Re: [Firebird-devel] Preventing error code collision
On Sun, Jul 26, 2015 at 5:15 PM, Vlad Khorsun wrote: Or is there a reason to ignore those higher bits for the facility and code? I have no idea why ENCODE_ISC_MSG written in this way. CLASS_MASK seems to not be used anywhere, or at least I can't remember ever having seen an error code with the bit 30 (warning) or 31 (info) set. Or is it used somewhere internally as an in-band channel? Looks like something planned at the past (before Firebird) but never used... 27.07.2015 1:24, Ann Harrison wrote: Firebird was based on InterBase which was based on Rdb/ELN, an implementation of DEC's [standard(!)] relational interface. As part of DEC's VAX software empire, DSRI used DEC's error message facility. Every project had a code and used it as a prefix to its error messages. On Mon, Jul 27, 2015 at 4:14 AM, Vlad Khorsun hv...@users.sourceforge.net wrote: Ann, thanks for explanation. But, it is still not clear why ENCODE_ISC_MSG is more strict than necessary. It limits number of facilities by 31 (bitmask used allows 255) and number of codes per facility by 16383 (instead of 65535). Currently, we have no problem with it, just curious... The high bits in the codes may have been intended for severity ... I really don't remember. The high bits in the facility were to distinguish relational database errors from other DEC product errors. I'd feel free to use any of them. Cheers, Ann -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
On 7/27/2015 11:31 AM, Ann Harrison wrote: 27.07.2015 1:24, Ann Harrison wrote: Firebird was based on InterBase which was based on Rdb/ELN, an implementation of DEC's [standard(!)] relational interface. As part of DEC's VAX software empire, DSRI used DEC's error message facility. Every project had a code and used it as a prefix to its error messages. On Mon, Jul 27, 2015 at 4:14 AM, Vlad Khorsun hv...@users.sourceforge.net mailto:hv...@users.sourceforge.net wrote: Ann, thanks for explanation. But, it is still not clear why ENCODE_ISC_MSG is more strict than necessary. It limits number of facilities by 31 (bitmask used allows 255) and number of codes per facility by 16383 (instead of 65535). Currently, we have no problem with it, just curious... The high bits in the codes may have been intended for severity ... I really don't remember. The high bits in the facility were to distinguish relational database errors from other DEC product errors. I'd feel free to use any of them. The only VMS documentation I could find identifies the high order bits as internal control flags. A piece of almost forgotten history is that when Interbase first shipped, it had bi-directional interoperability of Rdb/VMS and Rdb/ELN. Something we liked to do was fire up Database on VMS and throw queries at Interbase on an Apollo. Like Firebird, the Rdb y-valve loaded subsystems dynamically, so we could slip our remote (and embedded) subsystems under the DEC y-valve. DEC didn't support TCP/IP (TCP/IP will blow over before we could do an implementation, which almost turned out to be true), so we had to use a third party TCP/IP. -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
27.07.2015 1:24, Ann Harrison wrote: On Sun, Jul 26, 2015 at 5:15 PM, Vlad Khorsun wrote: Or is there a reason to ignore those higher bits for the facility and code? I have no idea why ENCODE_ISC_MSG written in this way. CLASS_MASK seems to not be used anywhere, or at least I can't remember ever having seen an error code with the bit 30 (warning) or 31 (info) set. Or is it used somewhere internally as an in-band channel? Looks like something planned at the past (before Firebird) but never used... Firebird was based on InterBase which was based on Rdb/ELN, an implementation of DEC's [standard(!)] relational interface. As part of DEC's VAX software empire, DSRI used DEC's error message facility. Every project had a code and used it as a prefix to its error messages. Ann, thanks for explanation. But, it is still not clear why ENCODE_ISC_MSG is more strict than necessary. It limits number of facilities by 31 (bitmask used allows 255) and number of codes per facility by 16383 (instead of 65535). Currently, we have no problem with it, just curious... Regars, Vlad -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
On 25-7-2015 15:08, Vlad Khorsun wrote: 25.07.2015 15:45, Mark Rotteveel wrote: How would I go about that if none of the errors in that facility are defined inside Firebird, We could define and reserve facility code for Jaybird and let you know it. That would be great. and how can I know/calculate the range of error codes I can use (a quick glance at the code seems to suggest that each facility has 24 bits for error codes)? Look at src\common\msg_encode.h : const ISC_STATUS ISC_MASK = 0x1400; // Defines the code as a valid ISC code const ISC_STATUS FAC_MASK = 0x00FF; // Specifies the facility where the code is located const ISC_STATUS CODE_MASK= 0x; // Specifies the code in the message file const ISC_STATUS CLASS_MASK = 0xF000; // Defines the code as warning, error, info, or other Looking at the code in msg_encode.h, I notice that the above masks don't align with the definition of ENCODE_ISC_MSG: return ((ISC_STATUS(facility 0x1F) 16) | (code 0x3FFF) | ISC_MASK); Shouldn't that be: return ISC_STATUS(((facility 16) FAC_MASK) | (code CODE_MASK) | ISC_MASK); Or is there a reason to ignore those higher bits for the facility and code? CLASS_MASK seems to not be used anywhere, or at least I can't remember ever having seen an error code with the bit 30 (warning) or 31 (info) set. Or is it used somewhere internally as an in-band channel? Mark -- Mark Rotteveel -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
26.07.2015 12:08, Mark Rotteveel wrote: On 25-7-2015 15:08, Vlad Khorsun wrote: 25.07.2015 15:45, Mark Rotteveel wrote: How would I go about that if none of the errors in that facility are defined inside Firebird, We could define and reserve facility code for Jaybird and let you know it. That would be great. Facility code 26, name JAYBIRD. and how can I know/calculate the range of error codes I can use (a quick glance at the code seems to suggest that each facility has 24 bits for error codes)? Look at src\common\msg_encode.h : const ISC_STATUS ISC_MASK= 0x1400; // Defines the code as a valid ISC code const ISC_STATUS FAC_MASK= 0x00FF; // Specifies the facility where the code is located const ISC_STATUS CODE_MASK = 0x; // Specifies the code in the message file const ISC_STATUS CLASS_MASK = 0xF000; // Defines the code as warning, error, info, or other Looking at the code in msg_encode.h, I notice that the above masks don't align with the definition of ENCODE_ISC_MSG: return ((ISC_STATUS(facility 0x1F) 16) | (code 0x3FFF) | ISC_MASK); Shouldn't that be: return ISC_STATUS(((facility 16) FAC_MASK) | (code CODE_MASK) | ISC_MASK); Or is there a reason to ignore those higher bits for the facility and code? I have no idea why ENCODE_ISC_MSG written in this way. CLASS_MASK seems to not be used anywhere, or at least I can't remember ever having seen an error code with the bit 30 (warning) or 31 (info) set. Or is it used somewhere internally as an in-band channel? Looks like something planned at the past (before Firebird) but never used... Regards, Vlad -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
On Sun, Jul 26, 2015 at 5:15 PM, Vlad Khorsun hv...@users.sourceforge.net wrote: Or is there a reason to ignore those higher bits for the facility and code? I have no idea why ENCODE_ISC_MSG written in this way. CLASS_MASK seems to not be used anywhere, or at least I can't remember ever having seen an error code with the bit 30 (warning) or 31 (info) set. Or is it used somewhere internally as an in-band channel? Looks like something planned at the past (before Firebird) but never used... Firebird was based on InterBase which was based on Rdb/ELN, an implementation of DEC's [standard(!)] relational interface. As part of DEC's VAX software empire, DSRI used DEC's error message facility. Every project had a code and used it as a prefix to its error messages. Cheers, Ann -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
[Firebird-devel] Preventing error code collision
Up to now Jaybird has always used the Firebird error codes, or specific error messages defined in code. I am currently looking into defining some Jaybird specific error codes, with associated messages and SQL states. To prevent collisions with future error codes in Firebird, I'd like to know if using codes below 3 would be sufficient to prevent collision, or would it be better to follow the same formal rules as used inside Firebird to generate error codes (eg reserve a facility code for Jaybird)? Mark -- Mark Rotteveel -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
25.07.2015 14:57, Mark Rotteveel wrote: Up to now Jaybird has always used the Firebird error codes, or specific error messages defined in code. I am currently looking into defining some Jaybird specific error codes, with associated messages and SQL states. Are you going to report is using Firebird API ? Are you going to add Jaybird specific messages into firebird.msg ? To prevent collisions with future error codes in Firebird, I'd like to know if using codes below 3 would be sufficient to prevent collision, or would it be better to follow the same formal rules as used inside Firebird to generate error codes (eg reserve a facility code for Jaybird)? I think, facility is what you need. Regards, Vlad -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
25.07.2015 15:16, Vlad Khorsun wrote: To prevent collisions with future error codes in Firebird, I'd like to know if using codes below 3 would be sufficient to prevent collision, or would it be better to follow the same formal rules as used inside Firebird to generate error codes (eg reserve a facility code for Jaybird)? I think, facility is what you need. I second that. Dmitry -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
Re: [Firebird-devel] Preventing error code collision
On 25-7-2015 14:16, Vlad Khorsun wrote: 25.07.2015 14:57, Mark Rotteveel wrote: Up to now Jaybird has always used the Firebird error codes, or specific error messages defined in code. I am currently looking into defining some Jaybird specific error codes, with associated messages and SQL states. Are you going to report is using Firebird API ? Are you going to add Jaybird specific messages into firebird.msg ? Jaybird doesn't use firebird.msg, I occasionally export the error messages and sqlstates from a Firebird build to the isc_error_msg.properties and isc_error_sqlstates.properties of Jaybird (and the error code constants to ISCConstants.java). However, the Jaybird specific error codes, error messages and sqlstate mapping will be within Jaybird itself. I want to reuse (extend) the existing error message mechanism inside Jaybird to obtain the messages and sqlstate mappings from multiple sources and that means that I have to avoid collisions. To prevent collisions with future error codes in Firebird, I'd like to know if using codes below 3 would be sufficient to prevent collision, or would it be better to follow the same formal rules as used inside Firebird to generate error codes (eg reserve a facility code for Jaybird)? I think, facility is what you need. How would I go about that if none of the errors in that facility are defined inside Firebird, and how can I know/calculate the range of error codes I can use (a quick glance at the code seems to suggest that each facility has 24 bits for error codes)? Mark -- Mark Rotteveel -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
