syntax anomaly.

However, I've upgraded from version 1.34 to 1.38 and now I get compilation errors when syntax abnormalities are encountered. It seems like the compile -ignore flag is not being honoured. Is this the case, or have I missed something after doing the
upgrade?

Any help would be appreciated,

Thanks,

Ray.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Tuesday, November 14, 2023 at 10:08:45 AM UTC-5, stafford raymond wrote:

I'm in the process converting interpreted code to YottaDB, and it is great that object files are still produced, even if the compiler finds a syntax anomaly.

However, I've upgraded from version 1.34 to 1.38 and now I get compilation errors when syntax abnormalities are encountered. It seems like the compile -ignore flag is not being honoured. Is this the case, or have I missed something after doing the

upgrade?

Any help would be appreciated,

Thanks,

Ray.

Ray, an example or two of the behavior would help. Thanks.

Regards
– Bhaskar

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Tuesday, November 14, 2023 at 4:19:47 PM UTC-5, stafford raymond wrote:

On Tuesday, November 14, 2023 at 4:00:02 PM UTC, K.S. Bhaskar wrote:

On Tuesday, November 14, 2023 at 10:08:45 AM UTC-5, stafford raymond wrote:

I'm in the process converting interpreted code to YottaDB, and it is great that object files are still produced, even if the compiler finds a syntax anomaly.

However, I've upgraded from version 1.34 to 1.38 and now I get compilation errors when syntax abnormalities are encountered. It seems like the compile -ignore flag is not being honoured. Is this the case, or have I missed something after doing the

upgrade?

Any help would be appreciated,

Thanks,

Ray.

Ray, an example or two of the behavior would help. Thanks.

Regards
– Bhaskar

Hi Bhaskar,

Thanks for the reply.

It looks like I've solved my issue, which related to spurious characters in the source code I'm converting.
Normally, when the object files were complied, some errors were shown during compilation, but the resulting object code ran fine.
After the upgrade, errors were also occurring at run-time. I've managed to isolate these characters, and everything compiles and runs OK again.

Maybe these were not detected previously due to the character sets I was using. I was running version 1.34 on Debian, where, because of an "ICU version not found error", I had set ydb_chset=M in order to gain access and had no issues compiling. The

upgrade to 1.38 fixed the "ICU version not found error" and I've now made the character set consistent, as UTF-8, so hopefully I'm sorted.

Regards,

Ray

Thanks for closing the loop. I'm glad the issue is solved.

Regards
– Bhaskar

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Tuesday, November 14, 2023 at 4:00:02 PM UTC, K.S. Bhaskar wrote:

On Tuesday, November 14, 2023 at 10:08:45 AM UTC-5, stafford raymond wrote:

I'm in the process converting interpreted code to YottaDB, and it is great that object files are still produced, even if the compiler finds a syntax anomaly.

However, I've upgraded from version 1.34 to 1.38 and now I get compilation errors when syntax abnormalities are encountered. It seems like the compile -ignore flag is not being honoured. Is this the case, or have I missed something after doing the

upgrade?

Any help would be appreciated,

Thanks,

Ray.

Ray, an example or two of the behavior would help. Thanks.

Regards
– Bhaskar

Hi Bhaskar,

Thanks for the reply.

It looks like I've solved my issue, which related to spurious characters in the source code I'm converting.
Normally, when the object files were complied, some errors were shown during compilation, but the resulting object code ran fine.
After the upgrade, errors were also occurring at run-time. I've managed to isolate these characters, and everything compiles and runs OK again.

Maybe these were not detected previously due to the character sets I was using. I was running version 1.34 on Debian, where, because of an "ICU version not found error", I had set ydb_chset=M in order to gain access and had no issues compiling. The
upgrade to 1.38 fixed the "ICU version not found error" and I've now made the character set consistent, as UTF-8, so hopefully I'm sorted.

Regards,

Ray

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)