Yeah, I don't get the difference either but I get a different error on
Ubuntu 13.10 after a "make dist"
/bin/bash ../../ylwrap `test -f 'src/thrifty.yy' || echo
'./'`src/thrifty.yy y.tab.c thrifty.cc y.tab.h `echo thrifty.cc | sed -e
s/cc$/h/ -e s/cpp$/hpp/ -e s/cxx$/hxx/ -e s/c++$/h++/ -e s/c$/h/` y.output
thrifty.output -- bison -y -d
thrifty.h is unchanged
make[3]: Leaving directory `/home/henrique/dev/thrift-gh/compiler/cpp'
cp: cannot stat ‘./thrifty.hh’: No such file or directory
On 1 February 2014 14:38, Jens Geyer <jensge...@hotmail.com> wrote:
> Not sure. I have touched thrifty.yy while working on the doctext stuff,
> but OTOH all the appropriate builds on apache.org have been successful:
>
> https://builds.apache.org/job/Thrift/1025/
> https://builds.apache.org/job/Thrift-Compiler-Linux64/943/
> https://builds.apache.org/job/Thrift-Compiler-Windows/1051/changes
>
> In your travis log, why is thrifty.h updated AFTER it is searched? Is that
> by intention? Shouldn't it be reversed: first create/update the file,
> before we use it?
>
> JensG
>
>
> -----Ursprüngliche Nachricht----- From: Henrique Mendonça
> Sent: Saturday, February 1, 2014 1:19 PM
> To: dev@thrift.apache.org
> Subject: Re: [jira] [Commented] (THRIFT-2338) First doctext wrongly
> interpreted as program doctext in some cases
>
>
> Hi Jens,
>
> Oops sorry, I didn't mean to send that!
> But perhaps that's related to this patch?
>
> https://travis-ci.org/apache/thrift/builds/18011132 line 1509
> thriftl.ll:64:21: fatal error: thrifty.h: No such file or directory
> compilation terminated.
> updating thrifty.h
> ...
> make: *** [all] Error 2
> The command "make -j$NUM_CPU" exited with 2.
>
> the job fails but after the make check runs normally again!
> Not sure what's causing it.
>
> - Henrique
>
>
> On 1 February 2014 13:16, Henrique Mendonça <henri...@apache.org> wrote:
>
> Hi Jens,
>>
>>
>> https://travis-ci.org/apache/thrift/builds/18011132 line 1509
>>
>> thriftl.ll:64:21: fatal error: thrifty.h: No such file or directory
>>
>> compilation terminated.
>>
>> updating thrifty.h
>> ...
>>
>> make: *** [all] Error 2
>>
>> The command "make -j$NUM_CPU" exited with 2.
>>
>> $ make check -j$NUM_CPU && sh test/test.sh
>>
>>
>>
>> On 1 February 2014 01:02, Hudson (JIRA) <j...@apache.org> wrote:
>>
>>
>>> [
>>> https://issues.apache.org/jira/browse/THRIFT-2338?page=
>>> com.atlassian.jira.plugin.system.issuetabpanels:comment-
>>> tabpanel&focusedCommentId=13888334#comment-13888334]
>>>
>>> Hudson commented on THRIFT-2338:
>>> --------------------------------
>>>
>>> SUCCESS: Integrated in Thrift #1025 (See [
>>> https://builds.apache.org/job/Thrift/1025/])
>>> THRIFT-2338 First doctext wrongly interpreted as program doctext in some
>>> cases (jensg: rev 813749dc50676c9d2ba778bf73bcb9bb872bd30c)
>>> * compiler/cpp/src/main.cc
>>> * compiler/cpp/src/logging.h
>>> * compiler/cpp/src/thriftl.ll
>>> * compiler/cpp/src/parse/t_doc.h
>>> * compiler/cpp/src/globals.h
>>> * compiler/cpp/src/main.h
>>>
>>>
>>> > First doctext wrongly interpreted as program doctext in some cases
>>> > ------------------------------------------------------------------
>>> >
>>> > Key: THRIFT-2338
>>> > URL: https://issues.apache.org/jira/browse/THRIFT-2338
>>> > Project: Thrift
>>> > Issue Type: Sub-task
>>> > Components: Compiler (General)
>>> > Affects Versions: 0.9.2
>>> > Reporter: Jens Geyer
>>> > Assignee: Jens Geyer
>>> > Fix For: 0.9.2
>>> >
>>> > Attachments:
>>> THRIFT-2338_First_doctext_wrongly_interpreted_as_
>>> program_doctext_in_some_cases.patch
>>> >
>>> >
>>> > One can never have enough test cases.
>>> > {code}
>>> > namespace * foo
>>> > /** the coolest enum ever */
>>> > enum bar { baz = 1 }
>>> > {code}
>>> > ends up with a program-level doctext saying "the coolest enum ever".
>>>
>>>
>>>
>>> --
>>> This message was sent by Atlassian JIRA
>>> (v6.1.5#6160)
>>>
>>>
>>
>>
>
