* any reason for "cmake_" prefix on cmake_runcmake_build()?
@ 2021-03-08 11:22 Robert P. J. Day
2021-03-08 18:37 ` [OE-core] " Khem Raj
0 siblings, 1 reply; 6+ messages in thread
From: Robert P. J. Day @ 2021-03-08 11:22 UTC (permalink / raw)
To: OE Core mailing list
collecting some examples of inheritance of class functions using
EXPORT_FUNCTIONS, and ran across this routine in cmake.bbclass:
... snip ...
cmake_runcmake_build() {
bbnote ...
eval ...
}
cmake_do_compile() {
cmake_runcmake_build --target ${OECMAKE_TARGET_COMPILE}
}
... snip ...
what puzzles me is that the routine cmake_runcmake_build(), despite
having a "cmake_" prefix, is not being exported with EXPORT_FUNCTIONS,
so while that's a perfectly respectable function, it's not clear why
it would have a "cmake_" prefix. is there some value to that prefix,
or is it just arbitrary?
rday
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [OE-core] any reason for "cmake_" prefix on cmake_runcmake_build()? 2021-03-08 11:22 any reason for "cmake_" prefix on cmake_runcmake_build()? Robert P. J. Day @ 2021-03-08 18:37 ` Khem Raj 2021-03-10 11:38 ` Robert P. J. Day 0 siblings, 1 reply; 6+ messages in thread From: Khem Raj @ 2021-03-08 18:37 UTC (permalink / raw) To: Robert P. J. Day, OE Core mailing list On 3/8/21 3:22 AM, Robert P. J. Day wrote: > > collecting some examples of inheritance of class functions using > EXPORT_FUNCTIONS, and ran across this routine in cmake.bbclass: > > ... snip ... > cmake_runcmake_build() { > bbnote ... > eval ... > } > > cmake_do_compile() { > cmake_runcmake_build --target ${OECMAKE_TARGET_COMPILE} > } > ... snip ... > > what puzzles me is that the routine cmake_runcmake_build(), despite > having a "cmake_" prefix, is not being exported with EXPORT_FUNCTIONS, > so while that's a perfectly respectable function, it's not clear why > it would have a "cmake_" prefix. is there some value to that prefix, > or is it just arbitrary? it does not intend to provide a default function like do_compile do_install exporting it is not necessary > > rday > > > > > ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [OE-core] any reason for "cmake_" prefix on cmake_runcmake_build()? 2021-03-08 18:37 ` [OE-core] " Khem Raj @ 2021-03-10 11:38 ` Robert P. J. Day 2021-03-10 16:35 ` Khem Raj 2021-03-10 21:27 ` Richard Purdie 0 siblings, 2 replies; 6+ messages in thread From: Robert P. J. Day @ 2021-03-10 11:38 UTC (permalink / raw) To: Khem Raj; +Cc: OE Core mailing list On Mon, 8 Mar 2021, Khem Raj wrote: > > > On 3/8/21 3:22 AM, Robert P. J. Day wrote: > > > > collecting some examples of inheritance of class functions using > > EXPORT_FUNCTIONS, and ran across this routine in cmake.bbclass: > > > > ... snip ... > > cmake_runcmake_build() { > > bbnote ... > > eval ... > > } > > > > cmake_do_compile() { > > cmake_runcmake_build --target ${OECMAKE_TARGET_COMPILE} > > } > > ... snip ... > > > > what puzzles me is that the routine cmake_runcmake_build(), despite > > having a "cmake_" prefix, is not being exported with EXPORT_FUNCTIONS, > > so while that's a perfectly respectable function, it's not clear why > > it would have a "cmake_" prefix. is there some value to that prefix, > > or is it just arbitrary? > > it does not intend to provide a default function like do_compile do_install > exporting it is not necessary no, that part i understand ... the nitpicky observation was the use of the "cmake_" prefix for a function which was not intended to be exported for the purpose of inheritance. sure, it's perfectly legal, but in a stylistic sense, it just leaves open the possibility of misinterpretation by someone reading the code. a simpler style rule might be, "reserve classname_ prefixes for only those routines you plan on exporting." but i admit this is a quibble. rday ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [OE-core] any reason for "cmake_" prefix on cmake_runcmake_build()? 2021-03-10 11:38 ` Robert P. J. Day @ 2021-03-10 16:35 ` Khem Raj 2021-03-10 21:27 ` Richard Purdie 1 sibling, 0 replies; 6+ messages in thread From: Khem Raj @ 2021-03-10 16:35 UTC (permalink / raw) To: Robert P. J. Day; +Cc: OE Core mailing list [-- Attachment #1: Type: text/plain, Size: 1803 bytes --] On Wed, Mar 10, 2021 at 3:38 AM Robert P. J. Day <rpjday@crashcourse.ca> wrote: > On Mon, 8 Mar 2021, Khem Raj wrote: > > > > > > > On 3/8/21 3:22 AM, Robert P. J. Day wrote: > > > > > > collecting some examples of inheritance of class functions using > > > EXPORT_FUNCTIONS, and ran across this routine in cmake.bbclass: > > > > > > ... snip ... > > > cmake_runcmake_build() { > > > bbnote ... > > > eval ... > > > } > > > > > > cmake_do_compile() { > > > cmake_runcmake_build --target ${OECMAKE_TARGET_COMPILE} > > > } > > > ... snip ... > > > > > > what puzzles me is that the routine cmake_runcmake_build(), despite > > > having a "cmake_" prefix, is not being exported with EXPORT_FUNCTIONS, > > > so while that's a perfectly respectable function, it's not clear why > > > it would have a "cmake_" prefix. is there some value to that prefix, > > > or is it just arbitrary? > > > > it does not intend to provide a default function like do_compile > do_install > > exporting it is not necessary > > no, that part i understand ... the nitpicky observation was the use > of the "cmake_" prefix for a function which was not intended to be > exported for the purpose of inheritance. sure, it's perfectly legal, > but in a stylistic sense, it just leaves open the possibility of > misinterpretation by someone reading the code. > > a simpler style rule might be, "reserve classname_ prefixes for only > those routines you plan on exporting." but i admit this is a quibble. This is more towards code styling issue and we have not really bothered to define such namespaces they cause more confusion than the problem they solve since such things should be checked by linters other wise they fall apart > > > rday > [-- Attachment #2: Type: text/html, Size: 2521 bytes --] ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [OE-core] any reason for "cmake_" prefix on cmake_runcmake_build()? 2021-03-10 11:38 ` Robert P. J. Day 2021-03-10 16:35 ` Khem Raj @ 2021-03-10 21:27 ` Richard Purdie 2021-03-11 9:15 ` Robert P. J. Day 1 sibling, 1 reply; 6+ messages in thread From: Richard Purdie @ 2021-03-10 21:27 UTC (permalink / raw) To: Robert P. J. Day, Khem Raj; +Cc: OE Core mailing list On Wed, 2021-03-10 at 06:38 -0500, Robert P. J. Day wrote: > On Mon, 8 Mar 2021, Khem Raj wrote: > > > > > > > On 3/8/21 3:22 AM, Robert P. J. Day wrote: > > > > > > collecting some examples of inheritance of class functions using > > > EXPORT_FUNCTIONS, and ran across this routine in cmake.bbclass: > > > > > > ... snip ... > > > cmake_runcmake_build() { > > > bbnote ... > > > eval ... > > > } > > > > > > cmake_do_compile() { > > > cmake_runcmake_build --target ${OECMAKE_TARGET_COMPILE} > > > } > > > ... snip ... > > > > > > what puzzles me is that the routine cmake_runcmake_build(), despite > > > having a "cmake_" prefix, is not being exported with EXPORT_FUNCTIONS, > > > so while that's a perfectly respectable function, it's not clear why > > > it would have a "cmake_" prefix. is there some value to that prefix, > > > or is it just arbitrary? > > > > it does not intend to provide a default function like do_compile do_install > > exporting it is not necessary > > no, that part i understand ... the nitpicky observation was the use > of the "cmake_" prefix for a function which was not intended to be > exported for the purpose of inheritance. sure, it's perfectly legal, > but in a stylistic sense, it just leaves open the possibility of > misinterpretation by someone reading the code. > > a simpler style rule might be, "reserve classname_ prefixes for only > those routines you plan on exporting." but i admit this is a quibble. No, please lets not do that. Having come kind of standard prefixing for variables/functions in classes does help give people direction to know where to find code from tracebacks or variable definitions and the like. The use of EXPORT_FUNCTIONS is a bit of a special case which requires the prefixing but I really don't want to see any recommendation we drop all prefixes otherwise. Cheers, Richard ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [OE-core] any reason for "cmake_" prefix on cmake_runcmake_build()? 2021-03-10 21:27 ` Richard Purdie @ 2021-03-11 9:15 ` Robert P. J. Day 0 siblings, 0 replies; 6+ messages in thread From: Robert P. J. Day @ 2021-03-11 9:15 UTC (permalink / raw) To: Richard Purdie; +Cc: Khem Raj, OE Core mailing list [-- Attachment #1: Type: text/plain, Size: 2169 bytes --] On Wed, 10 Mar 2021, Richard Purdie wrote: > On Wed, 2021-03-10 at 06:38 -0500, Robert P. J. Day wrote: > > On Mon, 8 Mar 2021, Khem Raj wrote: > > > > > > > > > > > On 3/8/21 3:22 AM, Robert P. J. Day wrote: > > > > > > > > collecting some examples of inheritance of class functions using > > > > EXPORT_FUNCTIONS, and ran across this routine in cmake.bbclass: > > > > > > > > ... snip ... > > > > cmake_runcmake_build() { > > > > bbnote ... > > > > eval ... > > > > } > > > > > > > > cmake_do_compile() { > > > > cmake_runcmake_build --target ${OECMAKE_TARGET_COMPILE} > > > > } > > > > ... snip ... > > > > > > > > what puzzles me is that the routine cmake_runcmake_build(), despite > > > > having a "cmake_" prefix, is not being exported with EXPORT_FUNCTIONS, > > > > so while that's a perfectly respectable function, it's not clear why > > > > it would have a "cmake_" prefix. is there some value to that prefix, > > > > or is it just arbitrary? > > > > > > it does not intend to provide a default function like do_compile do_install > > > exporting it is not necessary > > > > no, that part i understand ... the nitpicky observation was the use > > of the "cmake_" prefix for a function which was not intended to be > > exported for the purpose of inheritance. sure, it's perfectly legal, > > but in a stylistic sense, it just leaves open the possibility of > > misinterpretation by someone reading the code. > > > > a simpler style rule might be, "reserve classname_ prefixes for only > > those routines you plan on exporting." but i admit this is a quibble. > > No, please lets not do that. Having come kind of standard prefixing > for variables/functions in classes does help give people direction > to know where to find code from tracebacks or variable definitions > and the like. > > The use of EXPORT_FUNCTIONS is a bit of a special case which > requires the prefixing but I really don't want to see any > recommendation we drop all prefixes otherwise. fair, i was probably getting overly pedantic there. moving on ... rday ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2021-03-11 9:16 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-03-08 11:22 any reason for "cmake_" prefix on cmake_runcmake_build()? Robert P. J. Day 2021-03-08 18:37 ` [OE-core] " Khem Raj 2021-03-10 11:38 ` Robert P. J. Day 2021-03-10 16:35 ` Khem Raj 2021-03-10 21:27 ` Richard Purdie 2021-03-11 9:15 ` Robert P. J. Day
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.