D12074: Solve problem of appearance input of interpreter in interpreter's output in Lua backend
Nikita Sirgienko
noreply at phabricator.kde.org
Wed Apr 25 21:03:58 UTC 2018
sirgienko added a comment.
In D12074#253885 <https://phabricator.kde.org/D12074#253885>, @asemke wrote:
> This all looks a bit strange...
Well, in fact this is a problem of incorrect parse end condition, which produce false positive errors.
I add printing of readed line from backend whatever make it clear.
final command to be executed "function fib(n)\n if n < 2 then return 1 end\n return fib(n - 2) + fib(n - 1)\nend\n\n-- Closures and anonymous functions are ok:\nfunction adder(x)\n -- The returned function is created when adder is\n -- called, and remembers the value of x:\n return function (y) return x + y end\nend\na1 = adder(9)\na2 = adder(36)\nprint(a1(16)) --> 25\nprint(a2(64)) --> 100\n"
QSizeF(937, 397) QSizeF(937, 297)
newLine: "function fib(n)\n"
newLine: ">> if n < 2 then return 1 end\n"
newLine: ">> return fib(n - 2) + fib(n - 1)\n"
newLine: ">> end\n"
newLine: "> \n"
newLine: "> -- Closures and anonymous functions are ok:\n"
newLine: "> function adder(x)\n"
newLine: ">> -- The returned function is created when adder is\n"
newLine: ">> -- called, and remembers the value of x:\n"
newLine: ">> return function (y) return x + y end\n"
newLine: ">> end\n"
newLine: "> a1 = adder(9)\n"
newLine: "> a2 = adder(36)\n"
newLine: "> "
parsing the output "function fib(n)\n>> if n < 2 then return 1 end\n>> return fib(n - 2) + fib(n - 1)\n>> end\n> \n> -- Closures and anonymous functions are ok:\n> function adder(x)\n>> -- The returned function is created when adder is\n>> -- called, and remembers the value of x:\n>> return function (y) return x + y end\n>> end\n> a1 = adder(9)\n> a2 = adder(36)\n> "
final output of the command "function fib(n)\n if n < 2 then return 1 end\n return fib(n - 2) + fib(n - 1)\nend\n\n-- Closures and anonymous functions are ok:\nfunction adder(x)\n -- The returned function is created when adder is\n -- called, and remembers the value of x:\n return function (y) return x + y end\nend\na1 = adder(9)\na2 = adder(36)\nprint(a1(16)) --> 25\nprint(a2(64)) --> 100" : ""
Command `a2 = adder(36)` not last, but after this command we read `> ` (lua printing lag) and it satisfy parse end condition. So, backend finish the first command and run next, and then:
QSizeF(937, 157) QSizeF(937, 126)
newLine: "print(a1(16)) --> 25\n"
newLine: "25\n"
newLine: "> print(a2(64)) --> 100\n"
newLine: "100\n"
newLine: "> "
parsing the output "print(a1(16)) --> 25\n25\n> print(a2(64)) --> 100\n100\n> "
final output of the command "function bar(a, b, c)\n print(a, b, c)\n return 4, 8, 15, 16, 23, 42\nend\n\nx, y = bar('zaphod') --> prints \"zaphod nil nil\"" : "print(a1(16)) --> 25\n25\nprint(a2(64)) --> 100\n100"
We set tail of previous command as result for two command.
And then:
newLine: "function bar(a, b, c)\n"
newLine: ">> print(a, b, c)\n"
newLine: ">> return 4, 8, 15, 16, 23, 42\n"
newLine: ">> end\n"
newLine: "> \n"
newLine: "> x, y = bar('zaphod') --> prints \"zaphod nil nil\"\n"
newLine: "zaphod\tnil\tnil\n"
newLine: "> "
parsing the output "function bar(a, b, c)\n>> print(a, b, c)\n>> return 4, 8, 15, 16, 23, 42\n>> end\n> \n> x, y = bar('zaphod') --> prints \"zaphod nil nil\"\nzaphod\tnil\tnil\n> "
final output of the command "function bar(a, b, c)\n print(a, b, c)\n return 4, 8, 15, 16, 23, 42\nend\n\nx, y = bar('zaphod') --> prints \"zaphod nil nil\"" : "zaphod\tnil\tnil"
The backend set result again.
In D12074#253921 <https://phabricator.kde.org/D12074#253921>, @asemke wrote:
> In D12074#253885 <https://phabricator.kde.org/D12074#253885>, @asemke wrote:
>
> > This all looks a bit strange...
>
>
> The problem was in LuaSession::readOutput(). The patch F5822387: lua_multiline_output_3.patch <https://phabricator.kde.org/F5822387>
>
> handles the outputs correctly. @sirgienko Does it also work for you?
No, output have changed, but still have problems:
F5822390: изображение.png <https://phabricator.kde.org/F5822390>
REPOSITORY
R55 Cantor
REVISION DETAIL
https://phabricator.kde.org/D12074
To: sirgienko, #cantor, asemke
Cc: #cantor, #kde_edu, narvaez, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20180425/a45e6d41/attachment.html>
More information about the kde-edu
mailing list