make test-baseline fails

classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

make test-baseline fails

Thomas Morley-2
Hi,

I just tried `make test-baseline´ form current master:
17abdc1c92a4d2efc188aa98ddd265b8c15d6889

It fails with:

[...]
Making input/regression/lilypond-book/out-test/suffix-latex.pdf < tex
Error: /invalidfileaccess in --file--
Operand stack:
   (/home/hermann/.local/share/fonts/C059-Bold.otf)   (r)
Execution stack:
   %interp_exit   .runexec2   --nostringval--   --nostringval--
--nostringval--   2   %stopped_push   --nostringval--
--nostringval--   --nostringval--   false   1   %stopped_push   1974
1   3   %oparray_pop   1973   1   3   %oparray_pop   1961   1   3
%oparray_pop   1817   1   3   %oparray_pop   --nostringval--
%errorexec_pop   .runexec2   --nostringval--   --nostringval--
--nostringval--   2   %stopped_push   --nostringval--
Dictionary stack:
   --dict:732/1123(ro)(G)--   --dict:0/20(G)--   --dict:76/200(L)--
Current allocation mode is local
Last OS error: Permission denied
Current file position is 238
GPL Ghostscript GIT PRERELEASE 9.51: Unrecoverable error, exit code 1
/home/hermann/lilypond-git-guile-2.2/./make/lilypond-book-rules.make:35:
recipe for target 'out-test/suffix-latex.pdf' failed
make[3]: *** [out-test/suffix-latex.pdf] Error 1
make[3]: *** Deleting file 'out-test/suffix-latex.pdf'
/home/hermann/lilypond-git-guile-2.2/./input/regression/lilypond-book/GNUmakefile:22:
recipe for target 'local-test' failed
make[2]: *** [local-test] Error 2
/home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:326: recipe for
target 'test' failed
make[1]: *** [test] Error 2
/home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:337: recipe for
target 'test-baseline-pre' failed
make: *** [test-baseline-pre] Error 2

I use selfcompiled ghostscript 9.51 and guile-1.8.8 (don't get
confused by the repo-name)


Any insights?

Cheers,
  Harm

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Thomas Morley-2
Am Fr., 6. Dez. 2019 um 14:38 Uhr schrieb Thomas Morley
<[hidden email]>:

>
> Hi,
>
> I just tried `make test-baseline´ form current master:
> 17abdc1c92a4d2efc188aa98ddd265b8c15d6889
>
> It fails with:
>
> [...]
> Making input/regression/lilypond-book/out-test/suffix-latex.pdf < tex
> Error: /invalidfileaccess in --file--
> Operand stack:
>    (/home/hermann/.local/share/fonts/C059-Bold.otf)   (r)
> Execution stack:
>    %interp_exit   .runexec2   --nostringval--   --nostringval--
> --nostringval--   2   %stopped_push   --nostringval--
> --nostringval--   --nostringval--   false   1   %stopped_push   1974
> 1   3   %oparray_pop   1973   1   3   %oparray_pop   1961   1   3
> %oparray_pop   1817   1   3   %oparray_pop   --nostringval--
> %errorexec_pop   .runexec2   --nostringval--   --nostringval--
> --nostringval--   2   %stopped_push   --nostringval--
> Dictionary stack:
>    --dict:732/1123(ro)(G)--   --dict:0/20(G)--   --dict:76/200(L)--
> Current allocation mode is local
> Last OS error: Permission denied
> Current file position is 238
> GPL Ghostscript GIT PRERELEASE 9.51: Unrecoverable error, exit code 1
> /home/hermann/lilypond-git-guile-2.2/./make/lilypond-book-rules.make:35:
> recipe for target 'out-test/suffix-latex.pdf' failed
> make[3]: *** [out-test/suffix-latex.pdf] Error 1
> make[3]: *** Deleting file 'out-test/suffix-latex.pdf'
> /home/hermann/lilypond-git-guile-2.2/./input/regression/lilypond-book/GNUmakefile:22:
> recipe for target 'local-test' failed
> make[2]: *** [local-test] Error 2
> /home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:326: recipe for
> target 'test' failed
> make[1]: *** [test] Error 2
> /home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:337: recipe for
> target 'test-baseline-pre' failed
> make: *** [test-baseline-pre] Error 2
>
> I use selfcompiled ghostscript 9.51 and guile-1.8.8 (don't get
> confused by the repo-name)
>
>
> Any insights?
>
> Cheers,
>   Harm

Ok, I downgraded to ghostscript 9.26.

Then I did `make test-clean´ without complaints.

Though `make doc-clean´ returned:
$ make doc-clean
/home/hermann/lilypond-git-guile-2.2/./Documentation/po/GNUmakefile:30:
warning: overriding recipe for target 'po-update'
/home/hermann/lilypond-git-guile-2.2/stepmake/stepmake/podir-targets.make:14:
warning: ignoring old recipe for target 'po-update'
Which is strange.

Trying `make test-baseline´ now succeeded.

Alas, `make check´ on _unchanged_ master:

[...]
Making input/regression/lilypond-book/out-test/texinfo-papersize-docs.pdf < texi
comparing input/regression/out-test-baseline
       to input/regression/out-test
  2098 signature
     0 midi
  1344 log
  1344 profile
     1 gittxt
comparing input/regression/midi/out-test-baseline
       to input/regression/midi/out-test
    30 signature
    39 midi
    30 log
    30 profile
     0 gittxt
comparing input/regression/midi/out-test-baseline/recovered
       to input/regression/midi/out-test/recovered
     0 signature
     0 midi
     0 log
     0 profile
     0 gittxt
no source for input/regression/midi/out-test/sequence-name-scoping-1.midi
no source for input/regression/out-test/tree.gittxt
no source for input/regression/midi/out-test/sequence-name-scoping-2.midi
no source for input/regression/midi/out-test/sequence-name-scoping-5.midi
no source for input/regression/midi/out-test/sequence-name-scoping-3.midi
no source for input/regression/midi/out-test/sequence-name-scoping-4.midi
no source for input/regression/midi/out-test/sequence-name-1.midi
no source for input/regression/midi/out-test/sequence-name-3.midi
no source for input/regression/midi/out-test/dynamic-initial-1.midi
no source for input/regression/midi/out-test/sequence-name-2.midi
output-distance summary:
     2 changed
     0 in baseline only
    52 below threshold
  3981 unchanged
     0 new
writing /home/hermann/lilypond-git-guile-2.2/build/out/test-results/input/regression/out-test-baseline/test-output-distance.png
writing /home/hermann/lilypond-git-guile-2.2/build/out/test-results/input/regression/out-test/test-output-distance.png
writing /home/hermann/lilypond-git-guile-2.2/build/out/test-results/index.txt
Validating /home/hermann/lilypond-git-guile-2.2/build/out/test-results/index.html
line 73 column 76 - Warning: <img> attribute "alt/" lacks value
line 79 column 169 - Warning: <img> attribute "alt/" lacks value
line 73 column 76 - Warning: <img> lacks "alt" attribute
line 79 column 169 - Warning: <img> lacks "alt" attribute
/home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:376: recipe for
target 'local-check' failed
make: *** [local-check] Error 1

Well, I'd need some help here...
No clue what's going on and why, all I wanted was test a possible
patch localy...

I try `make LANGS='' doc´ on unchanged master now.


Cheers,
  Harm

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Dan Eble
On Dec 6, 2019, at 12:32, Thomas Morley <[hidden email]> wrote:
>
> Validating /home/hermann/lilypond-git-guile-2.2/build/out/test-results/index.html
> line 73 column 76 - Warning: <img> attribute "alt/" lacks value
> line 79 column 169 - Warning: <img> attribute "alt/" lacks value
> line 73 column 76 - Warning: <img> lacks "alt" attribute
> line 79 column 169 - Warning: <img> lacks "alt" attribute
> /home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:376: recipe for
> target 'local-check' failed
> make: *** [local-check] Error 1

This is tidy complaining about index.html.  What version of tidy do you have (tidy --version)?  My build environment has 5.6.0 and it isn't complaining.

Dan


Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Thomas Morley-2
Am Fr., 6. Dez. 2019 um 19:50 Uhr schrieb Dan Eble <[hidden email]>:

>
> On Dec 6, 2019, at 12:32, Thomas Morley <[hidden email]> wrote:
> >
> > Validating /home/hermann/lilypond-git-guile-2.2/build/out/test-results/index.html
> > line 73 column 76 - Warning: <img> attribute "alt/" lacks value
> > line 79 column 169 - Warning: <img> attribute "alt/" lacks value
> > line 73 column 76 - Warning: <img> lacks "alt" attribute
> > line 79 column 169 - Warning: <img> lacks "alt" attribute
> > /home/hermann/lilypond-git-guile-2.2/GNUmakefile.in:376: recipe for
> > target 'local-check' failed
> > make: *** [local-check] Error 1
>
> This is tidy complaining about index.html.  What version of tidy do you have (tidy --version)?  My build environment has 5.6.0 and it isn't complaining.
> —
> Dan
>

$ tidy --version
HTML Tidy for Linux version 5.2.0

Harm

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Dan Eble
On Dec 6, 2019, at 13:54, Thomas Morley <[hidden email]> wrote:
>
>> This is tidy complaining about index.html.  What version of tidy do you have (tidy --version)?  My build environment has 5.6.0 and it isn't complaining.
>
> $ tidy --version
> HTML Tidy for Linux version 5.2.0

I'll try to get it and fix the issue.  I'm not sure if you can work around it by some argument to configure.sh, but you should be able to work around it by changing TIDY to "false" in config.make in the top-level build directory.

Dan


Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Thomas Morley-2
Am Fr., 6. Dez. 2019 um 19:58 Uhr schrieb Dan Eble <[hidden email]>:

>
> On Dec 6, 2019, at 13:54, Thomas Morley <[hidden email]> wrote:
> >
> >> This is tidy complaining about index.html.  What version of tidy do you have (tidy --version)?  My build environment has 5.6.0 and it isn't complaining.
> >
> > $ tidy --version
> > HTML Tidy for Linux version 5.2.0
>
> I'll try to get it and fix the issue.  I'm not sure if you can work around it by some argument to configure.sh, but you should be able to work around it by changing TIDY to "false" in config.make in the top-level build directory.
> —
> Dan
>

configure has the option
--disable-FEATURE       do not include FEATURE (same as --enable-FEATURE=no)

No clue whether --disable-tidy will work

Anyway, I'll first wait for `make doc´ to finish (which will last a
long time) before I give it a try.


Thanks,
  Harm

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Dan Eble

>>> $ tidy --version
>>> HTML Tidy for Linux version 5.2.0
>>
>> I'll try to get it and fix the issue.

https://codereview.appspot.com/551250043/

Dan


Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Thomas Morley-2
Am Fr., 6. Dez. 2019 um 20:45 Uhr schrieb Dan Eble <[hidden email]>:

>
>
> >>> $ tidy --version
> >>> HTML Tidy for Linux version 5.2.0
> >>
> >> I'll try to get it and fix the issue.
>
> https://codereview.appspot.com/551250043/
> —
> Dan
>

Hi Dan,

after applying your patch all of `make´, `make test-baseline´ and
`make check´ (on unchanged master) work.
I now start testing my patch...

To be complete, configure with --disable-FEATURE , here --disable-tidy
didn't do the trick.

Btw, to get access to the "flip"-functionality I need to allow it in
NoScript of my firefox. Intended?


Nevertheless we seem to have a problem with ghostscript 9.51 (see my
initial post)


Thanks,
  Harm

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

James Lowe-3
Hello

On 07/12/2019 09:36, Thomas Morley wrote:

> Am Fr., 6. Dez. 2019 um 20:45 Uhr schrieb Dan Eble <[hidden email]>:
>>
>>>>> $ tidy --version
>>>>> HTML Tidy for Linux version 5.2.0
>>>> I'll try to get it and fix the issue.
>> https://codereview.appspot.com/551250043/
>> —
>> Dan
>>
> Hi Dan,
>
> after applying your patch all of `make´, `make test-baseline´ and
> `make check´ (on unchanged master) work.
> I now start testing my patch...
>
> To be complete, configure with --disable-FEATURE , here --disable-tidy
> didn't do the trick.
>
> Btw, to get access to the "flip"-functionality I need to allow it in
> NoScript of my firefox. Intended?
>
>
> Nevertheless we seem to have a problem with ghostscript 9.51 (see my
> initial post)

I am using gs 9.26, I don't use Tidy when testing (simply because it
isn't required).

So apart from 'required' things breaking as newer versions come out (e.g
Ghostscript), should I be testing patches with required AND non-required
components installed? Or should I be testing them with ONLY required
components but then we should make patchy-staging use required AND
non-required?

Else we get into a mess with who has what on which system they happen to
be building from.

James



Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Thomas Morley-2
Hi James,

Am Sa., 7. Dez. 2019 um 11:43 Uhr schrieb James <[hidden email]>:

>
> Hello
>
> On 07/12/2019 09:36, Thomas Morley wrote:
> > Am Fr., 6. Dez. 2019 um 20:45 Uhr schrieb Dan Eble <[hidden email]>:
> >>
> >>>>> $ tidy --version
> >>>>> HTML Tidy for Linux version 5.2.0
> >>>> I'll try to get it and fix the issue.
> >> https://codereview.appspot.com/551250043/
> >> —
> >> Dan
> >>
> > Hi Dan,
> >
> > after applying your patch all of `make´, `make test-baseline´ and
> > `make check´ (on unchanged master) work.
> > I now start testing my patch...
> >
> > To be complete, configure with --disable-FEATURE , here --disable-tidy
> > didn't do the trick.
> >
> > Btw, to get access to the "flip"-functionality I need to allow it in
> > NoScript of my firefox. Intended?
> >
> >
> > Nevertheless we seem to have a problem with ghostscript 9.51 (see my
> > initial post)
>
> I am using gs 9.26, I don't use Tidy when testing (simply because it
> isn't required).

I didn't understand how to exclude tidy, other than as adviced by Dan
further aboce in this thread.
How do you do so?

> So apart from 'required' things breaking as newer versions come out (e.g
> Ghostscript), should I be testing patches with required AND non-required
> components installed? Or should I be testing them with ONLY required
> components but then we should make patchy-staging use required AND
> non-required?
>
> Else we get into a mess with who has what on which system they happen to
> be building from.

I'd be fine testing with required components only. If optional
components fail, than it's a bug, worth reporting (as I did above) and
fixing (as done by Dan).

The ghostscript 9.51 problem will likely beat us in future, though
this version isn't released as far as I know, it was selfcompiled out
of their current master.

My two cents,
  Harm

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

James Lowe-3
Hello

On 07/12/2019 10:57, Thomas Morley wrote:

> Hi James,
>
> Am Sa., 7. Dez. 2019 um 11:43 Uhr schrieb James <[hidden email]>:
>> Hello
>>
>> On 07/12/2019 09:36, Thomas Morley wrote:
>>> Am Fr., 6. Dez. 2019 um 20:45 Uhr schrieb Dan Eble <[hidden email]>:
>>>>>>> $ tidy --version
>>>>>>> HTML Tidy for Linux version 5.2.0
>>>>>> I'll try to get it and fix the issue.
>>>> https://codereview.appspot.com/551250043/
>>>> —
>>>> Dan
>>>>
>>> Hi Dan,
>>>
>>> after applying your patch all of `make´, `make test-baseline´ and
>>> `make check´ (on unchanged master) work.
>>> I now start testing my patch...
>>>
>>> To be complete, configure with --disable-FEATURE , here --disable-tidy
>>> didn't do the trick.
>>>
>>> Btw, to get access to the "flip"-functionality I need to allow it in
>>> NoScript of my firefox. Intended?
>>>
>>>
>>> Nevertheless we seem to have a problem with ghostscript 9.51 (see my
>>> initial post)
>> I am using gs 9.26, I don't use Tidy when testing (simply because it
>> isn't required).
> I didn't understand how to exclude tidy, other than as adviced by Dan
> further aboce in this thread.
> How do you do so?

I simply never installed it, it pops up when I run configure as a
'consider installing' option.

So I guess (depending on distro) apt-get uninstall tidy (?)

My configure output in context of version numbers shows this:

--snip--

checking python version... 2.7.15
checking /usr/bin/python version... 2.7.15
checking gcc version... 7.4.0
checking g++ version... 7.4.0
checking bison version... 3.0.4
checking /usr/local/bin/guile-config version... 1.8.8
checking /usr/bin/fontforge version... 20170924
checking pkg-config version... 0.29.1
checking for pangoft2 >= 1.38.0... yes
checking for fontconfig >= 2.4.0... yes
checking for freetype2 >= 2.1.10... yes
checking for glib-2.0 >= 2.38... yes
checking for gobject-2.0 >= 2.38... yes
checking guile version... 1.8.8
checking /usr/bin/gs version... 9.26
checking gs version... 9.26
checking makeinfo version... 6.5
checking texi2html version... 1.82
checking dblatex version... 0.3.10
--snip--

I use Ubuntu 18.04. Apart from Guile and texi2html, I don't compile
binaries we need for building LP and Doc unless I am asked to 'test'
something specific. So everything else should be what the distribution's
repos provide.

James

Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Dan Eble
In reply to this post by Thomas Morley-2
On Dec 7, 2019, at 04:36, Thomas Morley <[hidden email]> wrote:
>
> Btw, to get access to the "flip"-functionality I need to allow it in
> NoScript of my firefox. Intended?

Yes.  It requires JavaScript, as do the filter links at the top.

Dan


Reply | Threaded
Open this post in threaded view
|

Re: make test-baseline fails

Dan Eble
In reply to this post by James Lowe-3
On Dec 7, 2019, at 05:43, James <[hidden email]> wrote:
>
> So apart from 'required' things breaking as newer versions come out (e.g Ghostscript), should I be testing patches with required AND non-required components installed? Or should I be testing them with ONLY required components but then we should make patchy-staging use required AND non-required?
>
> Else we get into a mess with who has what on which system they happen to be building from.

My 5¢ (no more pennies here):

It's enough to test with or without tidy, whichever happens to be the condition of your system.

It would be nice if patchy-staging used tidy, but it isn't a high priority.

It is wise for a contributor making changes to output-distance.py to install tidy.

Dan